Home

Thesis - Peter Cross - screen viewing

1. 33 DEVELOPMENT OF A PRACTICAL DRUG DELIVERY COMMUNICATION AND MONITORING SYSTEM eese eee eene 45 3 1 Introductio estre tete terere aera i e a dp eap avv 45 3 22 Contemporary Controlled Drug Release Design Philosophy 46 3 3 Requirements of the Proposed Drug Delivery System 48 3 4 Overview of DMU Operation uini ioci captare ec Es bes oa vea tease 50 3 5 Drug Delivery Contr Olas ad a os at a o ete vetoes 53 3 6 Power Management us ton cd em dae ee oct ue 66 Vii External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals viii 3 7 CCommuficatiom n oec eco eee eee cere ever eee eee aee doce aee eo ee eaa ya eee ee eoe on 74 3 8 Executive Control ere a eee eer ads ods exec eo e dee aae 94 39 Environmental Monitored ss eee oe emet tese duos 98 3 10 Software Design Methodology sse 104 31 Base Statio Nie e x pe ee epe be eee e o be eed 106 3 12 Construction o oo elec eee E E 108 3 13 Bencli T 6sti gz so tete qno teet t tr ed re EPA SER tera edo 111 S 14 Specificati nS coc suena sults uidet fuas tue te desea eruat 117 4 EXPERIMENTATION vssscssccdecceabeveccisccsoscsavcocccaccccecsoavcceccceccseasocbcscscauesceccoave 119 4 1 Materials ese act Ate ptite Ade etna ea 119 4 2 Method sick coset ettet bote 120 4 3 Results and Discussion ccccceccecccccccccccececccccccecececececeeeeeseeevecs 124 4 4 X Ex
2. 206 xiii List of Tables Table 1 Description of Microcontroller Modes senes 68 Table 2 Current Consumption in Sleep Mode sse 71 Table 3 Current Consumption in Fully Awake Mode sess 12 Table4 tase Station Help Sereeti oan ae aed ea added dont dle edes 92 Table 5 Logged Data Result Set os i dp pleb vete usi beediekt 97 Table 6 User Requesting a Snapshot From DMU 4 Over Wireless Link 112 Table 7 DMU Specifications ener 117 Table 8 Description of Observations Results and Manipulations for DMU 3 and Control Son pe ueste Nei e NS Een a E 174 Table 9 User Requesting Setup Information From DMU 4 Over Wireless Link 175 Table 10 DMU PCB Specifications et eten ette ttti nebat teet to ep ep ab reae Re eR o LEHRER HL AIRE 185 Table 11 DMU Bill of Materials eere 189 Table 12 Development Board Cross Reference Wiring Table 204 Table 13 Jumper Settings for the M16C Microcontroller Evaluation Board 205 XV AI Anestrous Anestrus CIDR Corpus luteum DMU Estrous Estrus Flash memory Gascell MEMS Glossary Artificial Insemination Non cycling behaviour in non human mammals in which the ovaries fail to exhibit follicular development and ovulation State or interval of sexual inactivity between two periods of estrus A T shaped intrava
3. Figure 48 Network for Remote Design and Monitoring of Drug Delivery Programmes 165 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 6 4 166 Normally the results would be sent once a day from each active farm to the drug programme administrator s central database on the sponsor s premises Any pending reply or system status enquiries could be made at the same time This would take about 10 minutes per farm and occur automatically in the early hours of the morning for least disruption to farm activities Procedures on the farm could be controlled remotely For example the central monitoring drug programme administrator might be in Hamilton New Zealand If a serious problem developed in a site somewhere in Australia all remaining sites in the world could be told to suspend drug deliveries until further notice This detection and control function could be set up to happen without local human intervention There would be some time delay due to the fact that DMU needs to be transported by the cow back within range of the local base station before the command can take affect If this was a problem a repeater system could be deployed to extend the range of the DMU communication range to the entire farm A system of farm wide cow to base station communication already exists in a simplified one way form The commercially available HeatWatch system from DDx Inc mentioned previously transmits mou
4. Format is command sub command from ID to ID where items in the brackets are represented by a single letter Always start the command with a Always end the command with a In between comes the command and sub command letters followed by the from ID number followed by the to ID number A to ID of 9 commands all DMU s at once A from ID of 0 indicates the base station The case of the letters is not important If there is no sub command any letter will do Example 1 bb03 means request a verbose summary from DMU ID 3 and reply on both the radio and wired comms link Example 2 dt01 means request to download all result sets from DMU ID 1 replying on the radio link only Command Table Command Sub Command test t none any letter will do transmit ID e wake up w download all result sets d reply on radio t snap shot CSV S reply on wired comm port c summary verbose b reply on both 5 setup info verbose i flash LED f roll call r gascell override o make active a Start delivery override q make inactive i gascell override value p off if override active start delivery override value z on if override active n restart delivery j confirm k Development of a Practical Drug Delivery Communication and Monitoring System Command Set The commands can reply on the wired comm port the wireless link or both The method of reply is specified by the sub command where
5. PE IINE E ii f Fe TEL g Figure 49 DMU Schematic Top Level Yo sorore pue 1oAroouvi peog Sseriojug pue JOATSOUBI 178 Appendix B DMU Schematics r z 1 qndmo Joy pomo am IITETIUTECT LIE i pap amios jou smd NORW me Suneog prasad of Bview aeeoa wie sm Syon 42jj041u020421Jy pavog qpr QING WL aT PRAY AN TOAPTOORN E PE PONT VACHE Sd x aivese E IR C Omer uassi b x asa oc Id Ww sca AND aesa myetod PIA E s As X 0 EDILNYIL O14 ESL v FO Nvo ota iSo ri ILUSNYN OLA iSo v EO oburNv ord Sora amseaw asm APUG pp ENVIEOLE GIYA v ENVITOLd sve r LU ELLE Ez sg INVI Ola LIYA v ZF woo 91YA v P wesSagvi6d StviL TIAS WL PIOS IXSNYIS 64 siya N Prag poro BLA mmoixanws 6 EIS a Epa YA 6d ZIYA a iRUOYQE 64 n OTLLTIDOS gy Xi msma ed otv a TTALNO LINDOS eas gI 6d evn a X B simos 64 svota 60LL3320S WXL BE LYIL24 p TALNO LINDOS mooyjs 84 waza s DUUS 94 SYS z SO LNW ea wea b gt CO LINW 84 Y amp zd i WOLT OMe 84 cR zd ied Xi XL OHO TINE gt lt gery ea lyn 2a amoreod ovmza ar PL woa E secvno ta tEiq 14 ARMS AL tc OPEET SEEE iE yy YTOR onissaaq EINA LINDOS uaria PES aeia CLAY 1108 1 Ld sanis Lm TOOVIATGSRIXLO Ld aq la ENa XL AXL A ms z BL UUDXLLmpos 9 sora ILESA
6. 267 7416 227 2639 267 7416 227 2639 589 0552 589 0552 0805CS 101XJBB 0805CS 560XJBB 227 2639 227 2639 Top mark Comment two small holes white dot Approximate price price is for 2 pins of the 32 way break off strip and one jumper Price is for one socket and one pin 32 way socket strip 267 7416 32 way socket strip RS 227 2639 Price is for 3 sockets and 1 pin 32 way socket strip 267 7416 32 way socket strip RS 227 2639 Price is for one socket and one pin 32 way socket strip 267 7416 32 way socket strip RS 227 2639 price is approximate price is approximate price is for 1 pin of the 32 way break off strip price is for 1 pin of the 32 way break off strip Function Gascell Releases hydrogen gas under load to drive syringe piston General IO connector For selecting program or run MCU boot mode Aux comm port connector Hardware debugging and serial programming port connector External power connector MCU power supply filtering TXR Lrfb Transceiver power supply filtering Lesd parallel tuning and ESD protection for RFIO pin Lat antenna tuning series inductor Vdd PCB interconnect GND PCB interconnect Cost Ex GST NZ 3 00 1 08 3 82 0 32 0 48 0 32 0 34 0 34 1 00 1 00 0 24 0 24 191 External Control and Monitoring of Intravaginal Drug
7. 30K not present LDR 0603 0603 0603 0603 0603 0603 0805 0603 0603 0603 0603 0805 0805 0603 Manf Part Number NSL19 MS51 CRG0603 Series CRG0603 Series CRG0603 Series CRG0603 Series CRG0603 Series RL73 Series CRG0603 Series CRG0603 Series CRG0603 Series CRG0603 Series Manufacturer Supplier Neoohm Neoohm Neoohm Neoohm Neoohm Meggitt Polymers amp Composites Neoohm Neoohm Neoohm Neoohm MultiComp MultiComp RS RS RS RS RS RS Farnell RS RS RS RS Farnell Farnell Supplier Order Code 596 141 213 2676 213 2430 213 1982 213 2676 213 2676 213 2418 213 2676 213 2143 213 1982 213 2367 771 510 771 510 Top mark Comment 105 153 105 105 2R2 105 101 472 303 303 Farnell Feb 2000 catalogue page 1717 Farnell Feb 2000 catalogue page 1715 Farnell Feb 2000 catalogue page 1715 Cost Function Ex GST NZ Sensor for light level 2 89 indication Pull down resistor for 0 05 LDR to minimize current flow into MCU IO pin sets time delay between 0 05 power fail indication and reset signal If the TC1073 is the only 0 00 regulator fitted this ORO resistor ensures that the regulator is always enabled In that case disconnect Vreg shutdown else it will be connected to greater
8. Barn housed animals are easily heat detected because of the great peak in activity during estrus compared to the rest of their relatively sedentary lifestyle In pasturally based systems there is a higher level of background activity Heat detection is still possible but requires some post processing of the data and there is a residual higher false positive indication Body Temperature Because temperature is such an obvious and easy variable to measure it is no surprise to find that remote telemetry of vaginal temperature in cows has been done many time previously Kyle Kennedy and Small 1998 49 1437 1449 Redden et al 1993 76 713 721 Other studies of vaginal temperature in cows have been reported de Mol 2001 126 99 103 Gil Szarek and Kural 1997 65 25 29 Kyle Kennedy and Small 1998 49 1437 1449 Redden Kennedy Ingalls and Gilson 1993 76 713 721 Core temperature range is 37 8 C to 40 0 C but usually does not stray far from 38 6 C during most of the cycle Webster 1993 2 given a moderate climate Background Temperature increases nearly a whole degree Celsius 12 hours before ovulation and drops back half a degree at the time of ovulation Bobowiec Studzinski and Babiarz 1990 44 573 579 It has been suggested in unpublished Verkerk 2002 work that the increase in temperature is due to the increased physical activity associated with normal heat and that the drop is attributable to ovulation For cows e
9. mode whenever it is not actually transmitting A from address of zero represents the base station A to address of 9 represents a broadcast message to all DMU s Bit Synchronisation The receiver needs to know where the bit boundaries are This is complicated by the fact there may be up to five bits of the same value in a row When bits of the same value are sent in a row there is no transition available to help keep the receiver synchronised Although both the transmitter and receiver have highly accurate crystal controlled oscillators controlling the timing of the synchronisation algorithm the problem lies in the fact that there is significant jitter introduced by the transmission and reception process Even at moderate signal to noise ratios the output of the data slicer will exhibit some jitter in the position of the logic transitions In addition random noise spikes can generate false edges at any time Development of a Practical Drug Delivery Communication and Monitoring System To counter this a software phase lock loop is employed On the receiver side a timer is configured to overflow with the same period as the basic bit period If no transition occurs in the expected window the counter is not affected If the transition occurs a little before the expected time the count up timer value is incremented by 12 of its maximum value If the transition occurs a little after the expected time the count up timer value
10. the effective symbol rate meaning that it s only 50 efficient Eight Bit to Twelve Bit Encoding Eight bit to twelve bit encoding maps each of the eight bit patterns into 256 carefully chosen 12 bit patterns out of the available 4096 Maximum run length is 6 and this code is 67 efficient Eight Bit to Ten Bit Encoding Gigabit Ethernet IEEE 802 3z IEEE 802 3z Gigabit Ethernet uses a more complicated version of this basic principle This eight bit to ten bit code patented and licensed from IBM originally for the Fibre Channel standard has a maximum run length of five consecutive 0s or 1s Franaszek Peter and Widmer 1984 The code also has a maximum digital sum variation from 3 to 3 Digital sum variation is a running count of the zeroes and ones counting a zero as minus one and a one as plus one There are extra coded words for link control and of the 268 codes 134 are disparity neutral It works by encoding the first five bits as a six bit group according to a lookup table The second group of three bits are encoded into a four bit group from one of two tables the choice of which depends on the running disparity Franaszek and Widmer 1983 27 440 451 This eight bit to ten bit code is also a moderately strong error checking mechanism This is due to the fact that only certain code groups can be selected at any given time the choice of which code comes next is partially determined by previously received data Devel
11. 1993 ntel Corporation Winsel A Device for electrochemical generation of gases for the transportation of fluids and similar mediums US 5242565 1993 Woolford M W J H Williamson and H V Henderson 1998 Changes in electrical conductivity and somatic cell count between milk fractions from quarters subclinically infected with particular mastitis pathogens Journal of Dairy Research 65 187 198 Wyssmann M Device for the intentional and controllable distribution of a liquid or viscous material US 5741275 1998 Zahn J D A A Deshmukh A P Pisano and D Liepmann 2001 Continuous on chip micropumping through a microneedle The 14th IEEE International Conference on Micro Electro Mechanical Systems 2001 503 506 Zhao L L K Hung and T T Choy 1997 Delivery of medication by iontophoresis to treat post burn hypertrophic scars investigation of a new electronic technique Burns 23 Suppl 1 S27 S29 Zhao X 2000 Gene transfer and drug delivery by electronic pulse delivery A nonviral delivery approach Methods Mol Biol 133 37 43 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Ask dr tech Daily Comic HEY NEED yp N SOMEONE TO A THIS BABYIS SIGN FOR THIS GONNA MOOOO LER By kind permission of Steve Kearsley O 2000 Steve Kearsley 220
12. 2 least significant bits This means that measurements of very small currents can be inaccurate by a high percentage Integrated over a long time these inaccuracies can turn into large errors The DS2438 monitor IC provides a method for filtering out potentially erroneous small signals so that they are not accumulated If the current is less than two least significant bits it is not accumulated however if it is over two least significant bits it is accumulated with full accuracy Calibration Being only 3 9 Ohms the voltage across the current sense resistor is small in terms of the potential DC offset error of a typical analogue to digital converter 65 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 6 66 input op amp The effect of DC offset error is minimized by calibrating it out every 40 minutes This is achieved by reading the no load error current and storing the negative value of that in the DC offset error register Whenever the gascell monitor IC takes a current measurement 36 times a second it adds the value in the DC offset error register to the result before accumulating it 3 5 9 Manual Override Via Wireless The gascell load can be switched on or off manually over the wireless link subject to the safety check The delivery profile can also be controlled by a manual over ride suspending or resetting the delivery schedule 3 5 10 Retention Mechanism To retain the DMU in
13. 37 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 58 3 5 5 Gascell Load Control Variable Rate Mechanism The gascell is connected in series with the gascell switch load resistor and current sense resistor The schematic is presented in Figure 52 in Appendix B on page 181 Recall that gas production is directly proportional to current drawn assuming temperature is constant Some time after the electrical load on the gascell is removed the pressurised gas finishes expanding until it is just above external atmospheric pressure Static piston friction prevents it from equalising completely At this time volume delivered is directly proportional to gas production and hence integrated current Hence integrated current is used to estimate the volume delivered The mechanism for accurately measuring integrated current is discussed in detail later This theory includes some approximations but they prove to be more than adequate after calibration To turn the fixed rate mechanism into a variable rate one requires some control circuitry Every five minutes the microcontroller checks the target volume against the estimated volume it has delivered If the estimated volume is less than the target volume the gascell is turned on If the estimated volume is greater than the target volume the gascell is turned off This is a pulsatile release mechanism but the low pass filter effect caused by syringe
14. Unfortunately since only one DMU lasted to the first piston measurement it is not possible to statistically assess the accuracy of delivery 4 3 5 Integrated Current Results As you would expect the integrated current graph is a scaled version of the estimated volume delivered graph since they are related by a simple constant 131 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 120 100 80 60 40 Integrated Current mAhr 20 Flat Profile Target Integrated Current and Target Integrated Current Rate Day of Trial 0 1 2 3 4 5 6 7 60 o Integrated Current Rate mAhr day 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Gascell Actual Integrated Current mAHr Gascell Target Integrated Current mAHr s Gascell Target Integrated Current Rate mAHr Day A When the DMU is reset it immediately starts to pressurise the syringe For diagnostic purposes the integrated current accumulator is enabled so that a record of the B integrated current required to obtain pressurisation is acquired When the syringe has pressurised the integrated current accumulator is zeroed so that its value now C represents the amount of volume delivered from that point onwards The integrated current accumulates in step like fashion since the gascell switch decision is made at D discrete interva
15. and ultimately the rate at which the drug is delivered The control of this device was successfully demonstrated in the laboratory at varying rates Kim 1999 reports of a more sophisticated pump an ambulatory infusion device developed to provide drug delivery at a precisely controlled rate The pump module contains medication separated by a flexible membrane from a gas generating chamber with an electrolytic cell The control module alters the driving current level depending on the infusion rate required A microprocessor calculates the current level required based on selected infusion rate ambient pressure and ambient temperature The accuracy and precision of the device was verified over varying environmental conditions and flow rate rates 0 1 to 100 mL Hr Flow rate was found to be accurate within 5 Electrolytic gas powered pumps are easy to construct have been proven to deliver complex profiles and can cope with a large range of required flow rates Development of a Practical Drug Delivery Communication and Monitoring System 3 1 3 Development of a Practical Drug Delivery Communication and Monitoring System Introduction Drug delivery in farm animals for uses such as controlling the reproductive cycle in cows has centred on injections and or silicone inserts which permit the slow release of a hormone over a number of days In these commercial protocols the reproductive cycle is usually monitored on a manual b
16. net incomes have declined significantly over the last decade Producer Board Project Team 1999 Background 2 2 Current and Future Trends in the Global Dairy Market 2 2 1 Global Competition The decline in farm incomes is related to the fact that New Zealand has a continued reliance on low margin commodity exports While New Zealand milk output has increased significantly this volume increase has been offset by falling commodity prices With 70 percent of New Zealand dairy exports unbranded New Zealand remains highly exposed to fluctuations in international commodity prices In 1999 Promar International UK was commissioned to write a report titled International Forces Shaping The New Zealand Dairy Industry by the New Zealand Ministry of Agriculture amp Forestry and other government departments This report claims that as domestic demand is essentially constant New Zealand must look internationally to sustain growth Promar International 2001 This will be difficult given that international competition is predicted to increase According to this report the New Zealand dairy industry will have to continue to develop these international markets if it is to maintain a competitively priced product On the plus side there are some positive global influences that will take effect in the medium term Under the agreed terms of the World Trade Organization agricultural subsidies in other countries are to be phased out eventually B
17. than the MCU s Vdd Top trim resistor for 0 05 supervisor reset threshold sense Top main resistor for 0 05 supervisor reset threshold sense Bottom resistor for 0 05 supervisor reset threshold sense Current sense resistor 1 11 for batteries that power the electronics Connects open drain 0 05 error output on TC1073 3 0V regulator to maintain correct output when in shutdown mode Rrfb Transceiver 0 05 power supply filtering TXR to tie AGCCAP 0 05 high Rtxm alters transmitter 0 05 output power Rifp LPFADJ resistor 0 11 receiver low pass filter bandwidth adjust Rth2 sets DS2 data 0 11 slicer trip point Not fitted in OOK design 0 00 Minimizes power consumption in sleep mode when using ASK page 39 of designer s Guide Appendix D DMU Bill of Materials Designator R207 R208 R209 R210 R211 R212 R213 R214 101 102 SOCKET101 Description Resistor 200K SMT 0805 Format 5 tol 0 1W Resistor 270K SMT Format 196 tol 0 1W Resistor 270K SMT Format 196 tol 0 1W Resistor 100K SMT Format 196 tol 0 1W Resistor Resistor Resistor Resistor Tilt switch non mercury max diff angle 15 degrees Switch SPST momentary action 4 8x5 0mm SMD 24V 100mA Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section SOCKET102 Connector 1 s
18. the circuit works adequately given the power filtering and decoupling suggested in transceiver IC datasheet Regulator Control Very low quiescent current and dropout voltage from the voltage regulator is a requirement for long battery life Quiescent current is the current in the regulator ground pin that is essentially wasted Since the voltage regulator is always connected to the batteries it must be an absolute minimum to prolong battery life In simplified terms dropout voltage is the voltage difference across the regulator when the battery starts to run low The higher the drop out voltage the higher the minimum battery voltage required for reliable operation A lower dropout voltage translates into extra battery life when the batteries begin to go flat The two regulators are U107 and U108 in Figure 51 on page 180 The Complementary Metal Oxide Semiconductor CMOS TelCom TC1073 regulator U108 has an unacceptably high 50uA quiescent current when operating but an excellent 25mV dropout voltage at 30mA load The Motorola MC78FC30 73 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 7 74 regulator U107 has a very low 7uA quiescent current but an unacceptably high dropout voltage The solution is to use both The TC1073 has a shutdown function which turns it off This reduces quiescent current to lt 0 luA When the batteries are low it is switched on reducing the dropout volta
19. 0 000 mAHr Day Target Volume Rate 0 00 mL Day Sensor Data Light Intensity 89 2 of Full Sun Temperature From Monitor 11 01 deg C Temperature From Pressure Sensor 11 10 deg C Pressure 1009 8 mbar Motion Cycles 10 0 cycles Motion Open 34 2 Communication Data Cow Number Proximity no Cow Number Proximity yes Cow Number Proximity yes Cow Number 4 Proximity no Anonymous Cow Proximity no 5 DMU 4 provides user Gah ER prompt for next command 4 input gt 112 Development of a Practical Drug Delivery Communication and Monitoring System 3 13 2 Variable Rate Delivery Testing Volume Rate Vs Pressure 35 30 25 J 20 4 Actual Volume Rate mL Day oo ve PO ERI TUUM MICA 1090 1100 1110 1120 1130 1140 1150 1160 1170 1180 Internal Syringe Pressure mbar Data points taken over 3 days at 38 6 C from two separate pressurisation depressurisation cycles while ambient pressure varied between 1018 and 1031 mbar Excludes data points during initial pressurisation Figure 24 Volume and Pressure During Bench Test of Variable Rate Delivery The graph above demonstrates the relationship between internal syringe pressure and volume rate Relating pressure directly to rate does not work well at high or low rates and would be a poor method of directly controlling drug delivery because of this The next graph demon
20. 4 a E E J wy 2 o l 80 X T 950 M i o 60 2 E 900 2 Temperature l E p 14 5 i 7 i 00 gt 850 7 1 20 MM Lil Estimated Volume Delivered p x i i i iv vo 800 0 7265 O 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Continuing from the previous diagram at point E this point shows a sharp decrease and recovery E when the DMU was removed for observation When removed the hydrogen cools and contracts very quickly resulting in this pressure decrease During the long period where the gascell was held off by external control the pressure drops down F toward ambient When the catch up phase starts pressure increases rapidly back to the delivery pressure Again the hydrogen is cooler when out of the cow resulting in contraction of the gas The delivery G programme also finishes at this point meaning no further hydrogen is produced Figure 38 Downloaded Results Pressure Temperature and Estimated Volume Delivered Full Duration 136 Experimentation 4 3 8 Cow Proximity Results Cow Proximity Recorded By DMU 3 Day of Trial Cow ID Number Detected 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Figure 39 Downloaded Results Cow Proximity Recorded By DMU 3 In these cow proximity graphs the base station is represented with an ID number of zero The reception just after time equals zero is a manu
21. 41 no 10 937 942 Simatec AG Miniature Gas Production Device 2001 Simatec AG Soede N M W Hazeleger J Broos and B Kemp 1997 Vaginal temperature is not related to the time of ovulation in sows Anim Reprod Sci 47 no 3 245 252 Sternesjo A C Mellgren and L Bjorck 1995 Determination of Sulfamethazine Residues in Milk by a Surface Plasmon Resonance Based Biosensor Assay abio 10 1006 175 181 Tang Z B Smith J H Schild and P H Peckham 1995 Data transmission from an implantable biotelemeter by load shift keying using circuit configuration modulator IEEE Transactions on Biomedical Engineering 42 no 5 524 528 Verkerk G Vaginal Temperature During Silent Oestrus in Cows Unpublished work 2002 Walker W L R L Nebel and M L McGilliard 1996 Time of Ovulation Relative to Mounting Activity in Dairy Cattle J Dairy Sci 79 no 9 1555 1561 217 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 218 98 Webster J 1993 Understanding the Dairy Cow 2 ed Blackwell Science 99 Wehner G R C Wood A Tague D Baker and H Hubert Efficiency of the Ovatec Unit 100 101 102 103 104 105 106 for Estrus Detection and Calf Sex Control in Beef Cows Staff Paper SP 96 004 1996 Northeast Missouri State University Agricultural Science Division Williams J Low Voltage Embedded Design Intel Corporation Application Note AP 477
22. Acknowledgements Alltech Communications Providing OPNET network simulation software free of charge for this project Dr Craig Bunt InterAg Providing instruction in the theory and practice of experimenting with controlled drug release and its relation to the estrous cycle in cows Dr Michael Rathbone InterAg Joint Supervisor of this project and Director of Research at InterAg who have given freely of their time and resources to make this project possible InterAg supplied the vast majority of material resources for the development and construction of devices used in experimentation Mark Watt M2 Technology Ltd Providing the M21 real time operating system for the M16C microcontroller free of charge for this project Professors Dick and Mary Earle Massey University Providing the Dick and Mary Earle Scholarship in Technology without which I would not have pursued this degree Mike Benson Mitsubishi Electric Australia Extraordinary effort in providing components not normally available and for providing the KNC30WA compiler assembler for the M16C microcontroller free of charge for this project Dr Rod Claycomb Sensortec Advice on the application of the robotic milking machine cow reproductive theory and thesis editing Dr Dale A Carnegie The University of Waikato Joint Chief Supervisor Special help in thesis composition and editing along with practical advice throu
23. D DMU Bill of Materials Appendix D DMU Bill of Materials Table 11 DMU Bill of Materials Cost Manf Part Supplier Designator Description Manufacturer Supplier Top mark Comment Function Ex GST Number Order Code NZ B201 Battery button cell 11 5mm EPX76 Energizer Cory s EPX76 EPX76 Or other Electronics power 3 38 1 5V Energizer brand Diameter Electrical equivalent silver supply 1 of 3 silver oxide EPX76 Button Wholesaler oxide battery Cell B202 Battery button cell 11 5mm EPX76 Energizer Cory s EPX76 EPX76 Or other Electronics power 3 38 1 5V Energizer brand Diameter Electrical equivalent silver supply 2 of 3 silver oxide EPX76 Button Wholesaler oxide battery Cell B203 Battery button cell 11 5mm EPX76 Energizer Cory s EPX76 EPX76 Or other Electronics power 3 38 1 5V Energizer brand Diameter Electrical equivalent silver supply 3 of 3 silver oxide EPX76 Button Wholesaler oxide battery Cell C101 Capacitor 100nF X7R 100nF 0603 Philips Farne 432 210 MCU power supply 0 17 16V SMD Ceramic Components decoupling C102 Capacitor 100nF X7R 100nF 0603 Philips Farnel 432 210 MCU AVss pin 0 17 16V SMD Ceramic Components decoupling C103 Capacitor 8 2pF 596 8 2pF 0603 Philips Farnel 721 955 For 32kHz MCU sub 0 13 NPO SMD Ceramic Components clock C104 Capacitor 8 2pF 5 8 2pF 0603 Philips Farnel 721 955 For 32kHz MCU sub 0 13 NPO SMD Ceramic Components clock C105 Capac
24. Delivery in Farm Animals Cost F Manf Part Supplier Designator Description Part Type Manufacturer Supplier Top mark Comment Function Ex GST Number Order Code NZ PIN203 Connector 1 pin SIL PIN CON RS 227 2639 price is for 1 pin Battery PCB 0 24 header pin strip single Battery PIN1FP of the 32 way interconnect row straight 32 way break off strip circular cross section PIN204 Connector 1 pin SIL PIN CON RS 227 2639 rice is for 1 pin Battery PCB 0 24 header pin strip single Battery PIN1FP of the 32 way interconnect row straight 32 way break off strip circular cross section PIN205 Connector 1 pin SIL PIN CON RS 227 2639 rice is for 1 pin Gascell PCB 0 24 header pin strip single Gascell PIN1FP of the 32 way interconnect row straight 32 way break off strip circular cross section PIN206 Connector 1 pin SIL PIN CON RS 227 2639 rice is for 1 pin Gascell PCB 0 24 header pin strip single Gascell PIN1FP of the 32 way interconnect row straight 32 way break off strip circular cross section PIN207 Connector 1 pin SIL PIN Rx CON RS 227 2639 rice is for 1 pin Rx Data PCB 0 24 header pin strip single Data PIN1FP of the 32 way interconnect row straight 32 way break off strip circular cross section PIN208 Connector 1 pin SIL PIN Tx CON RS 227 2639 rice is for 1 pin Tx Data PCB 0 24 header pin strip single Data PIN1FP of the 32 way interconnect row straight 32 way break of
25. Design Development and Optimisation of Veterinary Intravaginal Controlled Release Drug Delivery Systems The University of Waikato Olejnik O C G Wilson J Stevens and J G Hardy 1985 Ocular Retention of Ophthalmic Vehicles in Rats Evaluated by Gamma Scintigraphy J Pharm Pharmacol 37 118 Pennington J A J L Albright and C J Callahan 1986 Relationships of sexual activities in estrous cows to different frequencies of observation and pedometer measurements J Dairy Sci 69 no 11 2925 2934 Plade Holdings SMARTT1 Intelligent Breeding Device 1999 Plade Holdings Producer Board Project Team Producer Board Project Team Briefing for the Incoming Minister 1999 Promar International International Forces Shaping The New Zealand Dairy Industry 2001 Promar International Radio Spectrum Management Department of the Ministry of Commerce RFS29 Specification for Restricted Radiation Radio Apparatus 1998 Rasby R and G Deutcher Synchronizing Estrus In Beef Cattle 2001 University of Nebraska Lincoln Institute of Agriculture amp Natural Resources Rathbone M J 2000 The pharmaceutical scientists role in animal production An historical and future look at the evolving area of controlled drug delivery in animal production 215 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 216 7T 78 79 80 81 82 83 84 85 86 Rathbone
26. Drug Delivery and Monitoring Units The intended function operation and construction of the DMU s and base station have been extensively described in the previous chapter Schematics printed circuit board layout and bill of materials including individual component descriptions and their purpose are presented in appendices A through E Electronic versions of these and the source code for the programs are provided on the accompanying CD ROM inside the back cover 4 1 3 CIDR Each DMU and the control syringe had an active CIDR attached for retention 119 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 4 2 120 purposes Using an active CIDR meant that the cows would be prevented from reaching estrus during the experiment This meant that behavioural and physiological changes due to estrus were not expected to occur e g vaginal temperature peaks due to ovulation would not occur 4 1 4 LiveLube LiveLube was used as a lubricant during DMU insertion 4 1 5 Animals The five cows were part of the Dexcel Ltd research trial herd on the Number 4 station on Ruakura Road Hamilton New Zealand All cows were of the Friesian breed between two and five years old weighing approximately 500kg each Ovaries were intact and the cows had been cycling normally prior to the experiment Methods To answer the thesis objectives the delivery and monitoring performance of the DMU was demonstrated in a
27. ER REEERE TEENE 121 Location of DMU in Cow Adapted from Webster 1993 2 ssses 122 Variable Rate Profile Used In DMU s 1 and 2 sss 123 Sample Graph for Familiarisation eene 125 Downloaded Results Temperature esee 127 Downloaded Results Temperature in Frequency Domain 128 Flat Profile Target Estimated and Observed Drug Volume 130 Flat Profile Target Integrated Current and Target Integrated Current Dc M a 132 Downloaded Results Gascell Voltage and Current 133 Downloaded Results Pressure Temperature and Estimated Volume Dehveted Closes Upias osos uestes uH eus e SE eet 135 Downloaded Results Pressure Temperature and Estimated Volume Delivered bulbDutgatioti niece e p cie eeu 136 Downloaded Results Cow Proximity Recorded By DMU 3 137 Downloaded Results Cow Proximity Recorded By DMU 1 138 Downloaded Results Cow Proximity Recorded By DMU 2 138 Downloaded Results Motion eene 139 Downloaded Results Light Intensity esses 141 DMU Power Supply Voltages essen 142 Discharge Curve of Energizer EPX76 Silver Oxide Battery Adapted from EPX76 data sheet Eveready Battery Company 2001 e 143 Closed Loop Drug Delivery Claycomb 2001 e
28. ESR 3 OHMS 100Hz 3 2x1 6mm Capacitor 47uF 4V tantalum SMD case size B ESR 2 4 OHMS 100Hz 3 5x2 8mm Capacitor 10uF 10V Tantalum SMD case size A ESR 3 OHMS 100Hz 3 2x1 6mm Capacitor 100pF 5 NPO SMD Ceramic Capacitor 27pF SMD 5 NPO Ceramic Capacitor Capacitor 15nF SMD 5 NPO Ceramic Capacitor 15nF SMD 5 NPO Ceramic LED Dual red green SMD 3x2 1mm Diode Switching Schottky barrier 200mA max forward V drop 100uA 240mV Fuse 125mA FF very fast acting Part Type 100nF 100nF 10uF 10WV 15nF 10uF 10WV 47uF 4WV 10uF 10WV 100pF 27pF not present 15nF 15nF LSG T670 HK BAT254 125mA FF Manf Part Number Foot print 0603 0603 CAP A TAJA106K 010X 0603 CAP A TAJA106K 010X CAP B TAJB476K 004X CAP A TAJA106K 010X 0603 0603 0603 0603 0603 T670 LSG T670 HK SOD110 BAT254 1206 429 125 Manufacturer Supplier Philips Components Philips Components AVX Philips Components AVX AVX AVX Philips Components Philips Components Philips Components Philips Components Siemens Infin eon Technologies Philips Semiconducto rs LittelFuse Farnel Farne Farnel Farnell Farnell Farnell Farnell Farnell Farnell Farne Farne Farne RS RS Supplier Order Code 432 210 432 21
29. Foote and N R Scott 1976 Detection of estrus in dairy cows by electrical measurements of vaginal mucus and by milk progesterone J Dairy Sci 59 no 5 982 985 Gil Z J Szarek and J Kural 1997 Detection of silent oestrus in dairy cows by milk temperature measurement Animal Science 65 25 29 Given Imaging Ltd 2001 Camera Pill An Ingestible Diagnostic Imaging System Available from http www givenimaging com Groning R and U Kuhland 1999 Pulsed release of nitroglycerin from transdermal drug delivery systems nt J Pharm 193 no 1 57 61 Groning R and S Weyel 1993 Electronically controlled release of drugs from capsules Eur J Pharm Biopharm 39 no 3 102 104 Gross J and S Zucker Electrolytic dispensing device US 5186805 1993 Gurny R V Kaltsatos V A A Deshpande M Zignani C Percicot and V Baeyens V 1997 Ocular drug delivery in veterinary medicine Adv Drug Deliv Rev 28 no 3 335 361 Hacisalihzade S S 1989 Control Engineering and Therapeutic Drug Delivery EEE Control Systems Magazine 9 no 4 44 45 Herres C R Electrochemical controlled dispensing assembly and method US 5368571 1994 Howell T D 1998 Statistical Properties of Selected Recording Codes BM Journal of Research and Development 33 no 1 60 73 References 44 45 46 47 48 49 50 51 22 53 54 InterAg 2000 Estrous Synchrony in Cattle Available
30. Highlights of the microcontroller are 16 bit architecture designed specifically around efficient execution of C code 4 stage pipeline Special low noise IO pin hardware design and packaging for high ElectroMagnetic Interference EMI acceptance and low EMI emissions Low power 2uA in sleep mode with 32kHz sub clock operating timers 40pA executing code with 32kHz sub clock 7mA executing code at 10MHz divide by 1 no wait state bus 10MHz operation down to 2 7V 256K flash memory which is in system and self re programmable 20K Random Access Memory RAM 14x14mm Low profile Quad Flat Pack LQFP 100 pin package with 0 5mm lead pitch 95 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 96 INTERNAL 10 BIT EIGHT 8 PERIPHERAL FUNCTION TO 10 CHANNEL SYSTEM ADC CLOCK GENERATION Xin Xout 16 BIT TIMER TAO 8 BIT THREE CHANNEL XCin XCourt 16 BIT TIMER TA1 SIM IER UART CLOCK 16 BIT TIMER TA2 SYNCHRONOUS SIO 16 BIT TIMER TA3 16 BIT TIMER TA4 CRC OPERATION 16 BIT TIMER TBO CIRCUIT CCITT 16 BIT TIMER TB1 16 BIT TIMER TB2 M16C 60 SERIES CPU CORE REGISTERS 15 BIT 19 0 8 WATCHDOG TIMER PROGRAM TWO CHANNEL COUNTER DMAC VECTOR TABLE INTB 8 BIT TWO STACK POINTER CHANNEL DAC TE MULTIPLY 8 CIRCUIT 15 0 Figure 16 Mitsubishi M16 Microcontroller Block Diagram Mitsubishi Electric Corporation 2001 3 8 2 Real Time Clock A timer runs off the 32kHz sub clock w
31. M J C R Burke C Ogle C R Bunt S Burggraaf and J A Magruder 2000 Design and development of controlled release intravaginal veterinary drug delivery systems In Controlled Release Veterinary Drug Delivery edited by M J Rathbone and R Gurny 1 ed Amsterdam Elsevier Science Rathbone M J P J Burns C Ogle C R Bunt and S Burggraaf 2000 Controlled release drug delivery systems for control of domesticated livestock In Controlled Release Veterinary Drug Delivery edited by Rathbone M J and R Gurny 1 ed Amsterdam Elsevier Science Rathbone M J and R Gurny 2000 Controlled Release Veterinary Drug Delivery Edited by Rathbone M J and R Gurny 1 ed Amsterdam Elsevier Science Rathbone M J J E Kinder K Fike F Kojima D Clopton C R Ogle and R Bunt 2001 Recent advances in bovine reproductive endocrinology and physiology and their impact on drug delivery system design for the control of the estrous cycle in cattle Adv Drug Deliv Rev 50 no 3 277 320 Redden K D A D Kennedy J R Ingalls and T L Gilson 1993 Detection of estrus by radiotelemetric monitoring of vaginal and ear skin temperature and pedometer measurements of activity J Dairy Sci 76 no 3 713 721 RF Monolithics Inc ASH Transceiver Designer s Guide 2001 05 05 2001 RF Monolithics Inc Riviere J E and M G Papich 2001 Potential and problems of developing transdermal patches for veterinary
32. Nervous excitable and restless More likely to ride other cows and be ridden May stand with back arched and tail in air Spend less time than usual grazing Display moist red vulva Clear mucus discharge from vulva May hold milk Background Frequently change from their usual order coming into the shed often leading or lagging Visually detecting all standing heats is next to impossible though because standing behaviour represents less than 1 of the total heat period This fact has promoted the development of systems to detect this mounting behaviour which then bring it to the attention of the farmer There are two complicating factors that can make this difficult One is the fact that silent heat is common among high producing dairy cows Shipka 2000 66 153 159 The term silent heat refers to the condition where the cow ovulates but displays much a reduced observable behaviour The other difficulty is that many of the easily detectable signs of estrus are effects rather than causes For example a slight increase in body temperature is one effect of estrus but this can also be caused by an increase in ambient temperature or the onset of illness Fordham McCarthy and Rowlinson 1987 11 367 379 Ideally we want to detect causative factors for greater accuracy Methods of Estrus Detection Commercially available heat detection aids for on farm use are based around the detection of behavioural sig
33. Theratron is a purpose built syringe incorporating the gascell originally intended for human portable drug infusion Simatec AG Switzerland invented both the Theratron and gascell Different models are available which correspond to the particular infusion rate required The only difference is the fixed resistor the value of which determines the rate of delivery The Theratron is a low risk technique for fluid delivery As such it is an ideal mechanism to deliver the pharmaceutical vehicle in the DMU The syringe is modified to allow the antenna transmission line to exit through the Development of a Practical Drug Delivery Communication and Monitoring System pressure release mechanism while maintaining a relatively hydrogen gas tight seal Headroom for movement above the gascell allows the whole electronics assembly to slide in relation to the rest of the syringe body should the over pressure release cap activate This can be seen in Figure 22 on page 110 3 5 2 Drug Delivery Profile Management Internal Representation of Delivery Profile The target volume rate profile is stored as a rate of volume versus time graph inside the DMU The actual numbers prepared by the user represent the vertices on a trace consisting of straight line segments The smallest allowable horizontal distance of the straight line segments is one minute allowing the graph to closely approximate any desired curve Target Volume and Rate of Deli
34. a count source for a counter on the microcontroller Motion percent closed is the percentage of time that the ball has stayed in contact with the cage contacts The intention is to provide an indication of mounting behaviour This is accomplished by establishing the tilt switch as a gate signal for a counter on the microcontroller The free running 32 768kHz oscillator is on even when the microcontroller is in sleep mode for the purposes of maintaining the real time clock This oscillator also provides the count source for the motion percent closed counter only decrementing the counter when the gate is closed 101 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 102 The quality of information is not expected to stand alone in determining estrus but to act as an indicator in conjunction with other systems to increase accuracy of detection The data obtained could be processed in a number of ways In the introduction it was reported that cows in heat typically walk two to five times more than when they are not in heat Nebel and Walker 1998 The Afikim pedometer system takes the rolling average over the last ten days and compares it to the present day A 70 increase over the 10 day rolling average is thought to be a reasonable indicator of increased activity due to estrus but even then the results need interpretation on a cow by cow basis Further interpretation is required if the pedometer counter
35. and Monitoring of Intravaginal Drug Delivery in Farm Animals 4 4 144 presently defined load profile and a reasonable download usage of one minute per day If the user only downloaded once a week seven weeks of operation could be expected in most circumstances The reason that a more precise answer cannot be given is that it greatly depends on the usage pattern of the device and the fact that battery voltage is often not a good indicator of the amount of remaining battery capacity Battery voltage is influenced by the following factors The voltage curve of silver oxide batteries is very non linear as demonstrated in Figure 45 Battery voltage drops on a heavy load then recovers slowly so the recent load history affects the instantaneous voltage reading Voltage is affected by temperature to a large extent as demonstrated in Figure 44 If the sampling and download usage was very much less frequent it is predicted that the batteries would last over a year In practice three weeks is more than adequate since insertion time is restricted due to animal welfare considerations Experimental Results Summary DMU performance has been adequately demonstrated Although three of the four DMU s had stopped working by the last day enough data was gathered to prove the capability of the system All systems worked very well except for the light intensity function This had a restricted dynamic range due to the programmin
36. applications Adv Drug Deliv Rev 50 no 3 175 203 RS Components Ltd Radio Spares New Zealand Catalogue April 2000 2000 RS Components Ltd Santini jr J T M J Cima and R S Langer Microchip drug delivery devices US 5797898 1998 Santini J T R Langer and M J Cima 1999 A controlled release microchip Nature 397 no 6717 335 338 References 87 88 89 90 91 92 93 94 95 96 97 Santini J T A C Richards R A Scheidt M J Cima and R S Langer 2000 Microchip technology in drug delivery Ann Med 32 no 6 377 379 Scanlon W G N E Evans and Z M McCreesh 1997 RF performance of a 418 MHz radio telemeter packaged for human vaginal placement JEEE Transactions on Biomedical Engineering 44 no 4 427 430 Shah R A Y Bekker and S Reisman 2000 Development and testing of a simulated closed loop drug delivery system for congestive heart failure patients under milrinone administration Proceedings of the IEEE 26th Annual Northeast Bioengineering Conference 2000 49 50 Shipka M P 2000 A note on silent ovulation identification by using radiotelemetry for estrous detection Applied Animal Behaviour Science 66 153 159 Shults M C R K Rhodes S J Updike B J Gilligan and W N Reining 1994 A Telemetry Instrumentation System for Monitoring Multiple Subcutaneously Implanted Glucose Sensors IEEE Transactions on Biomedical Engineering
37. ar 211 2aa 2 tania m m ui i zo a aoro ES d ING enpe ai uo pasn ye oF Ball 7 mn posue aro sqers amp m xoop qns pue weur at En preoq 1on inep ima Ajpiqueduroo uweze pue uaimo spout ale Pon m daajs 1821 01 819208 OJ Aq pa edor aan SWA PIJEN m suoreonrpo q E nuod ql t t 7 Figure 60 Modified M16C Evaluation Board Schematic Sheet 1 200 Appendix E Base Station Development DMU IPS T PNS prog EAT OOTIN PIPON Inr CEEE bv1ejul Fe Jo c vos ISOTPI ang rooc m or aeq I worst p roqumy tV cag panog paz 9I papoj w Ca cea Land di DING PU ap uo pasn yx 0 posue aro spejs amp m Yo 9 qns pue ureur aur pweoq 1ep anep wen Knpqueduroo weye pue juan spout daos 189 0 9320 DT Aq paoeydas MIM SWIN PIJIN suonesgipoy 2K lt am o Figure 61 Modified M16C Evaluation Board Schematic Sheet 2 a e 201 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Ys HOA pameog prg J9 6y1 u E 30 rwegs zrILbrp oump 1007 moray raqmy papog paq DOT parfipoyy ow IMS EAS n zi ldr t oie TLF qu i Q0 DODODO l aas aw ail ind QIN PM oi uo posn jp 0 posue 3124 se s A12 YOoTI qns pue Wew ILI Aymrqueduro wene pue quarmo apour daaqs 1321 0 S1908 JI Aq paouydar 213M SUJ PIJTEJN Suone po K Figure 62 Modified M16C Eval
38. current 120pA While integrating current at 36Hz and listening once a second for wireless data Peak current 17mA Electronics package 28 6x28 6x25 0mm Dimensions Body 30x30x150mm including drug reservoir but excluding antenna and retention device Payload 40mL Controlled 60mL day sustained up to 120mL day in short Maximum rate Release bursts Delivery voume 4 3 or 0 5mL whichever is greater accuracy Transceiver Transmit output power 6dBm 0 25mW Range 40m outside typical Raw data rate 23kBPS Modulation On Off keying Carrier frequency 433 92MHz ames tony soon Receiver dynamic 10 to 84 dBm range 4 Benchtop testing at constant temperature with pressure allowed to vary 5 mbar 117 Experimentation 4 Experimentation 4 1 Materials 4 1 1 Vehicle Formulation The delivered vehicle was aqueous HydroxyPropyl MethylCellulose HPMC solution The HPMC was United States Pharmacopoeia substitution type 2910 Shin Etsu Chemical Co Ltd at 4 weight per volume in double distilled water The manufacturer s data sheet gave the viscosity of the solution as 4000 cP at a body temperature of 38 6 C This formulation had previously been used as a vehicle for progesterone in the Theratron syringe Bunt Rathbone Ogle and Morgan 2001 28 but on this occasion was used without any pharmaceutically active agents This blank vehicle simulated the administration of a typical drug in solution 4 1 2
39. delivery profile external control was obtained over the wireless link The internal program was overridden on day 2 by an external command to stop delivery for a 24 hour period This process was repeated again for a 48 hour period starting on day 4 129 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Flat Profile Target Estimated and Observed Drug Volume Day of Trial 1 0 1 2 3 4 5 6 7 35 Volume mL 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Estimated Volume Observed Volume from Piston Position Target Volume mL with max 796 1 0 mask mL Basic in vivo delivery accuracy is 7 or 1 0 mL whichever is greater A mask of these dimensions is shown in light blue The DMU fell out of the cow during the last few hours This resulted in a piston position error in the last actual volume delivered point Figure 34 Flat Profile Target Estimated and Observed Drug Volume The number of removals for piston position measurements were kept to a bare minimum to reduce stress in the animals This means that there are only five check points on the actual volume delivered graph The points that were obtained prior to day 6 are within 7 of the desired delivery target The light blue lines are a mask of 7 or 1 0mL whichever is greater The method for determining actual volume dispensed consisted of measuring the piston
40. fluids themselves carry the modulated current induced by the transmitter Lindsey McKee Hull and Howell 1998 45 614 619 Electrodes need only be placed on the surface of the skin without penetration to effect communication with an embedded device This a technique requiring direct contact mentioned more for interest than for consideration in this application A cow ID system using this technique has been developed and uses the farmer s body to complete the connection between cow and reader There were some problems reported with communications failure during drier times of the year In addition cows are very sensitive to stray voltage and can be irritated by as little as 4V having an unusually low impedance to earth Ultrasound Cheap ultrasonic transponders would not tolerate being submersed and their frequency of operation would be affected even if not permanently damaged Inductive Capacitive Coupling The inductive technique is a well proven data transmission method for embedded biotelemetry devices Jeutter 1983 2 17 24 Nevertheless these are short range effects used effectively over a distance of only three metres at most with miniaturized circuits As such it is unsuitable for the current application requiring longer range However this method of communication might be well suited to a device in say the rumen if short range can be tolerated The inductive method of communication has the advantage of being able t
41. for a simplified inertial navigation to calculate path length FitSense Technology 2001 It uses a radio frequency IC from the same family as that used in the DMU No doubt these ideas are already under evaluation for commercial development but cost is the biggest barrier Perhaps estrus control is too narrow an application for all of this technology to be deployed at once The inclusion of an ID function is good example of how extra value can be included for little extra cost per cow to make some form of cow mounted information appliance a viable platform Conclusion 6 5 Conclusion 6 5 1 All Objectives Completed Advances in cow estrus detection and synchronisation via controlled drug release can provide efficiencies in the dairy industry Electronic control and monitoring of intravaginal delivery in farm animals can help meet this need The objectives of this thesis emphasised the need for an intelligent communicating device with the capability to Precisely control a variable rate delivery profile Provide external control of the delivery system Monitor the animal and delivery process A programme of design construction testing and experimentation proved the technical feasibility of such a system External control and monitoring of drug delivery in farm animals was successfully carried out in a week long trial with five cows Methods problems and solutions involved in developing and using the system were described Or
42. from www interag co nz Jeutter D C 1983 Overview of Biomedical Telemetry Techniques Engineering in Medicine and Biology 2 17 24 Kettlewell P J M A Mitchell and I R Meeks 1997 An implantable radio telemetry system for remote monitoring of heart rate and deep body temperature in poultry Computers and Electronics in Agriculture 17 161 175 Koelsch R K D K Aneshansley and W R Butler 1994 Analysis of Activity Measurement for Accurate Oestrus Detection in Dairy Cattle J Agric Engng 58 no 107 Kyle B L A D Kennedy and J A Small 1998 Measurement of vaginal temperature by radiotelemetry for the prediction of estrus in beef cows Theriogenology 49 no 8 1437 1449 Lefcourt A M and W R Adams 1998 Radiotelemetric measurement of body temperature in feedlot steers during winter J Anim Sci 76 no 7 1830 1837 Lefcourt A M J Bitman D L Wood B Stroud and W D Schultze 1993 Radiotelemetry temperature responses of mammary gland and body to intramammary injection of Escherichia coli endotoxin or Streptococcus agalactiae in lactating dairy cows Am J Vet Res 54 no 5 798 804 Lewis G S and S K Newman 1984 Changes throughout estrous cycles of variables that might indicate estrus in dairy cows J Dairy Sci 67 no 1 146 152 Li V H R W Wood J Kreuter T Harmia and J R Robinson 1986 Ocular drug delivery of progesterone using nanoparticles J Microencapsul
43. generation receiver interrupt processing noise spike rejection and initialisation of the device configure port pins in the appropriate time sequenced manner Above this the middle layer provides higher general purpose functions such as transmitting an array of byte values This involves the anti collision algorithm transmit mode setting preamble generation and transmission of bytes using the BIOS routines The top level can then interact with this Application Programming Interface APT at a higher level That is the application layer can deal with the relatively abstract concepts of sending and receiving messages rather than having to worry about the state of individual port pins or even which medium it is transmitting on wired wireless or both 105 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 11 Base Station 106 The purpose of the base station is to relay commands typed into the laptop for transmission over the wireless link As shown in Figure 20 the commands are parsed and sent to the transceiver over the RS232 serial cable at 57 600 Baud DMU reply packets are captured by the base station transceiver and reformatted for serial transmission over cable in the reverse direction to the laptop The wired link is duplex the transceiver link is half duplex at 23k Bits Per Second BPS The base station hardware is similar to the DMU It consists of the same components and schematic
44. is retarded by 1296 of its maximum value The assumption is that jitter is a random variable with a mean of zero Bit synchronisation is achieved during preamble ready for the first valid data pattern which is a byte synchronisation pattern No end of frame code is required since the message length is given in the packet header The output from the data slicer is sampled 10 times during the bit period and counts over 5 are considered as ones Byte Synchronisation The byte synchronisation pattern is a reserved 10 bit pattern from the 8 bit to 10 bit encoding scheme that does not occur anywhere in the actual data stream During the preamble phase each new incoming bit is left shifted into a register and the old one discarded At the end of every bit period the register is checked to see if it is the synchronisation pattern If it is the right pattern bit and byte synchronisation has been achieved Automatic Repeat Request Some communication protocols use a version of Automatic Repeat reQuest ARQ to ensure that the message gets through if not correctly received on the first try This can only be implemented in systems with a back channel The receiver may be out of range or find some unrecoverable errors in the data If the transmitter does not receive an acknowledgment from the receiver within a certain amount of time it re sends the information a few more times before giving up Many elaborate variations on this theme have bee
45. it is already collected from the dairy cow for its intrinsic value By monitoring for blood somatic cell count percentage solids and other factors an automatic treatment response could be carried out For example a high somatic cell count may indicate mastitis prompting the milking parlour management computer to automatically deliver anti biotic treatment This is an over simplified example and such a treatment would not be possible given this amount of information but it illustrates the general principle of what might be possible in future once more information is being collected from each milking on a routine basis and can be applied to automated controlled release drug delivery 161 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals systems Other applications include the automated release of drugs to control ectoparasites Vaginal temperature motion sensing inter cow grouping drug delivery Wireless network Base Station ne ww bj ID and a pedometer Milk pue Fully automated drug delivery sense control loop Relay Generic robotic milking outputs machine and herd management software Milk yield conductivity somatic cell count progesterone levels cow attendance activity Figure 46 Closed Loop Drug Delivery Claycomb 2001 162 A limited form of feedback is already a part of commercial reality in the Fullwood Conclusion 6 3 Lt
46. minus the DMU peripherals and with the addition of some dedicated RS232 level conversion In fact the development DMU board becomes the base station simply by downloading the base station software which ignores the extra peripherals This development board is documented more fully in Appendix E starting on page 199 Development of a Practical Drug Delivery Communication and Monitoring System Transceiver Packet Parser Base Station C ommand Parser 57600 Baud Serial Port Local Echo 23k Bits Per Second Wireless Link Transceiver v Command Parser Message Centre Command 1 Command 2 Command 3 Command N Packet Assembler E 57600 Baud Wired Serial Port for Diagnostics Local Echo v Figure 20 Base Station DMU Communication Flow 107 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 12 Construction 108 The hardware and software design is standard but the difficulty lies in finding a combination of devices that fit the special needs of this application These devices have to be compatible in terms of Small physical size Low power operation Obtainable in sample quantities Evaluations kits and development tools priced within tight budget Electromagnetically compatible with the transceiver in close proximity Layout is crucial to t
47. of Intravaginal Drug Delivery in Farm Animals TPS SHW pE DADNE md PUDjoaz man STSCFT nL TOOTA uoynungs DAMN Dayo ezre Sog ajwug Syaaqur 3 d d INNON X 09 TOZAM pou sou yeq ou p pamos Ano parpi pamo 20 Pasty ywgor pawmi As RET AS a Imod EMP T pamb s se sanea mpo Siero pue dir sozay wo 0s FOCNId cota zus F z AVM STE org 1 3pmo APS ASST 2 F s 9 Xd3 BABB X scLVd T 10a a F rmgeg NId OTT Adoot 7073 PPA AMOT MOT 1079 LOTNId end Xa Nid quasaud you 907NId p2stD NId THOSvp Toco Tone OCNId Tposto NId E TOTNId UND Nid TOCNId PPA Nid Transceiver and Batteries Board igure 53 DMU Schematic F 182 Appendix B DMU Schematics que ww iJ i ww lffcv aL Bus 2 X NT vroqun miles a qn SunumauSoig YA FL IM Cea NI S19 1NO SLa gt Cal NDHSD B 3S 8 LINDH s 1no cxi CA NI CXM CIR Coq 94 T1349 eM 9 9u 8nid ant 29 Ltd A 19 BY2122EXYN anez a 92 19 z T t9 Liud APO 6701747N i in m4 eu 3n Ip x1 129441 i LC tes BL amp Jz B5 Ll n EJ ow SI 8 tU E in bie EJ a ME p ZlHE UH E 5 E 11 sa eINO2 ax SSANO 1353 NS 1 hd3 CH 32 axe 198 2 Asna NOMW2OA e Program
48. only necessary to download the results at the end of the 7 day trial because there is enough 123 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 4 3 124 memory for 720 sets of readings In practice downloading was carried out once a day in case the devices failed unexpectedly To demonstrate the external control ability the delivery program for DMU 3 was manually overridden via the wireless link The first external control period aimed to inhibit delivery for a 24 hour period The second external control period aimed to inhibit delivery for a 48 hour period Results and Discussion The following results consist mainly of graphs generated by downloading logged data from DMU 3 over the wireless link during the experiment DMU 3 was the only DMU to last the whole week DMU 1 and 2 stopped communicating after 48 hours DMU 4 stopped communicating after the first day Further observations relating to these devices and the control syringe are presented in Appendix A Additional Experimental Results Most of the data was downloaded during the daily observation period with the cows confined to bales Data was also successfully downloaded from an adjacent paddock with the cows going about their normal activities 4 3 1 Trial Conditions The first day of the 7 day trial was 23 May 2001 Throughout the experiment ambient temperature ranged from a minimum of 1 C at night to a maximum of 16 C th
49. other countries means that the New Zealand dairy industry cannot rest on its laurels It has been generalised that other countries cost of producing milk solids is decreasing at a sustained rate of 2 per annum while New Zealand s cost is rising at 2 This is the basis behind the Meeting the 4 challenge campaign Dexcel Ltd 2001a to make New Zealand dairy farmers realise that they must continually strive for increased efficiencies Current global trends affecting the profitability of New Zealand s dairy and related industries are Producer Board Project Team 1999 Long term decline in commodity prices Rising consumer food safety concerns Demand for more product traceability Retail consolidation Future additional costs may increase as a result of The need to meet tougher environmental constraints such as reducing nitrogen run off in local water ways The impact of decisions regarding genetically modified organisms Background The introduction of greenhouse gas taxes methane emissions from animal digestion Developments in e commerce Changing consumer preferences and practices 2 3 Increasing Dairying Efficiency Through the Use of Controlled Drug Release By reviewing past trends and future predictions it has been established that the New Zealand dairy industry must continually strive for increased efficiencies if it is to remain successful The application of controlled drug release to
50. packet This is the mechanism by which the individual proximity of nearby cows are logged for each recording period Gascell Override Commands The gascell switch can be controlled by the gascell override value command if external control has been obtained through the gascell override command This is subject to the safety limit checks specified on page 57 Start Delivery Override Commands The start of delivery can be forced by the start delivery override value command if external control has been obtained through the start delivery override command Restart Delivery Command The delivery programme can be reset to the beginning of the delivery profile Executive Control All control and decision based processing occurs on board the DMU This allows for autonomous control of each DMU in a distributed network of intelligent delivery and sensing devices This is an absolute requirement since communication with the base station is intermittent 3 8 1 M16C Microcontroller To provide the intelligence and memory for the DMU the M30624FGLGP from the MI6C family of Mitsubishi microcontrollers was chosen This is U101 in Figure 50 on page 179 This represented the most capable microcontroller that Development of a Practical Drug Delivery Communication and Monitoring System could be found given the small amount of space available for the electronics and the limited budget for development tools
51. physical drug delivery task it was designed to do However this method of delivery is not suited to development into a more complex device capable of high accuracy and adherence to a complex profile incorporating both high and low delivery rates There are two reasons for this Due to the fact that so much room is taken up by the battery and electronics there is little room left for the main reservoir This means that the drug must be concentrated any error in absolute volume leads to a higher error in dosage Secondly the nature of the spring loaded controlled orifice release mechanism is such that any small blockage could adversely affect dosage after that point 39 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 40 2 5 5 Selected Electronically Controlled and Actuated Variable Rate Drug Delivery Technologies This discussion on devices and technologies is not specific to any animal or human application nor to any route Its purpose is to provide some background on previous experimental work on electronically controlled and actuated drug delivery before considering the design of the drug delivery platform used in this thesis Due to space limitations it covers only a few selected technologies This short review concentrates only on technologies that involve electronics close to the actuation of the dosage form whether it be by direct application of a voltage or via an electrically stimulate
52. piston friction turns this into a smooth dispensing action It is a very over damped response The maximum amount dispensed during a five minute period is around 300uL Finer control is possible by decreasing the gascell decision period An alternative would be to increase the value of the load resistor but this would decrease the maximum attainable rate of delivery Note that the resolution of measured estimated volume delivered is much finer than 300uL discussed later This is a closed loop control system Variability in gascell condition and the Development of a Practical Drug Delivery Communication and Monitoring System value of the load resistor are automatically compensated for The only significant errors in gas production volume are measurement of the integrated current and any syringe behaviour leaking gas etc A slightly defective gascell will produce just as accurate a result it will simply be told to turn on more often to make up for the short fall in expected current while on The practical significance of this is that each gascell can be used twice as long as particularly high volume delivery rates are not required on the second use Gascell Switch Decision The gascell switch decision made by the dispensing scheduler can be overridden by external manual control over the wireless link The exception is when the external command tries to turn the gascell on while safety limits are exceeded Safety checks alwa
53. products exist that exploit temperature by itself for sensing for estrus Surprisingly vaginal temperature rise is not a good estrus indicator for other common farm animals such as pigs Soede et al 1997 47 245 252 Milk Temperature Prediction using temperature of the milk sampled from cows using an on line sensor has met with a degree of success being closely correlated with body temperature Fordham McCarthy and Rowlinson 1987 11 367 379 Gil Szarek and Kural 1997 65 25 29 False positives from udders infected with mastitis can partly be eliminated by comparing the temperature from all four quarters Infection and therefore temperature rise due to mastitis generally occurs on a quarter by quarter basis Electrical Properties of Milk As with on line temperature sensing of milk taken from a milking machine on Background line milk conductivity measuring has also been able to predict estrus with limited success de Mol 2001 126 99 103 Unfortunately it suffers from the same disadvantage being correlated with mastitis Nielen et al 1995 78 1050 1061 Woolford Williamson and Henderson 1998 65 187 198 Electrical Properties of Vaginal Mucus Studies have shown that the occurrence of estrus is accompanied by an increase in electrical conductivity of mucus in the anterior part of the vagina peaking after temperature Bobowiec Studzinski and Babiarz 1990 44 573 579 Gartland et al 1976 59 982 985 Con
54. safe for use inside the cow The tilt switch oscillates when the cow is level and moving staying open circuit when the cow is inclined head up In retrospect mounting the switch so it would be closed when the cow 1s inclined head up would have been more useful because a head up inclination may indicate estrus mounting behaviour However a head up inclination of the device could also be simply due to the fact the cow is lying down at that angle Sloping pastures could also present a problem The maximum angle differential between open and closed is 15 degrees Although nowhere near as sophisticated as commercial devices based on MEMS accelerometers the hope was that it would be good enough for basic motion detection to assess activity level and mounting activity of the riding cow The switch is positioned with its long axis co incident with the long axis of the syringe which means that its behaviour would be the same irrespective of the rotation of the DMU within the cow Two different values are read at the end of each data logging period motion cycles and motion percent closed Motion cycles are the number times the switch contacts have been made and then broken giving an estimate of overall cow activity in the same way that pedometers fitted to the legs or collars of cows have been proven to do The intention is to provide an indication of overall activity This is accomplished by establishing the tilt switch signal as
55. system would be required implying a still greater investment in sensor technology and understanding and modelling of the biological system Such a system would find it difficult to produce an acceptable result under all of the exceptional conditions met in the real world and this might explain why few systems ever reach the stage of formal clinical evaluation Mason Linkens Edwards and Reilly 1996 13 243 252 Applications such as drug infusion for atracurium induced muscle relaxation for eye and brain surgery are safety critical demanding a robust and reliable solution which at present is not well developed Another example is the recent work on a closed loop drug delivery system to maintain cardiac output at set point range for congestive heart patients Again although said to be promising this has not been transferred through to a physical realisation with formal clinical evaluation Shah Bekker and Reisman 2000 49 50 So although externally operated biomedical machine human closed loop control systems are routine for many applications including anaesthesia Linkens and Abbod 1998 Digest No 1998 514 2 1 2 4 totally embedded solutions for drug delivery are in a comparatively early stage No doubt embedded closed loop drug delivery systems for safety critical systems will be successfully developed Hope can be found nearby in the success of similar closed loop technologies which are Conclusion employed in safety critica
56. the vagina a CIDR device see Glossary is attached to the syringe with adhesive and tape Being loaded with progesterone it prevents the cow from coming into estrus A blank CIDR with no progesterone can be used if this effect is undesirable Power Management 3 6 1 Batteries Button cells were chosen for the power source Three Energizer EPX76 silver oxide Zn Ag 0 batteries in series give a fairly flat discharge profile at a nominal on load voltage 4 5V These are represented in the DMU schematic as B201 B202 and B203 in Figure 53 on page 182 The batteries are the same size and shape as the gascell There were no suitable battery clips available so they were fabricated from strips of 0 9mm brass sheet 3 6 2 Low Voltage Power Supply Considerations The voltage from the batteries is regulated to 3V There are advantages and Development of a Practical Drug Delivery Communication and Monitoring System disadvantages operating at 3V On the plus side operating from 3V instead of 5V decreases current consumption by 40 Battery life increases by more than 40 because batteries have a higher effective capacity at a lower discharge rate More importantly power consumption increases as the square of the operating voltage so reducing the operating voltage of a device from 5V to 3V decreases power consumption by 64 Heat production is reduced which is important when trying to accurately measure body temperature from within t
57. to the contagious disease but is configured to be slightly more contagious resulting in just in time vaccination The example of a contagious disease has been used here but this method could equally apply to other situations where the organism is airborne or spread by some other proximity based distribution model Extending the Control Loop to the Organism 6 2 1 Measuring Control Factors in the Organism Previously it was stated that a closed loop control system provided the means for accurate volume and rate delivery in the DMU However this closed loop encompasses only the drug delivery mechanism itself A more sophisticated approach would include the organism in the control loop This requires measuring one or more control factors in the animal In the present case progesterone level in milk or blood would provide the most appropriate feedback information if progesterone was being delivered Rather than simply define the delivery profile by the rate at which progesterone exits the syringe it could be defined instead by the level in the animal This would allow more efficient use of the drug payload and provide a more targeted approach which partially includes environmental factors More frequent sampling and finer control of the delivery rate leads to greater reliability and less side effects because only the minimum amount of drug need be delivered Mason Linkens Edwards and Reilly 1996 13 243 252 Progesterone levels
58. up for lost time E The gascell is off from this point onwards because the delivery programme has finished Figure 36 Downloaded Results Gascell Voltage and Current Discussion The gascell switch decision is made every 5 minutes but the data is logged only every 15 minutes This explains why there are fewer non zero current values displayed than you might expect 133 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 134 4 3 7 Pressure Results Recall that the pressure readings are for internal syringe pressure which is an approximation of piston differential pressure It is assumed that ambient pressure does not change significantly The first graph Figure 37 is a close up to allow examination of the pressure fluctuations during initial pressurisation and delivery Experimentation Pressure Temperature and Estimated Volume Delivered Close Up Day of Trial 1 0 4 1150 460 4n B v 1100 TION LO s Ir fe 140 AN M i a Vs aaa i x D Pressure Oo m o 0t ON vorm vsus NUN UT 020 e e es eec Pees 1050 wA Chee Il eng ome tes T T Yo 100 5 2 1000 amp E e d s E v Start of Drug Delivery x Insertion 5 L Start of Syringe i i 80 a Pressurisation c Y 950 B 60 g o s Temperature 40 E o gt 850 il a un Estimated Volume Delivered 800 0 726 0 714 2154 Time Since Start of Delivery Minutes When the e
59. values are not read at regular intervals but at variable times throughout the day as with robotic milking attendance Results comparing pedometer data with observed behavioural patterns are good for barn housed animals but less so for pasture based animals 3 9 4 Light Sensor A Light Dependant Resistor LDR provides an uncalibrated light intensity function Vad Light Indicator Measure 1 100nF 1 AWG 24 tinned i copper leads gt L Ni 2 79 Physical Dimensions in mm DMU Schematic RS Components Ltd 2000 Figure 19 Light Sensor Physical and Schematic Development of a Practical Drug Delivery Communication and Monitoring System The LDR R128 in Figure 19 may be useful in determining the status of the device For example if the temperature decreases markedly this might be interpreted as a severe drop in body temperature of the animal If it corresponds with a sudden increase in light level however one might conclude that the device simply protruded from the vulva for a while The light intensity reading has a resolution high enough to determine protrusion even at night The light dependant resistor is mounted in close proximity to the LED This allows for self testing A light reading can be taken with the LED on and again with it off to give confidence that the light dependant resistor is operating correctly 3 9 5 Cow Proximity Since the DMU in each cow has the abili
60. week long field trial Four cows each had a DMU loaded with 40mL of HPMC inserted into their vagina A fifth cow had an HPMC filled syringe inserted without a DMU for control purposes Figure 28 shows the DMU in vivo Periodic withdrawal of the devices allowed for a visual check of piston position to compare actual and intended delivered volume Figure 27 shows a DMU temporarily removed for that purpose Removals were kept to a bare minimum to reduce stress in the animals Experimentation Figure 28 DMU In Vivo 121 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 122 4 2 1 Animal Manipulation Dexcel technicians carried out all animal manipulations Each day the cows were herded from the paddocks into bales to facilitate insertion removal data exchange and observation The DMU s were inserted by the trained Dexcel technicians with the aid of LiveLube by hand as per Figure 29 on the first day The cows were allowed to free roam in paddocks after their daily inspection rectum TA NECS 4 J ewe GAY id STEAM DMU FQ TN i bladder 8 pelvic bone Figure 29 Location of DMU in Cow Adapted from Webster 1993 2 4 2 2 Ethics Approval Ethics approval was obtained from both the Ruakura Research Centre Ethics Committee approval number IDC 039 14 and the Ethics Committee on the Welfare of Experimental Animals at The University of Waikato in compliance with thei
61. 0 197 130 722 248 197 130 196 964 197 130 722 080 722 017 722 248 722 248 250 867 216 9274 283 0308 Top mark Comment bar at cathode FB Blows after approx 5s at 250mA or 0 1s at 300mA Function Decoupling for LDR light indication voltage measure sets time delay between power fail indication and reset signal Input capacitor for voltage regulators Bypass capacitor for TC1073 3 0V regulator Output capacitor for voltage regulators Power supply filtering on output of 3 0V regulator Transceiver power supply decoupling Crfb2 Transceiver power supply decoupling TXR Crfb1 Transceiver power supply decoupling Not fitted in OOK operation Controls AGC operation when AGCCAP pin not tied to Vdd Cpkd peak detect attack and decay setting times Cbbout base band out to comparator in capacitor User feedback indicator Protects against reverse external voltage Protects against over voltage and reverse voltage when Zener conducts Cost Ex GST NZ 0 17 0 17 0 97 0 16 0 97 1 45 0 97 0 14 0 13 0 00 0 16 0 16 2 12 2 33 0 62 Appendix D DMU Bill of Materials Designator G201 H101 H102 J101 J102 J201 L101 L201 L202 L203 PIN201 PIN202 Description Gascell 11 6mm diameter Header for optional installation of jumper 0 1 i
62. 1 This capsule has already proven that quality images can be obtained with such a small sized low power device and may prove useful imaging inside the cow The correct operation of an experimental drug release Discussion mechanism could be verified without resorting to fistulated animals or dissection Add a specialised microphone or vibration sensor with appropriate filtering Cows coming on to heat are restless and bellow during the 6 to 10 hours before they will first stand to be ridden Hlectrical resistance of genital tissues and vaginal mucus Changes in electrical properties have been observed during reproductive events Applications for vaginal mucus conductivity include pregnancy checking In addition several studies show that electrical impedance in vaginal mucus may be used to pin point proper insemination time Edwards and Levin 1974 95 416 420 Add biosensors as they become available in a compatible format glucose hormone levels pH etc These sensor ideas are mostly speculative and represent thoughts about what might be achieved in the near future A choice to pursue such a path should only be made after carefully consideration of where effort is best expended since most of these would be expensive and time consuming to implement using currently available technology 153 Conclusion 6 1 6 Conclusion Future Systems 6 1 1 Modification for Other Animals It is tempting to make ass
63. 1 to DIN bottom setting ANO JP2 to DIN bottom setting AVcc JP3 to DIN bottom setting Vref JPA Eval board Vss top AVss setting JP5 Open MCU Reset JP6 Closed for RXD1 to DIN JP7 Closed for TXD1 to DIN Open if powered from Power source select If this is left closed JP8 daughter board when operating from batteries there will Closed if powered from be around 50pA leakage through the eval board electrolytic capacitors etc JP9 Closed BYTE JP40 Open or closed Selects serial program download or run mode JP15 and JP 16 Open LED digit select JP17 to JP 24 Open LED segment select 205 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 206 DR3100 Transceiver Module Modification By default the DR3100 transceiver modules are set up for 2 4k Bits Per Second BPS operation To change them so they be can be used at the chosen bit rate of 23k BPS the following changes were made Designator on New New DR3100 Datasheet Value Tolerance Circuit Board Top View C3 15nF 10 EIS 8 o UU LED R3 200K 196 z d R6 30K 5 MGE F z fed R8 30K 5 TEE em E L2 100nH 10 CTR_O c GND CTR 1 ve LPF Figure 64 Modifications to DR3100 Transceiver Module Appendix F CD ROM Contents Appendix F CD ROM Contents N This thesis in PDF format low resolution for online viewing thesis This the
64. 10us are applied in a square wave type fashion Drugs can be loaded into cells in this way allowing them to travel to the intended zone transported and protected by the body s own mechanisms Electroporation can also be used for gene transfer Zhao 2000 133 37 43 Sonophoresis Sonophoresis or phonophersis is the use of ultrasound to enhance delivery through the relatively impermeable skin barrier and down into the soft tissue where it can be absorbed Berner and Dinh 1998 Therapeutic action occurs in the ultrasonic near field which is effective at up to a lcm depth at a frequency of 1MHz further at higher frequencies A contact medium is necessary due to the fact that there would otherwise be complete reflection from the transducer air interface at such a high frequency While not sonophoresis per se another experimental use for ultrasound is to stimulate a pellet embedded in the eye This enables periodic ocular drug delivery without having to re insert a foreign object into the eye with each dosage Background Implantable Infusion Pumps These are often embedded under the skin to deliver drugs Some devices can be refilled percutaneously Applications include cancer chemotherapy and long term pain relief Componentry generally consists of a battery electronic controller reservoir and pump inside a biocompatible housing The pump mechanism itself may be a miniaturised peristaltic pump driven by a stepper m
65. 2 2 Base Station confirms message is sent Station confirms message is sent v it input gt output Transmitted message vb04 input DMU 4 replies with 4 output Message from base station received uuu it 4 output gt Get snapshot verbose command initiated 4 output Replying on both comm port and transceiver 4 DMU 4 replies with DMU Result Set command action output Time Time Since Pressurisation 2377 minutes Time Since Power On 3154 minutes Reading Number N A snapshot Identification DMU ID 4 DMU Status Battery Voltage 3 89 V Battery Current 16 0 mA Vdd 2 95 V AVdd Vref 2 98 V Voltage Regulator Unregulated N A TC1073 Voltage regulator shutdown Voltage Regulator Shutdown yes Any Reset no Power Fail Reset no Watch Dog Reset no Time at Pressurisation 777 minutes Pressure at Pressurisation 1105 2 mbar Integrated Current at Pressurisation 26 918 mAhr Gascell Management Gascell Switch off Gascell Override inactive Gascell Override Value off Start Delivery Override active Start Delivery Override Value on Gascell Voltage 0 55 V Gascell Peak Current 0 00 mA Gascell Target Integrated Current 46 468 mAHr Gascell Actual Integrated Current 46 820 mAHr Target Volume 14 99 mL Estimated Volume Delivered 15 10 mL Gascell Target Integrated Current Rate
66. 2 way circular cross section SOCKET109 Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section SOCKET110 Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section U101 M16C microcontroller from Mitsubishi LQFP square package 100P6Q A U102 Smart Battery Monitor IC symbol for X reflected placement on schematic U103 SPST Analogue switch U104 Pressure Sensor Module with serial interface U106 MCU supervisory reset IC U107 Voltage Regulator 3 0V 80mA 1 1uA quiescent current U108 Voltage Regulator 3 0V 100mA U201 Transceiver Amplifier Sequenced Hybrid 433 93MHz 196 SOCKET Gascell SOCKET Rx Data SOCKET Tx Data SOCKET CNTRLO SOCKET CNTRL1 M30624F GLGP DS2438 MAX4628 MS5535A MAX6308 MC78FC3 OHT1 TC1073 30VCH TR3000 Foot print CON PIN1FP CON PIN1FP CON PIN1FP CON PIN1FP CON PIN1FP LQFP10 0 P 1414 0 50 MISSIN G PADS SOIC 8 MISSIN G PADS SOT23 6 MS5535 AM BUMP IN HOLE SOT23 5 SOT89 SOT23 6 SM 20L Manf Part Number M30624FG LGP DS2438 MAX4628 MS5535A M MAX6308 UKO0D3 T MC78FC30 HT1 TC1073 30VCT TR3000 Manufacturer Mitsubishi Dallas Semiconducto r Maxim Intersema Sensoric SA Maxim Motorola TelCom RF Monoli
67. 2001 In receive mode RF amplifiers RFA1 and RFA2 are controlled by a pulse generator and the two amplifiers are coupled by a surface acoustic wave delay line The RF signal is filtered by a narrow band surface acoustic wave filter before RFAI to eliminate noise outside the pass band The pulse generator turns RFA1 on and RFA2 off After being amplified by RFAI the signal travels along the delay line Now RFAI is off and RFA2 is on applying more gain RFAI and RFA2 are never on at the same time resulting in excellent feedback isolation The architecture leads itself to very quick power up and power down times A wide dynamic range log detector digital automatic gain control and a compound data slicer are built in The delay line and surface acoustic wave filter are used for both the receive and transmit functions Transceiver Hybrid Configuration 80 The TR3000 transceiver IC can be set up for various transmit power receiver Development of a Practical Drug Delivery Communication and Monitoring System sensitivity modulation type and depth minimum symbol width maximum run length data slicer behaviour automatic gain control and squelch threshold level configurations through the choice of external components In this application the transceiver is configured for on off keying modulation at a 23k Bits Per Second BPS Transceiver Polling As explained in the power mode control section the transceiver draws too much current t
68. 3 no 3 213 218 Lindsey D P E L McKee M L Hull and S M Howell 1998 A new technique for transmission of signals from implantable transducers JEEE Transactions on Biomedical Engineering 45 no 5 614 619 Linkens D A and M F Abbod 1998 Intelligent control of anaesthesia JEE Colloquium on Intelligent Methods in Healthcare and Medical Applications Digest No 1998 514 2 213 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 55 56 57 58 59 60 61 62 63 64 65 214 1 2 4 Magnusson B M K A Walters and M S Roberts 2001 Veterinary drug delivery potential for skin penetration enhancement Adv Drug Deliv Rev 50 no 3 205 227 Magtrac Inc 2001 Magtrac Available from www magtrac com Mason D G D A Linkens N D Edwards and C S Reilly 1996 Development of a portable closed loop atracurium infusion system systems methodology and safety issues nt J Clin Monit Comput 13 no 4 243 252 Mcara K 2000 Sensor Feels the Heat The Waikato Times McCreesh Z and N Evans 1994 Radio telemetry of vaginal temperature Engineering in Medicine and Biology Society 1994 Engineering Advances New Opportunities for Biomedical Engineers Proceedings of the 16th Annual International Conference of the IEEE 904 905 Miller J A 2000 Controlled Release Products for Control of Ectoparasites of Livestock In Controlled
69. 32 30 B B o 28 o 26 g A 5 24 _ w 22 o 20 i o 4 4 Lr 18 ij i 16 he 14 J f 3 D 4 12 4 t Wi 10 J 8 i i ii iv v uw 6 l 1 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Temperature From Pressure Sensor deg C Temperature From Monitor deg C When the DMU is inserted both temperature sensors increase rapidly from ambient reaching A 38 4 C They both remain within 0 5 C of that value while inserted When the DMU is removed for a piston position check both sensors rapidly cool and take 30 minutes B or so to reheat once re inserted C This is where the DMU fell out of the cow of its own accord D The temperature falls rapidly to ambient in exponential fashion Figure 32 Downloaded Results Temperature Discussion Average temperature while inside the cow is 38 4 C which is in agreement with the nominal 38 6 C reported in the literature Webster 1993 2 Even when the temperature scale is magnified greatly no diurnal oscillation is 127 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals evident in the time domain However Fourier analysis shows there is a significant 24 hour oscillation present To prepare the data for Fourier analysis 512 temperature readings from days zero though five are processed as follows Six points are artificially low due to the fact that the DMU had been remov
70. 38 6 C The syringe was set up to drip over the edge meaning that a decrease in mass represented an increase in volume delivered The scales had a resolution of 0 01g and were set to output digital mass data directly into a spreadsheet every 15 minutes The accuracy of this technique allowed an estimate of mass loss due solely to evaporation out of the syringe orifice The syringe orifice is approximately 1mm in area producing an evaporative loss of around 30uL a day With temperature fixed at 38 6 C ambient pressure varied 5 mbar which is the temperature and pressure variation expected inside the cow Other factors not simulated on the bench were the constant movement more sudden pressure changes and slight mechanical squeezing of the syringe body that would be expected in the cow These were neglected to obtain an accurate understanding of the delivery mechanism in isolation The actual volume and actual volume rate traces have been advanced with respect to the targets by 100 minutes to take account of the average syringe delay This could alternatively be done in software The actual delay would still be present but the user would not have to manually shift the response trace The 3 accuracy is good especially considering the fact that this test profile was designed to pick out the shortcomings of the syringe based system At time equals zero a step change in desired volume rate is not suited to the gradual nature of the gas
71. 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20 Figure 21 Figure 22 Figure 23 List of Figures Anatomy of the Cow Webster 1993 2 ccsccsscsccsssescescescessesssscesesseesees 15 Cow Reproductive Process Follicle Development and Degeneration Webster 1003 0 oce cete De eoe e M G 17 Application of CIDR to Bovine Estrous Cycle InterAg 2000 38 SMARTTI Intelligent Breeding Device Plade Holdings 1999 39 Systemi OVERVIEW dersucsscsutascdusasrdasacsandasssatacsdusadsutacsdosaerausdabenetarersrnadasrsrensy 51 Block Diagram of Delivery and Monitor Unit DMU 52 Gascell Physical and Schematic s icto o DEP OU eap 54 Sample Delivery Profile Defined on PC for Download into DMU 55 DMU Gascell Switch Decision eese tenente ennt 60 Gascell Monitor Physical and Schematic sss 64 Power Mode Control cce ace Ska TA RR UR niiit a 69 Transceiver Hybrid Physical and Schematic s 79 Amplifier Sequenced Hybrid Transceiver Block Diagram RF Monolithics Inc 2001 oss cute eie i M M eL uU eR LIE E UE 80 Receiver Signal Processing RF Monolithics Inc 2001 esse 85 Ar Interface Packet SUUGluIe saa ada ona a inta mamada med 88 Mitsubishi M16 Microcontroller Block Diagram Mit
72. 9 Resistor not 0603 For optional connection 0 00 present of TXR Rx Data to RxD2 192 Appendix D DMU Bill of Materials Designator R110 R111 R112 R113 R114 R115 R116 R117 R118 R119 R120 R121 R122 R123 R124 R126 R127 Description Resistor ORO SMT Format 1 tol 0 1W Resistor ORO SMT Format 196 tol 0 1W Resistor 10K SMT Format 196 tol 0 1W Resistor Resistor 10K SMT Format 196 tol 0 1W Resistor 1M SMT Format 196 tol 0 1W Resistor 1M SMT Format 196 tol 0 1W Resistor 1M SMT Format 196 tol 0 1W Resistor 1M SMT Format 196 tol 0 1W Resistor Resistor 1M SMT Format 196 tol 0 1W Resistor 3R9 SMT 0805 Format 2 tol 0 1W Resistor 100K SMT Format 196 tol 0 1W Resistor ORO SMT Format 196 tol 0 1W Resistor 10R SMT Format 196 tol 0 1W Resistor 270R SMT Format 196 tol 0 1W Resistor 270R SMT Format 196 tol 0 1W ORO ORO 10K not present 10K 1M 1M 1M 1M not present 1M 3R9 100K ORO 10R 270R 270R 0603 0603 0603 0603 0603 0603 0603 0603 0603 0603 0603 0805 0603 0603 0603 0603 0603 Manf Part Number CRGO0603 Series CRGO0603 Series CRG0603 Series CRG0603 Series CRG0603 Series CRG0603 Series CRGO0603 Series CRG0603 Series CRG0603 Seri
73. As a result more work had to be put into the low level coding of communications routines The extra time devoted to this meant reduced messaging capability and sophistication Object oriented methodology was applied to the software design process The C language by itself does not directly support object oriented programming but with discipline the software engineer can organise the code in an approximate manner Applying scope rules and access modifiers appropriately can encapsulate variables and functions effectively turning them into member variables and methods Development of a Practical Drug Delivery Communication and Monitoring System A layered approach to the software design decouples the low level driver routines for peripherals from the specific use required by the application At the lowest level there is a driver routine for each individual hardware device which when put together form the Basic Input Output System BIOS For the more complex peripherals including the transceiver and the gascell monitor a middle layer provides a range of standard objects with which any general purpose application can interact with On top is the application layer that talks only to the layer immediately below itself Only the top layer is application specific This allows for greatest portability maintainability and re use of the software source code Taking the transceiver as an example the BIOS routines for mode control pre amble
74. Drug Delivery in Farm Animals 156 loop gascell operated syringe with controlled release piston and two bolus capsules has been developed by InterAg with this in mind Rathbone et al 2001 50 277 320 6 1 3 Other Uses for Controlled Drug Release in Farm Animals Another major application for future controlled drug release products is ectoparasite control This is a relatively unexplored field due to a perception of it being a less profitable application Miller 2000 1 229 249 Even if it is not economic to produce a separate product for ectoparasite control this represents an opportunity for an adaptable controlled release product to provide both estrus control and parasite control in the one device There are already separate devices located in on the ear for estrus and ectoparasite control respectively Perhaps a multi function device for both purposes would provide a competitive advantage since a large part of the cost of deployment of these technologies is the application monitoring and removal of the devices Instead of every animal being subjected to an anti parasitic drug dose drenching the dose would be automatically dispensed if required Such multi function devices may require external control after insertion to activate after lying dormant for months at a time Control signals may be generated as a result of measuring parasite activity Simpler means may be by manual control over a wireless link whenever the farmer
75. External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals A thesis submitted in fulfillment of the requirements for the degree of Master of Philosophy at the University of Waikato by Peter Cross University of Waikato Te Whare Wananga o Waikato 2002 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Peter Cross 2002 ii Abstract Advances in cow estrus detection and synchronisation via controlled drug release will improve efficiencies in the dairy industry To this end this thesis describes the development construction and testing of an advanced drug delivery communication and monitoring system A weeklong trial with this system in five cows was conducted The original aspects of the experimental work were social grouping determined by inter cow wireless communication precise delivery of a pharmaceutical vehicle over short or extended periods of time for controlled delivery in accordance with a complex profile external wireless control override of delivery and the ability to gather an array of different data at once from the cow vagina by wireless communication The conclusion was that external control and monitoring of controlled delivery in farm animals is possible and that future systems may extend the control feedback loop outside the drug delivery device the animal itself and the farm at which the animal resides iii Chris Yardley
76. HeatWatch system from DDx Inc consists of small spread spectrum radio transmitters operating in the 900MHz band These are placed inside disposable patches and glued onto the tailheads of cows As mounting activity occurs a radio receiver collects and transmits this information to a remote computer The data can be collected and rebroadcast by repeaters to increase the range to a few hundred metres The software on the computer at the end of the data link presents accumulated data on mounting activity The manufacturers of HeatWatch claim a 95 detection rate DDx Inc 2001 Also from DDx Inc MountCount is essentially a simpler version of HeatWatch but instead of transmitting the information it displays mounting status through three lights located on the device itself One disadvantage of these systems is that being pressure activated they will falsely trigger if the cow lies against a fence or similar object Despite this HeatWatch has been used in number of estrus studies and has been found to be useful as a research tool and as an on farm estrus detector Nebel and Walker 1998 Shipka 2000 66 153 159 Walker Nebel and McGilliard 1996 79 1555 1561 Motion Sensing As explained cows coming into heat will stand to be ridden ride other cows and generally be more active One study reported that cows in heat typically walk 2 to 5 times more than when they are not in heat Nebel and Walker 1998 Another experiment conc
77. P switch 4 101 DIP switch 3 Buzzer DIP switch 2 Pressure MCLK Pressure DIN 15 Gascell Monitor Data TXR CNTRL1 Vreg Shutdown Vdd Light Indicator Measure Load Test Gascell Switch IN 32 MCU Primary Pin Allocation P8 0 TA4out U P10 1 AN1 P7 1 RXD2 SCL TAOin TB5in s eme es P1 6 D14 INT4 P2 0 A0 pu EE E pu p CEN OEC ee po I Lines E ee a Pep peg m EE SS em es x9 x e ERN RR RENI ERES ERN UMEN URGES t S pour ceo 1 Vss 23 62 14 62 25 14 14 P7 6 TA3out MCU Alternate Pin Allocation E EE we BEEN DRE SURE EN ESI NECS Eco Tg ch uneecn XL E a a P8 2 INTO P7 2 CLK2ITATouV 204 Appendix E Base Station Development DMU Settings for the M16C MCU Evaluation Board There are two of these One has been modified so that some components can be removed This makes it pretty much 100 look like part of the circuit for the DMU For example the MAX232A IC can be removed so that it does not draw any current This facilitates sleep mode power measurements where currents of only a few uA need to be measured accurately To make this board as compatible as possible remove all IC s and other components with sockets except for the crystal and configure the jumpers as follows before connecting to the Veroboard daughter board Table 13 Jumper Settings for the M16C Microcontroller Evaluation Board Jumper State Comment JP
78. PO ltd puopaz man Jo Tpays TEGLPI ang rooc m or aA Mabe WOME Gee FURNS JVC ROD ae 3 I orsi p raqumy FW org n yIW es s PNE T E plvog Aa QJIN paoifipo gy wL m 4 ont s 58 8 9c w a2 N n M m ino 7 4 E z E h B Kar Joyousueg iuam Y Ka r n ms m has m oH ES C i aoiwaavaH ui u var Bn m tiid gt SaN w M Ul M mo acma ui iid 5 L3 T oem w m rts g 1 ots D jot 8 RE p zar B IS m m om m 0B e 2 B HE T wo 95M4 x omu zaa H C bona m v w Ho E sem BE oon1 caa Ha 3 e n m a z wo sO m me T wa u seid gt Ta Ny 0ld 4 tend w ul ap 0 004 s EL t suv sDld 17 tiMd m a sit rr Hoo imd o i id HW H a J 1 TEM 5 H old oo anid mo f6Nd gt lt unig ui ui d l vM lt i md QU ni T l gemd 33 img tzi a iara n 2 OFM tzi 4 u EN m Lm gt OQ1M4 odi es uu uu iid a 2 zaid ESE ul ui B EB r 3 lt 4 Zl ul E i Sis B aiee des Id H 8M IA m O 4 C ema m T7H388 ZC E gt Rng a w k a rt 2 FO Nid w ul m on sd EE 8i m DIM CO Qo 3 gt Sing i 5i War 2 MNid T n LLL H 241 2 biMid i Dj m MM C Qo S iMd 1 1 lar 7 tzuid H n N 7 omg i n ul ZZ gt HHT novvi jona seme ai m m iwa gt IN GBd Z Baid ul m ui SIMd gt 2 find f m Syme m m ui Sinia F 2 Nd m m m oma
79. Release Veterinary Drug Delivery edited by M J Rathbone and R Gurny 1 ed Amsterdam Elsevier Science Mitchell R S L A Smith and R A Sherlock 1995 An Investigation into the Use of Machine Learning for Determining Oestrus in Cows Mitsubishi Electric Corporation M16C 62 Group Microcontroller User s Manual Rev C3 2001 Mitsubishi Electric Corp Kitaitami Works MRL Research Group On Farm Milk Testing 1997 MRL Research Group Nebel R L M B Jobst M B G Dransfield S M Pansolfi and T L Bailey 1997 Use of a radio frequency data communication system HeatWatch to describe behavioural estrus in dairy cattle J Dairy Sci 80 no Suppl 1 179 Nebel R L and W L Walker 1998 Electronic Heat Detection Devices Available from http www heatwatch com virginiatech cfm References 66 67 68 69 70 71 T2 73 74 75 76 Nebel R L W L Walker M L McGilliard C H Allen and G S Heckman 1994 Timing of artificial insemination of dairy cows Fixed time once daily versus morning and afternoon J Dairy Sci 77 3185 Nielen M Y H Schukken A Brand S Haring and R T Ferwerda Van Zonneveld 1995 Comparison of analysis techniques for on line detection of clinical mastitis Journal of Dairy Science 78 no 5 1050 1061 Ogle C 2000 Design Development and Optimisation of Veterinary Intravaginal Controlled Release Drug Delivery Systems PhD
80. Structure Chip scale devices that can sense and or control the physical environment Random Access Memory can be re written in individual bits and bytes at high speed but does not retain data when power is removed The M16C microcontroller used in this thesis has 20K of RAM xvii External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals RFID Theratron xviii Radio Frequency IDentification A system of identification whereby the item animal or person to be identified carries a transponder A reader interrogates the transponder with a coded radio signal and the transponder replies with a unique identification number A commercially available purpose built syringe incorporating a gascell The gascell produces a constant supply of gas behind the syringe piston which provides controlled delivery of the pharmaceutical vehicle Originally intended for human portable drug infusion it was modified to form the basis of the drug delivery mechanism used in this thesis Introduction 1 1 1 Introduction Overview A review of current and emerging estrus detection and drug delivery technologies was conducted in relation to the local dairy industry This established a need for an investigative programme into external control and monitoring of controlled drug release in farm animals This review provided the basis for a feasibility and design study for the device used for in vivo experimentation Development
81. V regulator with low 8 83 voltage differential Transceiver sends and 62 25 receives data via RF under control of the MCU Appendix D DMU Bill of Materials Designator Description PCB pad for direct mounting of semi rigid Co ax Crystal 32 768kHz 8 2pF load capacitance 20ppm Crystal 10 000 MHz 6 8pF load capacitance RSX 3 30ppm Zener Diode 6 2V Part Type 32 768kH z 10 000MH RSX 3 2 Manf Part Number Foot print CO AX MOUNT DS10 0N DHEO8RS X 3 10 0 MHz BXZ84C6V 2 Supplier Manufacturer Supplier Order Code Micro Crystal Micro DS10 0N Switzerland Crystal Switzerland Rakon Rakon DHEOB8RSX 3 10 0 MHz Philips Farnell 251 604 Semiconducto rs Top mark Comment Do not exceed 280C Do not heat for more than 5sec Do not solder can Price is approximate 10 00M 0400N1 reflow soldering Z60 or ZAP or Z4T Total Component Cost Ex GST Able to withstand Cost Ex GST NZ Function Special PCB pad to mount transceiver transmission line semi rigid co ax directly 32kHz MCU sub clock 5 00 10 000MHz main clock 15 00 Helps protect voltage regulators from input over voltage 264 36 197 Appendix E Base Station Development DMU Appendix E Base Station Development DMU 199 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals r t ex IPS peus preog UA 29TIN PIP
82. a Practical Drug Delivery Communication and Monitoring System The accuracy of delivery relies heavily on obtaining accurate integrated current measurements This is complicated by the fact that gascell current varies somewhat with a constant resistive load due to bubbles that form on the gascell electrodes To combat this the instantaneous current is measured and accumulated 36 times a second with the gascell monitor IC 10 bit analogue to digital converter This occurs even when the microcontroller is asleep consuming only 30uA of battery current on average The voltage resolution of the integrating current analogue to digital converter is 244 1uV which translates into a current of 62uA Sampling and accumulating at 36Hz this would represent a theoretical estimated volume resolution of less than InL Nano litre accuracy is unattainable in practice due to analogue to digital converter error thermal and semiconductor noise residual error on the sense resistor and analogue to digital conversion errors Variability of piston friction and other physical attributes of the syringe swamp any attempt to make use of such a fine resolution The bench test results proved that the short term resolution of the overall system is on the order of 10uL A graph of bench test delivery is presented on page 114 Threshold When the gascell supplies very little current the error after offset correction due to the current analogue to digital converter is still
83. a that can be used The data slicer in the receiver is preceded by a low pass filter to limit the effect of noise spikes and coupled via a capacitor to provide adequate dynamic range This imposes a base band limited constraint on the data stream The minimum bit width must be wide enough to pass through the low pass filter setting the maximum bit rate However the data value must also change frequently so that the high pass filter effect of the coupling capacitor does not let the data slicer settle to an indeterminate state representing a value halfway between a zero and a one If the base band filter requirements and bit errors are ignored then the system would become practically useless Fortunately the information bits transmitted over a digital wireless link can be protected by introducing redundant bits and rearranging them into a more palatable data stream This collective process is known as channel coding In summary the aims of channel encoding are to Provide error checking possibly in combination with error correction Development of a Practical Drug Delivery Communication and Monitoring System Provide control codes that can be differentiated from the data Achieve DC balance Provide a minimum run length Be easy to implement Be efficient compared to other encoding methods Channel encoding should be preceded by source encoding Source encoding is the process of using a digital signal to represent an o
84. administer the technology to more and more animals Artificial insemination technicians are already in short supply Some travel to different parts of the country thereby lengthening the season over which they apply their skills Others are required to work double shifts at peak times Automated controlled drug release products will help artificial breeding technology become less labour intensive 2 3 4 Attaining Increased Per Cow Efficiencies The trend is toward larger herds using less skilled labour These production units do not the same level of contact between cows and people The contact time between cow and human being can be as little as five seconds a day That is not enough time to adequately monitor the general health of the animal let alone work out if a cow is in heat This is especially true of young cows in which estrus generally lasts for around seven hours Mcara 2000 21 11 2000 Evening Edition This shows a need for a combination of advanced automated estrus monitoring and control programmes and products ll External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 12 2 3 5 Improving Reproduction Performance Managing the estrous cycle more effectively can enable improved milk production with more calves on the ground earlier lower breeding costs fewer reproductive culls and fewer inductions All this while recognising that the welfare of animals and the environmental sustainability of a
85. al test initiated by the base station A DMU cannot receive a transmission from itself when it is busy transmitting This is why there are no transmissions recorded from DMU 3 The following graphs show cow proximity data downloaded from DMU s 1 and 2 137 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Cow Proximity Recorded By DMU 1 Day of Trial 1 0 1 2 3 4 5 6 7 5 4 HEIDE v g o g o g 3 LE o 2 E 3 z o 2 ewm z o 5 1 i ii ii iv v 0 e d i d 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Figure 40 Downloaded Results Cow Proximity Recorded By DMU 1 Cow Proximity Recorded By DMU 2 Day of Trial 1 0 1 2 3 4 5 6 7 Cow ID Number Detected i ii iii iv v 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Figure 41 Downloaded Results Cow Proximity Recorded By DMU 2 Discussion 13s Many successful transmissions prior to insertion were recorded After that time Experimentation only two inter cow transmissions were seen by DMU 3 With the transceiver on DMU 2 only having a limited range DMU 4 not responding after insertion and DMU 2 unable to respond after day 2 only inter cow transmissions from DMU 1 were recorded by DMU 3 4 3 9 Motion Results The vertical dashed grid lines are equally spaced at 0815 each day They are approximately a
86. ally Controlled Drug Delivery Edited by Berner B and S M Dinh CRC Press Bluetooth SIG Inc 2001 Bluetooth V1 1 Core Specifications Bluetooth SIG Inc Bobowiec R T Studzinski and A Babiarz 1990 Thermoregulatory effects and electrical conductivity in vagina of cow during oestrous cycle Arch Exp Veterinarmed 44 no 4 573 579 Bou Matic 2001 CowTrakker PID Estrus Detection Tag Available from 209 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 210 10 11 12 13 14 15 16 17 18 19 20 http www bou matic com spindb query listall2 bouview 6157 Bunt C R and P S Cross Herd Control and or Monitoring Procedures NZ428552 2000 Bunt C R M J Rathbone C Ogle and S Morgan 2001 An Electronically Modulated Bovine Intravaginal Device for the Delivery of Progesterone Proceedings of the International Symposium on Controlled Release of Bioactive Materials 28 Burke C R 1999 Managing Your Herd To Meet Reproductive Targets 999 Ruakura Farmer s Conference The Business of Farming Claycomb R 2001 Personal communication on milking robot as round the loop sensor actuator Cleland W P 2001 Opportunities and obstacles in veterinary dental drug delivery Adv Drug Deliv Rev 50 no 3 261 275 DDx Inc 2001 HeatWatch Website Available from http www heatwatch com dairybene cfm de Mol R M 2001 Automa
87. and Inter PCB Batteries Board Connector 1 5V Button Cell x3 Electronics Package from Microcontroller End with Digital Board Removed Figure 21 DMU Electronics Construction 109 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals c c ro TIT povisredt Eh SA Figure 22 DMU Electronics Package Fitted to Modified Theratron Syringe Sectioned for Display and Shown Actual Size DMU Antenna gt Piston y CIDR Retrieval Cord a CIDR Retention Device b 2 r Y Fi A P Electronics Package o Y x v R Figure 23 Complete DMU Ready for Deployment 110 Development of a Practical Drug Delivery Communication and Monitoring System 3 13 Bench Testing 3 13 1 Communication and Measurement Testing Several results from bench testing are presented The first is an interaction between the user and a DMU showing a command typed into the laptop by the user Once the command has been parsed and found to be valid the base station sends it informing the user of the transmission The DMU replies with confirmation that it has received the command and then carries it out 111 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Table 6 User Requesting a Snapshot From DMU 4 Over Wireless Link bod a User types commana 1 1 User types command types command Pese Station confirme message is sent
88. appropriate If the message has a to ID of 9 it is a broadcast message and all DMU s respond A broadcast message means the DMU s respond after a delay determined by their ID number This is to ensure that they do not all try and reply at once The following commands are available Snapshot Command The snapshot command allows for a snapshot of the state of the device and the current readings from the environmental sensors It is the same as a result set taken during data logging but is transmitted instead of being stored An example of the user requesting a snapshot is given in Table 6 on page 112 Download Command The download command transfers the entire data log to the base station Setup Information Command A setup command responds with all calibration figures threshold settings safety limits and time constants stored in the device during program download Table 9 in Appendix A page 175 shows the user requesting setup information Wake Up Command The wake up command keeps the DMU awake for 10 seconds instead of the usual 2 5 seconds This is useful for bench testing the digital board without a transceiver board fitted 93 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 8 94 D Command When this command is received it updates the cow proximity array for the current result set The relevant bit is set according to the from address of the received
89. ased in Hamilton New Zealand having forged strong bonds with The University of Waikato Government funded research agencies and overseas collaborators from a variety of backgrounds Their relationships with universities focus on providing a source of interesting and worthwhile projects to which postgraduate students can apply themselves This results in projects which are a mixture of pure research and commercially oriented content that prove mutually beneficial The companies provide substantial 3 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals materials resources and expert guidance In return they receive the benefits of the student s effort in exploring relevant new areas of research backed by the rigour of academic supervision from the University The University itself gains a reputation for producing immediately employable graduates who have experience with real world problems and an appreciation of the underlying theory This thesis is an example of this ongoing collaborative effort Background 2 Background 2 1 New Zealand Dairy Industry Structure Justification for research into the general problem of drug delivery in dairy cows for estrous control requires a brief understanding of the social and economic factors operating in the industry on a local and global scale For this purpose the current state of the New Zealand dairy industry is briefly described 2 1 1 Owne
90. asis InterAg has experimented with an intravaginal constant rate drug delivery device with the aim of facilitating continuous delivery over a period of up to two weeks This experimental system is based on a commercially available syringe the piston of which is moved by production of hydrogen gas from a gascell The value of a resistor determines the fixed rate of delivery No battery is required since the act of drawing current from this special button cell produces the gas This is a type of electrolytic gas powered pump described in Chapter 2 There it was stated that electrolytic gas powered pumps have proven to be easy to construct easy to control are accurate in delivery and can cope with a large range of required flow rates Such a system would be greatly enhanced by the addition of some intelligence memory and the ability to communicate with the outside world This proposed system of drug delivery would then be capable of providing a more complex drug delivery profile With two way communication recorded events such as temperature readings could be transferred to the outside world to allow for the condition monitoring of the animal More importantly for the subject of this 45 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 2 46 thesis external control of the device could be obtained Once this basic form of communication has been established an automated drug delivery and resul
91. atic symbol by itself The readings are requested and obtained by the microcontroller over a proprietary four wire synchronous serial interface A temperature sensor is provided in the same module for temperature compensation The MS5535 was chosen for its low power operation small size large pressure range adequate accuracy and high resolution output It was designed for diving computer watches and tyre pressure gauges The 0 to 2 bar version is used in altimeters and can register a change in atmospheric pressure when moved through a vertical height of only one metre The MS5535 has a pressure resolution of 1 2mbar and a temperature resolution of 0 015 C 9 i H MS55354 amp 9x9x4 3 mm Physical DMU Schematic Figure 17 Pressure Temperature Sensor Physical and Schematic The piezoresistive pressure sensor is a Micro Electro Mechanical Systems MEMS device MEMS are chip level devices that can either sense or control the physical environment They are commonly made out of silicon with moving 99 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals parts constructed by micromachining processes MEMS use a number of manufacturing steps derived from integrated circuit fabrication techniques The on chip 15 bit analogue to digital converter is used for both pressure and temperature readings The module is calibrated at two temperatures and pressures at the factory f
92. ation function is quite rudimentary Every half hour each cow transmits it own ID number If any other cows are in range they will record the fact that a reception from that particular cow has been received At the end of the data logging period a flag will have been set for all cow IDs that have been heard from in that period The cow proximity flags are stored as part of the current result set before being cleared for the next data logging period 3 10 Software Design Methodology 104 The source code for the base station and DMU is common being written almost entirely in the C programming language Assembly was used sparingly where C was not suitable To generate the executable for the base station instead of the DMU conditional compilation is used This aids in software development and maintenance due to the fact that there are many common software routines The use of code libraries encouraged structured software development and reduced compilation time considerably The M21 Real Time Operating System RTOS was obtained from M2 Technology in Auckland New Zealand The use of this product would have made programming the DMU and base station much easier since it has networking capability built in Unfortunately the conditions of use specified that it could only be applied to a non commercial project Once it became clear that the DMU would be further developed beyond the scope of this thesis the M21 RTOS was removed from the project
93. attempted because the rudimentary effort proved surprisingly accurate The estimated and actual volumes delivered in Figure 25 on page 114 are within 3 even though the profile is a demanding one Accuracy is more than adequate even though the profile involves lengthy periods of zero rate delivery and delivery at different rates Syringe Calibration Each syringe piston set has a different leak rate and piston friction coefficient due to dimensional manufacturing tolerances and wear The variance of these factors can be reduced to an acceptable level by the use of two methods Piston lubrication with silicone oil Careful attention to consistent application of silicone oil to the piston results in reducing the variance in piston friction Calibration constants for piston friction and leak rate The leak rate and moving piston friction coefficients are combined with the integrated current to volume delivered ratio This is explained in the section on leak compensation above The variance in static piston friction is partly eliminated by specifying an individual minimum pressure leak compensation limit for each syringe piston set The calibration process is straight forward The sample run is performed with the desired syringe piston set and a very low pressure leak compensation threshold To determine the leak rate factor the data is scaled in a spreadsheet until the actual delivered volume overlays the target delivered volume The
94. bility to dispense more than one drug The multi drug variable rate serially delivered platform would support complex protocols and be able to dispense bolus shots in parallel with the main controlled release payload Future systems may extend the control feedback loop outside the drug delivery device the animal itself and perhaps even the farm at which the animal resides These herd based systems would provide individual and collective data to a remote computer Integration with herd management software would close the control feedback loop and allow interaction with the devices over the Internet while they are in the cows Conclusion 6 5 4 Summary This thesis demonstrated the feasibly of external control and monitoring of drug delivery in farm animals The development programme produced a practical device which met or exceeded all initial design criteria in real world conditions generating four significant and original results The Delivery and Monitor Unit is a valuable tool for further research into drug delivery in farm animals 171 Appendix A Additional Experimental Results Appendix A Additional Experimental Results 173 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Table 8 Description of Observations Results and Manipulations for DMU 3 and Control Syringe Date Time Day of Time Time Since DMU 3 Actual Volume Estimated Control Gasc
95. break off strip price is for 1 socket of the 32 way break off strip price is for 1 socket of the 32 way break off strip price is for 1 socket of the 32 way break off strip Function Rth1 heps set squelch and threshold for the standard data slicer Rpr PRATE resistor sets interval between the falling and rising edge of an ON pulse to the first RF amplifier Rpw PWIDTH resistor width of the ON pulse to the first RF amplifier Rref helps set squelch Not required Not required Would stop this line floating in sleep and transmit modes but the MCU outputs a low to do that so not required Not required Senses mounting activity of cow Resets MCU Vdd PCB interconnect GND PCB interconnect Battery PCB interconnect Battery PCB interconnect Gascell PCB interconnect Cost Ex GST NZ 0 11 0 05 0 05 0 05 0 00 0 00 0 00 0 00 12 89 2 52 0 25 0 25 0 25 0 25 0 25 195 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Designator Description SOCKET106 Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section SOCKET107 Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section SOCKET108 Connector 1 socket of wire wrap SIL socket pin strip single row straight 3
96. cacos r S0LL3420 1qngs 94 sasos 1X1594 sasos IM uod IS FIMELA SOUS Love od raroa X Samsong oue 9d sarod qnae 9d zatoa i suo d laos PE smile PO asos Ess canes 7X Torn loe Tor prog auba TOTT PPA troller Icrocon tal Board M igi Figure 50 DMU Schematic D 179 q s4swndns pur aog md s Jo Pus IFZ tI 3URL TO07 AjV sc SRA T orto anqums tV cns Dv 91u mE cI 89g m Syn Muosiawadngy pun samog pavnog jwousiq Awa WL AOSIAMIANY AAPJOMUOIOAIUY m as SOCOXVIN orld Dj NI su INDIE Hx us as LL t L pan ant d neo Wr d CALEN LOI DINBAY A0 E wesaad you Tera AMP quip AMOT MOT Aust ER D io HOA Ot 2012 1 TOLLT3ODOS GND LINDOS Vutosc AMOT MoT TA9 ru TOrz dd vulszrt TOLLANDOS PPA LTiDOS told POLLANDOS sorn owga LINDOS 2 a Figure 51 DMU Schematic Digital Board Power and Supervisory External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 180 Appendix B DMU Schematics l By su i UOYUDITPUT SHIDIS AHS 051 PDOT jj2250 34njDioduia y PUD FAST OAT IPTG qn aana saso Deyna ao waja WiO CHASE SE Figure 52 DMU Schematic Digital Board Peripherals 181 External Control and Monitoring
97. ce is much lower This is achieved by measuring its value accurately during DMU assembly and storing this with the other calibration constants C110 and R122 form a low pass filter to help capture the effect of any particularly fast current transients 63 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals The DS2438 gascell monitor IC U102 in the same figure was originally developed for fuel gauging battery packs in cell phones It has the following features Unique serial number Integrated current accumulator Keeps a Direct to digital temperature running total of all current Md going into and out of the Realtime clock battery Two general purpose 40 bytes of non volatile analogue to digital voltage memory EEPROM inputs All these functions are provided in an 8 pin small outline IC format Communication with the device is via a bi directional one wire serial bus the DQ line Only the integrated current accumulator is used for integrated gascell current measurements The other functions of the gascell monitor IC will be discussed as they become relevant To gascell and gascell load control switch B Mosi GND 1 8 DQ Vsen 2 7 NC 421 Vsen 3 6 NC Vao 4 5 Ov SOIC 8 Pin 0 150 inch body width Physical DMU Schematic Figure 10 Gascell Monitor Physical and Schematic 64 Development of
98. cers more than a trillion US dollars per year Reproduction problems are the number one reason for culling on many dairy operations In fact more cows are culled for 19 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 20 reproductive reasons than for milk production reasons Bou Matic 2001 Reproductive experts calculate that it costs approximately three dollars per day for each cow in the herd that is open over 100 days In the USA once milking losses replacement cost calf losses veterinary services medication and added breeding cost is taken into account the total annual cost per cow of estrous mismanagement there is US 89 Bou Matic 2001 Figures for New Zealand farms are NZ 125 in lost production per missed estrus and a further NZ 110 per induction if subsequently required Dexcel Ltd 2001c Effective reproductive management means calving at an optimal interval This synchronises lactation to peak levels of efficiency When using artificial insemination this relies on achieving effective detection of estrus Observable Signs of Estrus The best behavioural indicator of estrus is standing to be mounted by another cow so called standing heat This mounting activity occurs on average 11 times per heat period with each mount lasting 3 seconds Walker Nebel and McGilliard 1996 79 1555 1561 A summary of mounting behaviour and other effects is as follows Dexcel Ltd 2001b
99. construction and testing of this practical drug delivery communication and monitoring system was described This Delivery and Monitor Unit DMU was designed to evaluate intelligent variable rate controlled release in the cow vagina along with a data logging system for diagnostic and environmental information Included is a two way radio link for external control and exchange of information with the base station and other cows while in vivo Methods materials and results from an experiment using the DMU in a weeklong trial involving five cows were presented The original aspects of the experimental work were social grouping determined by inter cow wireless communication precise delivery of a pharmaceutical vehicle over short or extended periods of time for controlled delivery in accordance with a complex profile three percent accuracy with safety limit override external wireless control override of delivery and the ability to gather an array of different data at once from the cow vagina by wireless communication An evaluation of the performance of the system identified it as meeting or External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 1 2 exceeding all of the initial design criteria Discussion then followed on how this and future systems might be developed Possible herd based and farm based systems were considered The conclusion was that external control and monitoring of controlled deli
100. contain loaded silicone but are intra vaginally located devices The CIDR is a T shaped device with a nylon core while the PRID is similar in concept but has a stainless steel core in the shape of a spiral Silicone devices have the disadvantage of having to be monitored while in situ and have to be removed In the case of the intravaginal devices this is a reasonably simple task Intravaginal devices offer unique advantages and disadvantages and are an under exploited delivery route in general Rathbone et Background al 2000 1 173 200 The silicone subcutaneous implant technology often located in the ear requires a minor surgical procedure which leaves the animal open to possible infection Implants can migrate under the skin and rogue implants are a possible source of rejection at the meat processing works when the cow is culled One alternative is a subcutaneous injection fluid enough to be administered by syringe but which hardens into an implant shortly afterwards The implant dissolves in a biodegradable fashion over time SABRE MATE is one such product Current best practise to provide synchronisation is achieved by a seven or eight day delivery of progesterone from a slow releasing in ear or intravaginal device followed by an injection of estradiol or prostaglandin to control follicular development While it is impossible to describe here how all of the listed products work two that follow this type of prot
101. controller and other IC s in the circuit is required Figure 11 depicts the various power management modes the circuit can be in The modes are the shaded boxes and give the amount of current consumed for the whole circuit while in that mode as well as the percentage of time spent there The microcontroller port pins configure peripherals for lowest power consumption at all times The mode transition causes are represented by arrows This is a simplified view of how mode switching occurs It is impossible to accurately show on one page what happens for each individual state and every possible combination of interrupt event Table 1 Description of Microcontroller Modes Mode Description Microcontroller core stopped timers being clocked by 32kHz sub clock for real time clock and motion activity counters gascell current sleep integration 36 times a second transceiver in sleep mode 10MHz oscillator is off Doze As for sleep mode but transceiver is in receive mode As for doze mode but microcontroller core is operating from 32kHz Low Speed sub clock Microcontroller core is clocked from 10MHz oscillator Transceiver is Fully Awake in receive or transmit mode Development of a Practical Drug Delivery Communication and Monitoring System Sleep mode 60uA 99 2 No valid data detected Receiver polling wake up Low speed mode 775 uA 4 5mA when Receiving 0 2 Valid transceiver receive
102. d robot milking system Fullwood Ltd 2001b Fullwood Ltd 2001a The Merlin robot milker and others like it account for 3096 of new milking system installations in the Netherlands Each cow wears an ID tag and pedometer When the cow feels like being milked it approaches the robot which dispenses the calculated amount and type of feed based on its accumulating database of statistical information about that particular cow and how it is performing relative to the rest of the herd On line sensing of the milk is limited to conductivity flow and volume per quarter at this stage If the robot senses the cow is sick e g significant drop in activity level due to lameness she is automatically directed by separation gates to an isolated pen for examination by the farmer when he arrives Such machine cow feedback is limited at this stage The software operates the robot and reports on feeding milk quality milk yield and estrus detection through activity level At the moment the farmer makes most of the decisions but as on line sensing and drug delivery technology improve more and more of the decision making will happen under autonomous control rather than human intervention Figure 46 depicts a situation where a such a milking robot and herd management software might communicate with a drug delivery device inserted inside the cow Claycomb 2001 The insert and base station could be an extension of the DMU and base station developed for th
103. d applied work needs to be performed before these demands are met It is the aim of this thesis to provide fundamental information on the capabilities Introduction 1 3 of controlling electronic drug delivery systems thereby building on knowledge in this growing area of research Specific objectives are to l Examine the feasibility of electronically controlled intravaginal delivery in farm animals including the ability to externally control the delivery system Report on the methods for doing so problems that might arise and the solutions to those problems Build a system capable of controlled release external control and monitoring of intravaginal delivery in farm animals The in vivo device should be able to precisely deliver pharmaceutical vehicles over short or extended periods of time according to a complex profile rather than simply on off or constant rate Assess and report on the performance usefulness capabilities possible improvements and future applications of the system as built Suggest improvements for development of future systems Consider the implementation of a herd based system providing individual and collective data to a remote computer Relationship With Industry InterAg is a research company specialising in intravaginal controlled drug release for estrous control in cattle Sensortec a sister company develops in line sensors for complex biological fluids Both companies are b
104. d transducer It is conceivable that an electronic circuit could be added to any form of drug delivery mechanism However this review is more concerned with electronic actuation because it is likely that this approach would lead to smaller simpler more reliable devices Stimulated Polymer Elution of the drug usually follows the standard square root of time law The existing knowledge and manufacturing experience using this technology can be extended to more complex delivery profiles by using electric or magnetic fields exposure to ultrasound light or enzymes and changes in pH or temperature Microchip Array A silicon based integrated circuit provides controlled release of single or multiple chemical substances on demand Santini jr Cima and Langer 1998 Santini et al 2000 32 377 379 Santini Langer and Cima 1999 397 335 338 Micro reservoirs filled with chemicals in solid liquid or gel form are covered by a cap consisting of a thin layer of gold film that also acts as an anode An electrical voltage applied between the anode and a cathode causes a reaction that dissolves the cap releasing the chemical Controlled and pulsatile release of chemical substances with this device has been proven in the laboratory and a company called MicroCHIPS has been formed to commercially exploit the new technology Potential advantages of this approach include small size low power consumption Background absence of moving parts and t
105. data real time clock tick or reset Fully awake mode 17 mA 0 1 Decide on action based on wake up event and other factors Receive data while valid data detected Take Perform other housekeeping measurements maintain real time clock current A D calibration gascell switch decision safety check Process and TC1073 voltage regulator enable Carry out store results transmit ID etc received command Doze mode 950uA 0 5 Receive data inactivity time out Figure 11 Power Mode Control 69 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 70 Current Consumption in Sleep Mode The transceiver draws too much current to be kept waiting for data in receive mode Instead it is placed in receive mode for only 8ms out of every 600ms It is asleep for the remaining part of the receive polling cycle The average current due purely to receiver polling is duty cycle x receive current microcontroller low speed current 8ms 600ms x 3 7mA 0 78mA 600A The total sleep current for the entire circuit when the transceiver is off is 60nA With receive polling this increases total sleep mode circuit current consumption to 120A Ideally pulse width modulation outputs would have been used to control receive mode polling This would have eliminated the need for the microcontroller to wake up once every second to initiate the on part of t
106. derately towards 1 and a change from one to zero drags it moderately down towards 1 Detector Low Pass Filter Data Data Out Slicer C Threshold Figure 14 Receiver Signal Processing RF Monolithics Inc 2001 The threshold provides a squelching action to keep the receiver quiet unless a signal of reasonable significance is detected Raising the threshold reduces the range but limits the degree to which the receiver unnecessarily wakes the microcontroller up out of low power sleep mode Before any data is sent a data slicer training pattern of 11001100110011001100 is transmitted at the start of each packet to set the slicer threshold to the halfway point There are various coding approaches to assuring minimum run length and DC balance during the remaining data transmission The trade off on using a more time efficient coding scheme is the amount of processing required at each 85 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 86 end Howell 1998 33 60 73 Eight bit to ten bit encoding was selected as the most appropriate coding scheme to implement but some of the other more common coding schemes are also discussed below Manchester Encoding Manchester encoding is the simplest scheme requiring a minimum of processing complexity Manchester encoding sends each bit twice the second copy being inverted This has a very good maximum run length of only two bits but halves
107. ductivity starts at around 7 6mS and approaches 12 0mS before dropping again to 10 0 mS at ovulation This conductivity fluctuation corresponds with the average temperate change but lags by around 12 hours and also has a reasonable correlation with milk progesterone Variability is large between cows however indicating that a single measurement of electrical resistance is unreliable in distinguishing physiological states Gartland Schiavo Hall Foote and Scott 1976 59 982 985 Electrical resistance of vaginal mucus on the cervix can be when the correct procedure is used However the procedure is time consuming and involves some skill It must be positioned expertly in order to obtain reliable results Wehner et al 1996 Staff Paper SP 96 004 a Hormone Levels The role of progesterone was described in the section explaining the estrous cycle Progesterone in milk or blood will provide excellent data on estrus status because it is a controlling variable rather than an effect Progesterone levels in milk and blood are often used as a gold standard in accurately assessing the state of the estrous cycle and particularly for identifying cows with silent heat Currently the milk or blood sample has to be sent to a laboratory for analysis but low cost on line milk sensing products are being developed and are expected to 27 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 28 become avai
108. e complex workings of the naturally occurring estrous cycle functions controlled drug delivery methods can be designed to take advantage of this new knowledge to increase fertility shorten the time to calve and improve quality safety and efficiency of the drug regimen Background 2 3 6 The Role of Estrous Control in Attaining Increased Dairy Efficiency The central theme is that New Zealand must strive for increased efficiencies if it is to continue to be successful and that better estrous control has an important part to play in many applications aimed at increasing efficiency at all levels Controlled drug release has traditionally been central to estrus control programmes and as these programmes develop they become more complex Drug delivery technology must keep pace being developed in conjunction with these programmes to provide new and appropriate solutions in a timely manner 2 3 7 Market Demand for Improved Estrus Products New Zealand farmers have recognised this need for improved estrous oriented products They operate with low capital investment in equipment meaning that any expenditure on new systems is rarely considered Accordingly adoption rate of new technology is low due to the expense involved even when it is perceived that it would be beneficial Due to the marked decline in profitability during the 1990 s farmers have been slow to maintain and upgrade even the essential items of their capital equipm
109. e gascell is very small this allows a large proportion of the device to be available as a drug reservoir In fact the gascell does not produce hydrogen gas in a normal open air environment it consumes oxygen instead If activated in an enclosed space however it starts producing hydrogen gas once all of the enclosed oxygen is depleted For this reason the amount of dead space in the syringe must be kept to a minimum If there is too much dead space a significant negative pressure is created during oxygen depletion This means a lengthy delay for subsequent hydrogen gas production and hence start of delivery 53 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 54 PIN Gascell PIN205 Gascell G201 GASCELL Gascell PIN Gascell PIN206 Physical 11 6mm Diameter 4 5mm height DMU Schematic Simatec AG 2001 Figure 7 Gascell Physical and Schematic The gascell in the DMU is represented in the schematic as G201 in Figure 7 Complete schematics are provided as Appendix B starting on page 177 There are different sized gascells available Large ones are used in providing the piston movement in grease dispensers Grease is supplied via a 120mL syringe attached to a standard grease nipple which obviates the need for weekly servicing on rotating machines This can be very cost effective on inaccessible plant equipment such as pumps on marine oil rigs The
110. e most accurate amount if inserted before pressurisation has commenced This means the start of actual drug delivery for each syringe will be slightly different due to the fact that they will take slightly different times to pressurise By inserting and then relying purely on a set time delay the syringes keep absolutely synchronised in time but there will be some error in the amount delivered at the start of the delivery schedule A more complicated approach eliminates both of the previous disadvantages To achieve this the syringes are keep on the bench initially The start decision override command is made active over the wireless link until the syringe is pressurised and observed to start delivering This keeps the gascell turned on but prevents the profile from starting When pressurisation occurs the gascell override command is initiated to turn the gascell off The syringes are now ready to start delivery at short notice When all syringes have been accounted for by this technique the DMU s are inserted and both overrides are inhibited This means they are all synchronised in time and have delivered exactly zero target and actual volume at the start of the delivery programme The disadvantage is the extra effort required Using this technique many Development of a Practical Drug Delivery Communication and Monitoring System DMU s could be prepared and stored for up to a week and be ready for this form of accurate depl
111. ead random errors that fall within the error correcting capability of the forward error correcting code The simplest way to achieve this is to read the data stream into a grid row by row and send it out column by column Coding for Minimum Run Length and DC Balance The data slicer in the receiver is coupled via a capacitor introducing a high pass filter effect This allows the intended logic value to be predicted by looking at the short run average value of the received signal with greatest weighting on the most recently received bit period Development of a Practical Drug Delivery Communication and Monitoring System This imposes two restraints on the data pattern It must be guaranteed to have frequent state transitions and it must be DC balanced If long strings of zeros or long strings of ones were sent in a row the high pass filter output would settle down to an indeterminate value between the voltages representing zero and one corrupting the data stream Frequent bit transitions are also necessary for the receiver to maintain synchronisation with the transmitter The maximum run length in the DMU has been set to five to provide a compromise between high bit rate and greater sensitivity Figure 14 shows the receiver signal processing in a simplified form Any data slicer input voltage below OV is converted to a zero at the Data Out port and any voltage above OV is converted to a one A change from zero to one lifts it mo
112. ecause New Zealand has already totally abandoned subsidy of agricultural production the country stands to gain when other countries remove their subsidies Prices for New Zealand dairy exports can be expected to rise very substantially once this happens however this could be a drawn out process Another positive trend is the international market for processed dairy products which is expanding in breadth and depth In particular yoghurt and flavoured milk have achieved very high growth on a global scale and the cheese market in Japan is small but developing In the past the Asian markets have been External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals considered poor but there is a new dairy consuming generation reaching maturity in Japan The Japanese Government has raised public awareness of the goodness of milk and subsidised it in schools for many years now The same strategy is now being trialled in China This will create a new dairy consuming generation in the world s most populous nation Already the predicted growth of 2096 per annum is being revised upwards closer to 50 While New Zealand has traditionally focussed on raw milk products this gives it a huge potential for gain if it can capitalise on these value added markets and invent new ones along the way 2 2 2 Global Trends Affecting the Industries Future global decreases in price due to globalisation biotechnology and increased efficiency in
113. ed on occasion to check the piston position These points are replaced by the value of measured body temperature immediately preceding them The points are smoothed with a median filter The mean temperature is subtracted from each point to remove any DC level This is necessary because the DC smudge would otherwise swamp the low frequency component of interest Temperature in Frequency Domain Relative Magnitude 0 50 100 150 200 250 50 Frequency Cycles per Day x 0 188 Figure 33 Downloaded Results Temperature in Frequency Domain 128 Experimentation The frequency resolution is reasonably coarse The 255th discrete value on the right of Figure 33 represents half the sampling frequency or 48 cycles a day Therefore the frequency resolution is 48 255 0 188 cycles per day The peak value occurs at 5 on the raw FFT frequency axis This corresponds to 5 x 0 188 cycles per day or a period of 1 06 days This is the diurnal variation in ambient temperature and animal metabolism Fourier analysis of the measurements taken from the other temperature sensor resulted in a very similar frequency spectrum 4 3 4 Volume Delivered Results The variable rate DMU s stopped working after day 3 however there was enough data gathered from them to demonstrate variable rate delivery in vivo The graph below is from DMU 3 which was set to deliver at a constant rate of 4 29 mL day On two occasions during the
114. ed silicone is used for controlled release It acts like an artificial CL by releasing amounts of progesterone similar to normal physiological levels This delays estrus to allow herd estrus synchrony and or encourages non cycling cows to commence the estrous cycle Estradiol Estradiol by injection performs two functions in a synchrony program first to induce follicular atrasia at the commencement of a synchrony program and secondly to initiate estrus upon the withdrawal of progesterone controlled release devices Prostaglandin F2 One method for synchronisation is to give two doses of Prostaglandin F24 by intra muscular injection 11 days apart If the CL is less than 6 days old when the first dose is administered there will be no effect and the subsequent second dose will be certain to regress the CL If the CL is more than 6 days old both the first and subsequent second dose will regress the CL In this scheme AI is undertaken following the second dose of Prostaglandin F2 Background Prostaglandin F2 may also be used as part of a synchrony program with other agents such as estradiol and progesterone The role of prostaglandin F5 in conjunction with other agents is to regress the CL should one be present To achieve this with a single dose of prostaglandin F it is administered towards or at the end of the synchrony program to ensure that any CL present will be greater than 6 days old 2 4 2 Estrus Detection Be
115. ee 162 Symbolisation of One Monitoring Site sss 164 Network for Remote Design and Monitoring of Drug Delivery Prora ES o qeu adde addu du dada 165 DMU Schematic Top Level s eset tee 178 DMU Schematic Digital Board Microcontroller 179 Figure 51 Figure 52 Figure 53 Figure 54 Figure 55 Figure 56 Figure 57 Figure 58 Figure 59 Figure 60 Figure 61 Figure 62 Figure 63 Figure 64 DMU Schematic Digital Board Power and Supervisory 180 DMU Schematic Digital Board Peripherals 181 DMU Schematic Transceiver and Batteries Board 182 DMU Schematic Programming Cable esses 183 DMU PCB Layer Stack Up en at AA AAA A s 185 DMU PCB Layout Digital Board Side 1 and Inner Layers 186 DMU PCB Layout Digital Board Side 2 Power and Ground 186 DMU PCB Layout Transceiver and Batteries Board Side 1 187 DMU PCB Layout Transceiver and Batteries Board Side 2 Power ang Ground Planes stans tstatntennt btt uS 187 Modified M16C Evaluation Board Schematic Sheet 1 200 Modified M16C Evaluation Board Schematic Sheet 2 201 Modified M16C Evaluation Board PCB Layout ss 202 Base Station Development DMU sess 203 Modifications to DR3100 Transceiver Module
116. eived signal strength decreases the ratio of sampled values within bit periods will approach 0 5 Transmit at different powers with a description of the output power level contained as part of the payload in the data packet The weakest one to be received successfully can be taken a relative measure of range Replace the RF transceiver board with one based on inductive coupling technology The LibertyLink product described on page 74 would be ideal Aura Communications Inc 2001 The electronics can then be repackaged without the external antenna to make it suitable for a device which has two way communications and control ability to be located in the rumen Replace the viscous flowing drug vehicle with a gel with allows negligible mixing of the active ingredient at body temperature Such a gel would allow a multi drug regimen to be followed using the single variable rate dispenser currently used Different solutions are deposited in layers inside the syringe and to ensure the boundaries between these are clearly defined a small neutral gel layer is deposited between the active layers This allows for some inaccuracies in the volume delivered while acting as 151 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 152 a barrier between the active layers to prevent mixing The layers could consist of solutions containing the same active ingredient but at different concentrations Alternatively
117. ell Comments Trial Since Pres Observed Delivered Volume Syringe Control Reset surisation Piston from Piston Delivered by Piston Position Position DMU from Position Integrated Gascell Current dd mm yyyy Day min min mm mL mL mm hh mm 22 5 2001 4755 0 191 58 0 0 60 Piston start position at power on reset As per Pressurisation complete Pressure is 1 programmed i decreasing to minimal after initial delivery profile 2155 240 49 63 0 0 0 5 60 pressurisation so some extra leakage will occur The vehicle delivery programme started 49 minutes previously Two hours before insertion This does not accurately reflect the exact piston position just 23 5 2001 700 785 594 64 0 7 2 0 60 As per after insertion because the change in 0 programmed temperature causes a change in volume and delivery profile the piston will move slightly 0940 945 754 Insertion in to cow As per 24 5 2001 Could not download 2240 2049 programmed 0715 delivery profile Download accomplished 25 5 2001 Not Gascell forced 3795 3604 81 12 0 10 2 Taken out measured put back in 0911 measured off override switched gascell off As per 26 05 2001 Download accomplished 3 5160 4969 programmed 0730 Override made inactive delivery profile 27 05 2001 Gascell forced Download accomplished 4 6555 6364 0730 off override switched gascell off Download accomplished 28 05 2001 Gascell forced 8100 7909 90 18 0 19 2 61 Taken out measured put back in 0845
118. ency RF spectrum can penetrate a thin layer of mud water and other partially conductive materials at up to 1 GHz Cheap transmitters are commonly used in such applications as car alarm activation keys and garage door openers These typically have ranges in excess of 50m and are mass produced in low cost moderate performing systems One commercial device has been developed for animal identification and temperature recording for the purposes of animal health and possible estrus detection This is a device to be located in the rumen using two way RF communication with a read range of about 3m Magtrac Inc 2001 Later developments were going to add satellite boluses that could be ingested to magnetically attach to a permanently resident iron core Dispenser satellites for antibiotics worming and data storage would have been developed but the product has since been discontinued for unknown reasons Previous experimental work demonstrated another device operating at 418MHz from inside a human vagina which had an effective operating range of 10m McCreesh and Evans 1994 904 905 Scanlon Evans and McCreesh 1997 44 427 430 Bergen 1997 also had success from transmitting RF at 27MHz Development of a Practical Drug Delivery Communication and Monitoring System from the retro peritoneal space inside a monkey Likewise a commercially available telemetry device operating at 303 8MHz was used successfully to transmit glucose monitor data in
119. ent In spite of this situation there is a real market demand for low cost on farm pregnancy testing and heat detection systems Improved estrus detection and control products are near the top of the wish list that farmers would be willing to pay extra for Farmers are willing to pay around NZ 500 per set of cups or between NZ 20 to NZ 50 per cow MRL Research Group 1997 These market demands along with the recent increase in payouts to farmers provide impetus for research and product development in this area 13 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 2 4 14 Estrous Control Reasons for wanting to control the estrous cycle in cattle have been discussed above in the context of the wider dairy industry and are also well documented by others Rathbone et al 2000 1 201 228 It is now timely to list the more specific applications in one place before looking at the estrous cycle and how to control it with controlled drug release for maximum production efficiency Specific reasons for estrous control are Reduce time and labour involved in estrus detection Allow the use of set time insemination where all selected females to be inseminated naturally or artificially are processed at once This is more cost effective Synchronise donor and recipient in embryo transfer programmes Encourage non cycling cows to start cycling Enhance estrus behaviour in cows which exhibit s
120. eration On load current is approximately 16mA At the end of the seven day experiment this had fallen to 4 28V on load These voltages are quoted at room temperature Voltage in the cow is generally 200mV higher due the increased temperature To assess its useful lifetime the device was kept operational for a total of 40 days at room temperature without replacing the batteries After this time the on load battery voltage was 3 92V The minimum battery voltage for full function operation is 2 70V 0 90V per cell TYPICAL DISCHARGE CHARACTERISTICS Simulated Test at 21 C 70 F 1 6 1 5 1 4 Typical discharge after one o c 5 1 3 ee month voltage on one cell B 12 o g 1 1 5 1 0 Minimum full function voltage 0 9 0 8 0 250 500 750 1000 1250 Time hours Figure 45 Discharge Curve of Energizer EPX76 Silver Oxide Battery Adapted from EPX76 data sheet Eveready Battery Company 2001 During the experiment first week about five minutes per day of downloading was carried out on the device to ensure that the maximum amount of data was obtained in case it stopped working at any time This uses up a lot of battery capacity For the remaining weeks a much more realistic download pattern was used This explains why the voltage fell so much in the first week One month is probably the maximum period of reliable operational time with the 143 External Control
121. es RL73 Series CRGO0603 Series CRG0603 Series CRG0603 Series CRG0603 Series CRG0603 Series Manufacturer Supplier Neoohm Neoohm Neoohm Neoohm Neoohm Neoohm Neoohm Neoohm Neoohm Meggitt Polymers amp Composites Neoohm Neoohm Neoohm Neoohm Neoohm RS RS RS RS RS RS RS RS RS Farnell RS RS RS RS RS Supplier Order Code 213 1982 213 1982 213 2418 213 2418 213 2676 213 2676 213 2676 213 2676 213 2676 310 4540 213 2531 213 1982 213 2008 213 2193 213 2193 Top mark Comment 103 103 105 105 105 105 105 3R9 Farnell Feb 2000 catalogue page 1717 104 10R 271 271 Function For optional connection of TXR Rx Data to INTO For optional connection of TXR Rx Data to INTO Pull up for MCU NMI pin Was power fail signal to MCU NMI pin but manual says not to reset MCU when NMI low pull down for MCU CNVss pin Holds RxDO line low to prevent floating Must be kept an input line so that interrupt can occur in sleep mode TxDO ESD protection Holds RxD1 line low to prevent floating Must be kept an input line so that interrupt can occur in sleep mode TxD1 ESD protection Pull up for DQ line on gascell monitor internal MCU pull up used instead Pull down resistor for tilt switch to minimize current flow into MCU 10
122. etection methods for dairy cows might be based on on line sensing of milk Since this biological fluid is already being collected The author is an employee of Sensortec who specialise in developing such on line sensors Sensortec s relationship to this thesis is described in the preface Background from the animal on a daily basis it would be convenient to sample it for estrus detection Highly specific and sensitive tests are being developed to reduce the response time and cost of these systems to measure pregnancy and other factors in milk Ellison 2001 Publication No 211 41 45 Progesterone levels remain elevated during pregnancy but should cycle if the cow is not pregnant False positives for pregnancy occur frequently but the important condition of not being in calf low progesterone is always picked up Webster 1993 2 The focus of this thesis is on a drug delivery device mounted in or on the animal so on line sensing is not appropriate to discuss at length However milking robots which have the capability of measuring milk temperature and conductivity from each quarter could be used to transmit activation signals to an on cow drug delivery device Future Trends in Estrus Detection For the commercial farmer the manual collection of data for analysis is more effort than it s worth In addition the more accurate and sophisticated tests based on hormonal changes i e causes can only be carried out in the labora
123. f there is a command present but the signal strengths of any subsequent packets are logged The rate of transmission is reduced during the night to conserve battery power Contagious Model of Vaccination Delivery While the experimental results presented earlier demonstrate a simple form of communication useful for determining social grouping there are many other areas of potential application There is an interesting parallel between the spread of information by inter cow communication and the action of a contagious disease Imagine that a multi drug delivery device exists and is present in each animal This device contains as one of the available drug types a vaccine which is effective for a particular disease There are disadvantages to applying the vaccine so it is never applied unless absolutely required e g the vaccine for foot and mouth disease However for it to be effective it must be applied before exposure Now when the ID number of a cow is transmitted to other cows around it a flag indicating exposure to the disease is sent with it If an ID with that flag set is ever received it activates the specific vaccine dose and propagates that same flag to others around it Only the original animal needs the flag set by the detection of the infection The flag could be set by automatic means or manually over the wireless link Conclusion 6 2 In essence the spread of vaccine deployment exhibits biomimicry with respect
124. f strip circular cross section PIN209 Connector 1 pin SIL PIN CON RS 227 2639 rice is for 1 pin CNTRLO PCB 0 24 header pin strip single CNTRLO PINTFP of the 32 way interconnect row straight 32 way break off strip circular cross section PIN210 Connector 1 pin SIL PIN CON RS 227 2639 rice is for 1 pin CNTRL1 PCB 0 24 header pin strip singe CNTRL1 PIN1FP of the 32 way interconnect row straight 32 way break off strip circular cross section R101 Resistor 270R SMT 270R 0603 CRGO0603 Neoohm RS 213 2193 271 Provides load for 0 05 Format 196 tol 0 1W Series voltage droop test R102 Resistor 10K SMT 10K 0603 CRGO0603 Neoohm RS 213 2418 103 For boot monitor 0 05 Format 196 tol 0 1W Series program R103 Resistor 10K SMT 10K 0603 CRG0603 Neoohm RS 213 2418 103 For boot monitor 0 05 Format 196 tol 0 1W Series program R104 Resistor not 0603 Pull down for MCU pin 0 00 present as required by boot loader R105 Resistor 10K SMT 10K 0603 CRG0603 Neoohm RS 213 2418 103 Pull down for MCU 0 05 Format 1 tol 0 1W Series CLK1 pin as required by boot loader R106 Resistor 10K SMT 10K 0603 CRG0603 Neoohm RS 213 2418 103 Pull up up for TxD2 0 05 Format 196 tol 0 1W Series open drain R107 Resistor 10K SMT 10K 0603 CRG0603 Neoohm RS 213 2418 103 Pull up up for RxD2 0 05 Format 196 tol 0 1W Series open drain R108 Resistor not 0603 For optional connection 0 00 present of TXR Tx Data to TxD2 R10
125. fore the estrous cycle can be controlled it must be detected and predicted with some accuracy While estrus detection may not seem immediately important in the process of drug delivery per se the theme of this thesis is the automated control and monitoring of drug delivery for estrous control In doing so there is an emphasis on automated control of the delivery profile based on the current state of the estrous cycle It is necessary therefore to investigate what estrus detection methods are available The Cost of Missed Detection There has been much research into estrus detection This is because the cost of missed estrus to the farmer and the industry is high The single most important factor contributing to reproductive failure in dairy herds is poor heat detection Higher levels of milk yield per cow and fewer man hours per cow increasing herd size have complicated this important chore and reduced the efficiency and accuracy of heat detection Nebel and Walker 1998 Expression of estrus behaviour is poor in lactating dairy cows compared to heifers Nebel Jobst Dransfield Pansolfi and Bailey 1997 80 179 Also lactating dairy cows exhibit a high rate of early embryonic loss so it is important to pick up post AI estrus as an indicator of failed pregnancy These two factors make it extremely important that estrus detection is effective on dairy farms Calving intervals longer than 12 5 months cost U S and Canadian dairy produ
126. g error already mentioned It was still adequate to perform its primary function which is to help in assessing whether the DMU is correctly lodged in the cow In vivo delivery accuracy degraded to 7 from the bench top performance figure of 3 This is a provisional accuracy figure since it was obtained from a single delivery run from the only DMU to last the full seven days Discussion 5 1 5 Discussion Performance Assessment As a first field trial the experiment was very much a success Three of the four devices stopped working by day seven but not before achieving all objectives In particular the following objectives were met External control of drug delivery via radio link while in vivo Precision controlled variable rate drug delivery in real world conditions to the specifications given in Table 7 on page 117 This means a basic accuracy of delivery of 3 on the bench or 7 in vivo Inter cow communication for social grouping The wireless control and data exchange was carried out while the cows were marshalled into bales in the yard Data download was also demonstrated successfully over a fence while the cows were in an adjacent paddock able to free range This ability greatly enhances the usability of the system All of the initial requirements listed for the proposed delivery system on page 48 were satisfied Delivery accuracy transceiver range and battery life were better than required These facto
127. ge over the MC78FC30 There was some doubt as to whether this would result in a correct design Application advice from the voltage regulator design team at TelCom resolved the issue and it works very well in practice The TC1073 also has a low battery voltage error pin which can be used for an asynchronous power failure interrupt when it is enabled The active state and error pin status of the TC1073 are logged for diagnostic purposes Communication 3 7 1 Communication Medium Various mediums were evaluated for this application which requires a non contact range of 30m in the presence of industrial noise found in the milking parlour environment along with the special needs of the device to operate while being almost entirely inside the cow Optical Infra red Optical links have been used successfully in the past to transmit information from in vivo through a thin layer of skin and flesh Lindsey et al 1998 45 614 619 However the most practical way to implement an optical link would be to have a protruding sensor and emitter since this is allowed The specification calls for a device capable of working through a layer of mud and faeces These materials are opaque to visible and infra red light and so infra red is unsuitable in this application whether embedded or protruding Development of a Practical Drug Delivery Communication and Monitoring System lonic Conduction This is a novel technique whereby the bodily
128. ghout Stewart Finlay The University of Waikato Lots of time and patience in assembling some of the electronic hardware Dr Rainer K unnemeyer The University of Waikato Joint Chief Supervisor Guidance and motivation Links with local industry which provide worthwhile projects and future employment for his students Bruce Rhodes The University of Waikato Fabricating a hand made prototype PCB to the most exacting design rules Table of Contents IDSITAC Ecosse star Eae nat Mea em EK LA ES iii JACO Wedge ie il Sud ere E evt i Ree habete e e sac Edi V Table Of Contents ssc cts sievicdesayeats andes sonore lv deter rere Gees teda RYE vii ISG EPO P m xi Listot Tables So dee eet Rb aay ls a vn Ren ne ei ee pe XV CIOSSdEV ca iE eet Pu ecd e as ud ea ese eae sd ee od xvil INTRODUCTION zoiceiobitis ED qa Dee rex EET 1 1 1 DV ORV EW uS tivi tits use n a estt eects 1 1 2 Thes OSC BS naso t Sten eens tat eu dnd acit ic 2 1 3 Relationship With Industry sssssseeeeeeeneerenes 3 BACKGROUNDS aei risstorgeyee ie TUUS E RPLE SURVIE EYE or OERUAERe EMEN U DRE NAVI S OMEN oni o bas N OA 5 2 1 New Zealand Dairy Industry Structure 5 2 2 Current and Future Trends in the Global Dairy Market 7 2 3 Increasing Dairying Efficiency Through the Use of Controlled Drug Release eoe Nea nee nad ec UOI 9 24 Bstrons Controle soc eo ieu Ot Na e NU EDU 14 2 5 Controlled Drug Release for Estrus Control
129. ginal nylon spined insert which is coated with progesterone loaded silicone The purpose of the device is to control the estrous cycle in cows An endocrine gland that forms once the dominant follicle in the ovary has released its egg under the influence of the lutenising hormone The corpus luteum phase of the estrous cycle then regresses In the pregnant animal it remains active throughout pregnancy The corpus luteum is destroyed by uterine secretion of prostaglandin F2 Delivery and Monitor Unit The intelligent delivering monitoring and communicating drug delivery system developed for this project The biological cycle regulated by the endocrine system that controls the physiological and behavioural aspects of a non human mammal s reproductive system Period of the estrous cycle during which the female is receptive to the male heat This is accompanied by behavioural and physiological changes in most animals Memory which retains program code and data even when power is removed The M16C microcontroller used in this thesis has 256K of flash memory which can be re written in blocks during program execution The same size and shape as a button cell calculator battery the gascell produces hydrogen when an electrical load is connected to it Assuming other factors are constant gas production is directly proportional to current drawn The gascell is used in the Theratron syringe to propel the drug vehicle Micro Electro Mechanical
130. glandin degeneration Figure 2 Cow Reproductive Process Follicle Development and Degeneration Webster 1993 2 After Ovulation Once an egg is released from the dominant follicle at the time of ovulation the remaining part of the follicle then develops into the Corpus Luteum CL The CL releases progesterone for the rest of the cycle which readies the uterus for implantation of a fertilised egg Progesterone is the green light for the start and continuation of pregnancy If there is no fertilisation prostaglandin F2a released by the uterus acts to regress 17 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 18 the CL resulting in a rapid decline in progesterone level An injection of prostaglandin F2a can likewise regress the CL This lowers progesterone quickly and shortens the estrus cycle However the CL can only be regressed by prostaglandin F if it is more than 6 days old i e later than day 5 in the cycle If fertilisation does occur the egg continues down the horns of the uterus and embeds in the uterus to develop The subsequent release of interferon T inhibits the release of Prostaglandin F there by maintaining the CL and inhibiting further ovulation Since the CL survives for the duration of the pregnancy progesterone levels remain elevated Estrus Synchronisation by Application of Artificial Hormones Progesterone A nylon insert coated with progesterone load
131. griculture is a top priority A successful mating programme enables a desirable calving pattern Farmers have less control over conception rates during an artificial breeding programme but these can still be enhanced by increasing the proportion of cows cycling before mating starts ensuring cows are correctly submitted for insemination High producing lactating dairy cows exhibit a high rate of non cycling and early embryonic loss Nebel et al 1997 80 179 The mating programme must therefore actively plan to address these issues As mentioned the New Zealand dairy production system is a comparatively low cost efficient system This is because it works with the seasons to directly capture the pasture grown to produce milk This system necessitates a 365 day calving interval with a condensed pattern of calving beginning in late winter so that utilisation of the spring pasture flush is maximised Reproductive performance determines the sustainability of this system from one year to the next The recent industry trends of increasing stocking rates improving genetic merit and lactation lengths with an emphasis on production per hectare are demanding more advanced breeding technologies to maintain a fertile herd in a sustainable system Essential aspects of any programme aiming to improve breeding efficiency are an in depth understanding of the estrous cycle and the relevant pharmaceutical science Ogle 2000 As we find out more about how th
132. h catches all 5 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 84 burst bit errors of 16 bits or less and all longer bursts with 99 998 probability Forward Error Correction Systems which detect errors during transmission and simply ask for data to be resent can suffer from excessive numbers of retries and collisions The more retries which occur the more likely it is that further collisions and retries will occur Sufficient error detection is performed by the use of the cyclic redundancy check however it is possible to add the capability to correct for some of these errors without the need for resending By employing Forward Error Correction FEC additional information is sent along with the data bytes to be transmitted allowing the receiving device to detect and attempt to correct for some or all of these errors In the proposed channel encoding scheme a simple block code is used where every group of six bytes has two bytes of redundancy added This allows correction of all single errors and all two bit adjacent errors A bit error rate improvement of between 3dB and 10dB can be achieved by forward error correction Interleaving Next comes interleaving Errors usually occur in bursts in a wireless environment so the interleaver rearranges the order of the encoded data bits so that these errors are widely dispersed After decoding at the receiver these burst errors appear as widely spr
133. h per cow initial cost of deployment Bou Matic 2001 This is probably only true for overseas farm conditions due to the fact that the passive ID is included in the same package If a passive ID tag is required on the animal then including motion detection for little extra cost makes sense In New Zealand however machine readable ID tags are rarely used so the cost involved to use such a device purely for estrus detection does not compare with the cheap and effective tail paint method Burke 1999 In future European legislation is expected to make machine readable livestock ID tagging compulsory due to a desire for traceability It is already technically 31 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 32 feasible to trace an animal from the moment of birth through processing to the checkout counter at the supermarket Organic produce authentication for the consumer is a prime example of the need for better traceability The use of antibiotics and chemotherapeutics in animal husbandry has led to the occurrence of veterinary drug residues in all types of food of animal origin Sternesjo Mellgren and Bjorck 1995 10 1006 175 181 Organic produce is increasingly in demand and since the consumer expects to pay a small premium this allows for the extra costs involved When this demand strengthens considerably there could be strong market pull for machine readable ID in European countries a
134. he ability to store and release multiple drugs or chemicals from a single device The main disadvantage is the tiny amount of room available for drug storage which makes it impractical for doses over 10uL Micro Needle Array Microneedles are microfabricated devices for minimally invasive drug delivery applications which can be integrated into a variety of devices Zahn et al 2001 503 506 A portable drug delivery device with integrated microneedles and on chip MEMS positive displacement micropump has been developed Santini jr Cima and Langer 1998 Santini Richards Scheidt Cima and Langer 2000 32 377 379 Santini Langer and Cima 1999 397 335 338 This works by generation and collapse of thermally generated bubbles which propels the liquid drug along the length of the needle while check valves assure pumping in the desired direction Very low rates of volume are possible down to 1 0 nL s Hydrogel Researchers are developing a device made from porous hydrogel Once inserted it is capable of expanding to several times its own size This device is about the size of a matchstick and perforated with microscopic hydrogel controlled holes The hydrogel is addressed by a backing of conductive plastic Current flowing through the backing makes the holes in the hydrogel open allowing it to release variable amounts of medication lontophoresis Ionic or non ionic drugs can be persuaded to move by applying a voltage gradient Drug flo
135. he last 20 of their useful life Reset IC input bias 2 network Reset IC 4 Pressure transducer 1 Gascell switch 1 Leakage on spare 4 microcontroller IO pins Zener diode reverse 2 leakage Light sensor LDR bias 3 Total 119uA Current Consumption in Fully Awake Mode With the microcontroller fully awake transceiver mode polling no longer occurs The transceiver is either in transmit mode or receive mode The following table summarises the changes in current draw when the microcontroller is fully awake 71 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 72 Table 3 Current Consumption in Fully Awake Mode Item Comments Current mA Microcontroller 10MHz running at divide by 1 no wait bus 8 cycle Transmitting 10mA peak 6mA average Transceiver Receiving 4mA 5 Overall average 5mA LED 3 Other activity including all loads present in sleep 1 mode Total 17mA 3 6 4 Power Fail Warning A brown out is always terminated by a proper reset signal rather than the Vdd simply dropping below that required for reliable operation This function is provided by the reset generator U106 in Figure 51 on pagel80 The reset delay is chosen so that if the batteries were suddenly disconnected stored energy in the power supply capacitors would allow operation at full clock speed to occur for at least 1 7ms 17 000 clock cycles This gives the power fail interrupt routi
136. he receiver polling cycle When the gascell is off the microcontroller takes the DQ line low before entering sleep mode This puts the gascell monitor in a low power mode where only the real time clock functions The battery monitor has it own real time clock This saves power without affecting integrated current measurements Although it may appear on the DMU schematic diagrams in Appendix B that spare microcontroller IO pins have been left floating that is not the case All unconnected IO pins are configured as outputs which reduces the current per pin to around 50nA Table 2 gives the measured current from all parts of the device in sleep mode Development of a Practical Drug Delivery Communication and Monitoring System Table 2 Current Consumption in Sleep Mode rren Item Comments Surrent uA Wait mode with two timers and one counter on 3 Microcontroller some hardware interrupts enabled 32kHz oscillator in high drive mode 10MHz oscillator off Duty cycle operation receive mode for 8ms 60 Receiver polling sleep mode for rest of the 600ms cycle Average current is 60uA Power supply Leakage current 2 decoupling capacitors Assumes monitor is in low power mode about 29 Gascell monitor 80 of the time The monitor draws 32u4A if on 22uA in shutdown Regulator combined ground currents TC1073 regulator in shutdown 8 Add 48pA if TC1073 regulator is on This occurs when the batteries reach t
137. he small enclosed space of the syringe body Current leakage from power supply decoupling capacitors and IC output pins is also reduced Noise margins at inputs are lower however the sources of on board noise generators are reduced For example ground bounce from high speed switching is reduced due to the lower aiming voltage of the switching device This means less spurious RF emissions Williams 1993 Application Note AP 477 There are some disadvantages Noise due to externally induced currents has increased ability to corrupt signal levels 3V devices are more fragile CMOS 3V devices have thinner silicon oxide gate layers making them more susceptible to electrostatic damage 3V parts are usually de rated in terms of speed due to the thinner gate layers increasing the capacitance The maximum speed of the chosen 3V microcontroller is 10MHz compared to its 5V counterpart which can go to 16MHz In summary although there is less noise immunity and a reduced maximum microcontroller clocking speed imposed by the system wide 3V power supply choice this is greatly outweighed by the increased battery life and the fact that less cells in series are required to achieve the required battery operating voltage 67 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 68 3 6 3 Microcontroller Mode Control To get a useful life out the batteries careful control of the power state of the micro
138. he success of the transceiver operation Normally microcontrollers are not used in close proximity to radio frequency components The spurious signals coupled from the microcontroller clocking elements into the radio circuits can affect both transmission and reception To mitigate this the transceiver and microcontroller are placed at opposite ends of the device Attention to power supply distribution and other design rules allows a successful arrangement of components The six layer PCB has dedicated power and ground planes on each board which along with other techniques means bulky shielding is unnecessary As a result the electronics package including batteries and gascell fit into a cylinder 2 8cm in diameter and 2 5cm in length Four copies of the DMU exist Development of a Practical Drug Delivery Communication and Monitoring System 3D PCB CAD Model of Side 2 of Digital Board for Visualisation of Fit Motion Sensor Pressure Sensor Sides 1 and 2 of Digital Board Shown Actual Size 28 6mm Diameter eet Antenna Antenna Connector Over Pressure Release Cap Over Pressure Release Cap Retainer Transceiver and Batteries Board Transceiver Hybrid Gascell Gascell Monitor IC Reset Switch M16C MCU Sub Clock Crystal 32 768kHz Main Oscillator Crystal 10MHz Electronics Package from Microcontroller End Digital Board Transceiver and Batteries Board Transceiver
139. hem difficult to remove during Background 2 5 meat processing It is therefore a minimum requirement that they be safe to ingest if this happens by accident If tags become unpopular other means of identification that lend themselves to an automated approach include nose print and iris identification Iris identification looks more promising than nose printing because it is a non contact method It is an experimental technique which works by assessing the 500 or so measurable metrics from an automatically acquired image of the iris Currently in commercial use for human eye recognition a version for animals has been trialled One disadvantage is that the iris pattern is not stable in cattle until 6 months of age Controlled Drug Release for Estrus Control It has now been established what the estrous cycle is why estrous control is important how estrus may be detected and the consequences if it is missed This section describes the aims and methods of controlled drug release for estrous control along with experimental and commercially available products for doing So Depending on the needs of the farmer the use of estrus control products may be restricted to just a few cows in every hundred If the farmer is not concerned with synchronisation but simply wishes to jump start non cycling cows then only 10 to 20 may require treatment 2 5 1 Controlled Drug Release A Definition Traditionally the easiest way to de
140. here is also a Light Emitting Diode LED that can flash in one of three colours This LED D101 in Figure 52 on page 181 actually consists of one green LED and one red LED but if they are both on it appears orange This is used as an indicator to convey one way status communication without the need for a base station The LED indicates as follows On power up reset the light goes orange then rapidly blinks red while the power on message transmits then flashes alternately red and green a few times At the end of the red green flashing the internal clock is zeroed At the end of every minute a very short green flash 30ms means the internal clock has ticked over another minute If the flash is longer Development of a Practical Drug Delivery Communication and Monitoring System 600ms this indicates that a result set is being recorded for data logging When receiving data the light is steady green When transmitting data the light rapidly flashes red Ifthe light is rapidly flashing orange on and off the batteries have run out and the DMU is continually resetting itself 3 7 6 Wired Communication Port If the transceiver fails wires can be soldered onto the DMU to download any important data that might be otherwise inaccessible 3 7 7 Programming and Debugging Port Setting the Program Run switch to program and pressing reset allows the DMU software to be upgraded over the programming serial port Se
141. hich is permanently on It generates an interrupt exactly once a minute to update the real time clock counter 3 8 3 Unique ID The gascell monitor integrated circuit U102 in Figure 52 on page 181 has a unique code 64 bits long This is lasered onto the IC at the factory and guarantees that every DMU ID will be unique The first 8 bits are a manufacturer family code The next 48 bits the unique serial number The last 8 bits are a CRC of the first 56 bits 3 8 4 Data Logging and Retrieval Every 15 minutes a full set of measurements is taken and stored The measurements taken are given in Table 5 Development of a Practical Drug Delivery Communication and Monitoring System Delivery Profile Management Table 5 Logged Data Result Set Gascell Actual Integrated Current mAHr Estimated Volume Delivered mL Profile Estimates Gascell Target Integrated Profile Targets Current mAHr Target Volume mL Gascell Voltage V Gascell Instantaneous Readings Gascell Peak Current mA Gascell Switch Gascell Load Control Flags Gascell Override Gascell Override Value Sensor Data Light Intensity Temperature From Gascell Monitor deg C Temperature From Pressure Sensor deg C Pressure mbar Motion Cycles cycles Motion Open Cow Proximity Flags Cow Number 1 Proximity Cow Number 2 Proximity Cow Number 3 Proximity Cow Number 4 Proximity Anonymous Cow Proxi
142. ibration Data Gascell Sense Resistor 3 90 Ohms V reg Sense Resistor 2 20 Ohms Pressure sensor offset 8 00 mbar Temperature from pressure sensor offset 0 00 deg C Temperature from gascell monitor offset 1 50 deg C Time Base Configuration Safety Limits Main time base reload value 61440 Main time base minutes increment 1 Time acceleration factor 1 00 One simulated minute lasts 60 00 seconds Transmit ID period 30 minutes Record result set period 5 minutes Maximum safe pressure 1350 0 mbar Maximum safe gas volume 100 0 mL Miscellaneous Wired comms port TC1073 Voltage regulator switch over 3 50 V MCU UART 1 input gt 5 DMU 4 provides user prompt for next command if required 175 Appendix B DMU Schematics Appendix B DMU Schematics 177 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals puvjpaz aw s jo 1 pets Bose oe ID RC NUES alg 7774 pow doy AWA mL v preog uisi L yogspesoydiieg YOg JOUIN qos Alostaadng pur mod a Pensa ps qp amsa 209 PU AT ases sop NBT INOG amesar INOG anssony NIC assag NIT aiti NION assag NION amsa WIDS amsa WIS omia wa G3 aD CT Pag CIT Py CST morass SoA mops Sia wuz EEN pg Saga NI pans pore NI prng posg EA ee trg awp poro ERG pio THIRD FEFE NE Serm 2 2999
143. iginal experimental results were Social grouping determined by inter cow wireless communication Precise delivery of a pharmaceutical vehicle over a short or extended period of time in accordance with a complex profile This was to an accuracy of three percent under controlled conditions or seven percent in real world conditions in vivo External wireless control override of drug delivery in the cow Normal and external controlled delivery subject to safety limit override criteria The ability to gather an array of environmental data at once from the cow vagina by wireless communication 169 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 170 6 5 2 All Initial Design Criteria Met or Exceeded System performance was assessed and found to meet or exceed all initial design criteria Attention was paid to contemporary controlled release mechanism design philosophy The system exceeded the initial design criteria in the following areas Real world in vivo accuracy of 7 was better than the required accuracy of 10 Range of communication was up to 60m line of sight compared to the required 30m Battery life was seven weeks greatly exceeding the three week requirement 6 5 3 Future Work Possible improvements to the existing system were noted These included improvements in mechanical reliability alteration of the software to support longer term data gathering and the a
144. ilent heat for the purpose of easier heat detection Make artificial insemination more practical AI is best carried out herd wide simultaneously and its wider use promotes less risk of transmitting diseases To time calving patterns closer to the desired optimum schedule 2 4 1 Estrous Cycle in Cows The purpose of the following discussion 1s to establish a familiarity with a useful and immediate application of drug delivery in farm animals rather than being an authoritative and complete reference on the latest in estrous management For the sake of brevity it makes various simplifications and generalisations about the Background complex nature of the estrous cycle A more comprehensive description on estrous control and veterinary drug delivery can be found in the book Controlled Release Veterinary Drug Delivery Rathbone and Gurny 2000 1 While not the main topic of this thesis a basic understanding of the cow reproductive organs and endocrine system is essential before designing a drug delivery system to interact with it pelvic bones A vagina horns of uterus vulva pubic bone oviduct bladder left hornof o uterus o 9 o9 o0 cotyledons ei o 7 x oviduct W body of O lt uterus KON ovary with infundibulum ripe follicle and fimbriae f opening of vagina cervix opened j il Ih il NN __ vulva opened orifice ventral commisure clitoris S URINE Figure 1 A
145. in a statistical sense but short run variation can periodically lead to unacceptable results The extent to which the short run variation departs from the required channel characteristics is also dependant on the instantaneous transmission range and other factors Efficiency is related but not equal to the scramble key payload size ratio Some systems are more efficient because they do not explicitly have to transmit the scrambler key The Bluetooth standard does not need to transmit a scramble code on each packet transmission due to a synchronised clock in both the master and slaves The scrambler key is generated from a combination of the master 48 bit 87 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 86 unique ID and the synchronised clock value Bluetooth SIG 2001 Packetisation The packets are now assembled and sent to the transceiver modulation input pin as follows Cycling packet number To From Payload Preamble address address length Protected by FEC interleaving and eight bit to ten bit encoding in that order Figure 15 Air Interface Packet Structure Normally the preamble is only 2ms long but when the transmission is from the base station to the DMU the first packet to be sent is preceded by 800ms of preamble This tickle wakes up all DMU s in range and keeps them awake in full receive mode The base station does not require tickling since it is in receive
146. in the animal are influenced not only by the amount delivered artificially but also by naturally occurring levels in the animal Webster 1993 2 which is in turn dependant on a variety of external factors These factors include diet extremes of weather age calving history general condition and predominantly the current point in the estrous cycle There are two main barriers to such an approach The first is the difficultly of 159 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 160 sensing useful levels of the desired control signals At present there is no easy way to sense hormone levels with a remote embedded device that is sensitive easy to integrate durable and non intrusive Instead the routine approach is take a blood sample with a needle and send it to a laboratory The second barrier to developing a control system which includes the organism is the added complexity of device animal interaction Hacisalihzade 1989 9 44 45 Rates of delivery in terms of mL day are immediately understood and easily verified By contrast a control loop involving the animal requires a deeper level of understanding of the animal itself and a mathematical model of the transfer function for the actuator feedback path Further it is unlikely that the end goal of the drug regimen can be quantified with only one measured variable biological systems are rarely that simplistic Instead a multi variate control
147. ing the environment in which drug delivery systems must operate in what they are exposed to and the biological feedback mechanisms it attempts to interact with 149 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 150 5 4 3 Hardware Changes The current system can be described as a general computing communications platform interfaced to specialist drug delivery and monitoring hardware There is ample capacity for additional computational software and low power hardware requirements in future upgrades Beyond the trivial changes mentioned above hardware software changes could specialise the existing platform for specific use experiments In line with this InterAg are interested in further studies employing the device to measure the pressures in the cow vagina Currently the DMU only senses pressure inside the syringe for self diagnostic purposes The next phase of experimentation will be to add the capability of measuring pressure external to the syringe body Other ideas are to Develop a data network and configuration software for in situ alteration of fundamental drug delivery device behaviour This includes downloading new delivery profiles and control programs over the wireless link currently achieved by attaching a wire This is technically possible given the right software since the microcontroller can re program its own flash memory over the wireless link Develop a desktop applicat
148. inge internal pressure relaxes towards ambient slowly but the total volume delivered is within the greater of 3 or 0 5mL of the target value at all times After a successful bench test it is natural to question whether this accuracy will transfer through to an in vivo application of the device Fortunately Bunt Rathbone et al 2001 had previously determined that in vitro and in vivo rates of vehicle delivery using the Theratron syringe are essentially the same Their experiment used a fixed value load resistor over a 7 day period with a sample size of three Figure 26 shows their results 25 t 20 ti D 9 15 3 T 0 E E 5 0 0 1 2 3 4 5 6 7 Time days Figure 26 In vitro L and In vivo Rate of Delivery Bunt et al 2001 28 To save space here all other data that the DMU is capable of generating is presented in the form of real data downloaded over the wireless link from the in Development of a Practical Drug Delivery Communication and Monitoring System vivo experimental results in the next chapter 3 14 Specifications The major specifications and benchtop performance of the DMU system are presented in the following table Table 7 DMU Specifications Operational temperature range 10 to 50 C dependant on properties of solution being delivered Operating voltage 3 0V nominal Battery life 4 to 7 weeks dependant on usage profile II ile i j Overa Sleep
149. integrated current to volume dispensed ratio is then altered by this empirically determined scale factor The calibration factors for each individual unit are stored in flash memory They are selected using the machine readable ID number during power on reset initialisation The pressure threshold at which to first apply leak compensation is determined simply by looking at the pressure trace during a zero delivery period to determine Development of a Practical Drug Delivery Communication and Monitoring System where the volume just stopped being delivered 3 5 6 Gascell Voltage Measure The voltage on the gascell varies between 0 4 and 1 8V depending on the composition of surrounding gas load on the cell and remaining capacity Gascell voltage is logged for diagnostic purposes only 3 5 7 Gascell Current Measure Instantaneous gascell current is likewise logged for diagnostic purposes only This happens once every 15 minutes and should not be confused with gascell current integration discussed next 3 5 8 Gascell Integrated Current Gascell integrated current is used to provide an estimate of how much volume has been dispensed With the gascell turned on current flows through the gascell current sense resistor R121 in Figure 10 A voltage is developed across this resistor which is measured to find instantaneous gascell current Although this resistor has a rather large 2 manufacturing tolerance the effective toleran
150. ion of estrus behaviour is significantly enhanced by the second treatment of CIDIROL injection e Majority of animals will show estrus on Day 10 Figure 3 Application of CIDR to Bovine Estrous Cycle InterAg 2000 SMARTT1 Intelligent Breeding Device The SMARTT1 Intelligent Breeding Device IBD from Plade Holdings has batteries and an electronic timing system to deliver one reservoir of 5mL of drug over an extended time period A solenoid operates a valve that controls the drug under pressure from a spring pressurized piston The drug is released in small bursts several hours apart and timed to compensate for decreasing spring force as 38 3 The relationship between InterAg and this thesis is stated in the preface Background the reservoir is depleted In addition there are three small spring loaded bolus pistons that are activated by melting a plastic thread over a resistor This device attempts to deliver a drug programme comparable in nature to that used with the CIDR but in a more automated fashion Later versions of the device have more sophisticated delivery regulation gs 4 Silicone covered a Spring load electronics y bolus capsules amp 2x4 i Solenoid M d a a f Drug reservoir Spring Retrieval cord W il not pictured Figure 4 SMARTT1 Intelligent Breeding Device Plade Holdings 1999 There is no doubt that the SMARTT1 IBD should perform well in the kind of
151. ion to plan and download drug regimens and store the results in a database accessible over the company intranet and or Internet This is currently achieved by using a variety of off the shelf applications Drug profiles are altered in a textual format using a standard text editor A standard terminal program downloads the profiles and retrieves the results These results are then manually manipulated using a spreadsheet Integration of these tools into a single application would make it easier to use Utilise the flash memory for data logging Currently only 10 of available memory is used because all sensor data is accumulated in RAM only There is approximately 200K of flash memory that could be used for Discussion additional data logging Inter Device Range Estimation Output power is constant over the life of the battery However for a fixed distance some variation of received signal strength is to be expected due to variations in build component tolerances placement in the animal and particular propagation effects such as multi path fading There is no direct indication of received signal strength from the transceiver integrated circuit but received signal strength could be estimated using one of the following techniques Over sample the data bits being received Once the receiver clock has synchronised with the transmitter clock each sample should be of the same value throughout the duration of the bit period As rec
152. ional functions such as status indication external control and telemetry Temperature telemetry is very common in this respect since it is easy to include and is an important biological parameter 2 5 3 Established and Experimental Routes of Controlled Drug Delivery Subcutaneous and intravaginal routes have been used successfully in the control of the estrus cycle in cows and ewes on a commercial scale Rathbone Burns Ogle Bunt and Burggraaf 2000 1 201 228 For a drug to be useful via the intravaginal route it must be have rapid absorption a low minimum effective concentration be relatively soluble and be non irritant Alternative routes used for other experimental and or commercial applications in animals include ocular Gurny et al 1997 28 335 361 Li et al 1986 3 213 218 Olejnik et al 1985 37 118 ruminal Rathbone and Gurny 2000 1 oral dental Cleland 2001 50 261 275 rectal and intra muscular Mason et al 1996 13 243 252 Transdermal patches and topical treatments for veterinary use have also been investigated experimentally There are limitations with current transdermal technology developed for human use when applied to animals due to differences in body mass and skin transport properties Riviere and Papich 2001 50 175 35 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 36 203 Topical treatments are currently in veterinary commercial use and more are being deve
153. iquid form is present only in slugs distributed along the tube separated by sections of air As the gas production proceeds at a constant rate at the sealed end 43 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 44 of tube the slugs are delivered periodically in a pulsatile fashion at the other end Naturally delivering air along with the drug limits the field of application Nonetheless a delivery device with a coiled tube for the pulsed release of nitro glycerine has been developed and demonstrated successfully for transdermal application Groning and Kuhland 1999 193 57 61 To go beyond constant rate or simple predetermined pulsed delivery requires the addition of some control circuitry Such devices allow complex drug delivery profiles to be designed before activation commences Modification to that profile can occur during the programmed delivery based on external control or information from environmental sensors Practical devices along these lines have been built and assessed for reliability and accuracy of delivery A rudimentary device of this type consists of a small tube 25mm long with batteries a magnetically operated reed switch a gas production cell and a drug reservoir Groning and Weyel 1993 39 102 104 A battery supply can be turned on and off by a reed switch external to the patient A computer controls an electromagnet up to 10mm away which in turn affects the reed switch
154. is thesis Thus the individual elements of such a system already exist in working form To realise the system in the diagram requires only system integration and testing Remote Data Gathering From Many Sites At present information is obtained by addressing each animal on a manual basis To totally automate the process a network of information gathering sites could be built The transponder in each cow would be interrogated by a base station placed in a position such that each cow comes within adequate range often enough for the 163 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 164 system to be of use The typical range is 30m so a base station placed inside a milking shed for a milk producing herd would be ideal The base station can also be made portable if required Looking at one system installed on a farm as a whole it can be symbolised as follows Herd Station Figure 47 Symbolisation of One Monitoring Site The base station periodically asks if any transponders are in range If so it determines if any need servicing and then asks those requiring service to respond in turn Results such as those already achieved by the DMU are recorded in the local database for each cow The base station in this set up is not very intelligent and is essentially only an interface to the controlling program and database on the nearby personal computer that makes all of the important decisi
155. itor 100nF X7R 100nF 0603 Philips Farnel 432 210 Decoupling for Vref on 0 17 16V SMD Ceramic Components MCU C106 Capacitor 6 8pF 596 6 8pF 0603 Philips Farnel 721 943 load for MCU main 0 13 NPO SMD Ceramic Components clock C107 Capacitor 6 8pF 596 6 8pF 0603 Philips Farnel 721 943 load for MCU main 0 13 NPO SMD Ceramic Components clock C108 Capacitor 100nF X7R 100nF 0603 Philips Farnel 432 210 Noise filter for gascell 0 17 16V SMD Ceramic Components voltage measure C109 Capacitor 100nF X7R 100nF 0603 Philips Farnel 432 210 Power supply 0 17 16V SMD Ceramic Components decoupling for gascell monitor C110 Capacitor 100nF X7R 100nF 0603 Philips Farnel 432 210 Part of low pass filter for 0 17 16V SMD Ceramic Components gascell current accumulator measurements C111 Capacitor 100nF X7R 100nF 0603 Philips Farnell 432 210 Power decoupling for 0 17 16V SMD Ceramic Components pressure sensor 189 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Designator C112 C113 C114 C115 C116 C117 C201 C202 C203 C204 C205 C206 D101 D201 F101 190 Description Capacitor 100nF X7R 16V SMD Ceramic Capacitor 100nF X7R 16V SMD Ceramic Capacitor 10uF 10V Tantalum SMD case size A ESR 3 OHMS 100Hz 3 2x1 6mm Capacitor 15nF SMD 5 NPO Ceramic Capacitor 10uF 10V Tantalum SMD case size A
156. l Monitoring 3 9 1 Temperature Sensor Given that vaginal temperature has proven to be such an important physiological factor a temperature sensor has been included in the DMU In fact there are two independent sources of temperature measurement available from opposite sides of the digital board Accuracy is 0 1 C for both One source comes from the gascell monitor IC and the other from the pressure sensor IC The gascell monitor IC temperature sensor U102 in Figure 52 on page 181 is of the digital IC type relying on the difference in frequency between two oscillators One oscillator is designed to be sensitive to temperature variation and one not The difference in frequency is converted to a digital form and stored in a register for subsequent access when requested The temperature sensor on the pressure sensor IC is described later Development of a Practical Drug Delivery Communication and Monitoring System 3 9 2 Pressure Sensor Since the syringe relies on differential pressure across the piston to deliver the drug the ability to obtain pressure readings helps in understanding the experimental system In fact only internal syringe pressure is measured No account is made of fluctuations in ambient external pressure which have the potential to affect the delivery rate slightly The MS5535 is a 0 to 14 bar pressure sensor from Intersema Switzerland and is represented in the schematics as U104 Figure 17 shows the schem
157. l functions Although not drug delivery devices per se internal pacemakers which include some form of automated defibrillation are notable These devices prove the immediate benefits for the patient in developing such embedded portable systems The effort involved in discovering and applying the much needed application specific knowledge can be justified on a number of additional grounds Assessing the behaviour of the actual device animal interaction and comparing it to the model would lead to a greater understanding of the biological system under consideration 6 2 2 Measuring Control Factors from Excretions Instead of placing the measuring apparatus in the animal an indication of internal status can be provided by animal excretions This greatly simplifies the need to handle the animal or subject it to implantation of devices Faeces and urine are example of excretions which have been used to determine the presence of various chemical states within the cow Some promising work has been carried out which gauges the stress level and general health from examining the ketone levels in the cow s breath These herd based systems would provide individual and collective data to a remote computer Integration with herd management software would close the control feedback loop allowing the devices to be controlled over the Internet A more practical excretion for a production unit is the on line testing of milk the advantage being that
158. l impact due to discarded devices Capable of delivering multiple drugs Capable of delivering complex drug profiles including variable rate stop start and bolus While the attitude displayed in such a design philosophy is admirable it is exceedingly difficult to implement from an engineering perspective After initial implementation choices are made for the core technology the main design decisions are usually centred around compromise in order to achieve adequate performance Often equally important design criteria contribute to opposing design goals A device may need to be reduced in size to fit into a particular cavity but the act of making it smaller decreases the accuracy of the intended delivery because the formula must then be more concentrated Radically new technology is sometimes required to leapfrog over such limitations The philosophy above was noted in the course of developing the following system but compromises were made in order to produce a device within timeframe and resource constraints To limit the amount of risk involved standard techniques and off the shelf components were used where possible A critique of this design process is presented in the discussion section 47 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 3 Requirements of the Proposed Drug Delivery 46 System Analysis of InterAg s requirements their present experimental delivery device and the
159. l interrogations Although more complex this modification more accurately determines received signal strength since there are two transmissions per exchange along the same path This reduces the amount of network traffic and increases battery life since close animals are not unnecessarily interrogated Broadcast Multi Hop Hot Potato At regular intervals each animal asynchronously transmits its ID number to those 157 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 158 around it Those receptions are logged with the payload containing the received signal strength time the original packet was transmitted and ID number of the originating animal The ID number of the receiving animal is appended to the data and it is rebroadcast after a time delay based on the ID number of the receiving animal so that collisions do not occur with other animals responding in the same way If an animal receives a packet and it sees its own ID in the payload it does nothing This prevents data packets from continually circulating throughout the herd This data is of the same quality and can be processed in a similar manner as with the one hop passive technique The originator also logs any of the forwarded packets that originated from itself for another estimation of animal grouping The packet may take one of several paths or even arrive via multiple paths to a given animal Only the first packet to arrive is processed i
160. lable for every day use at each milking within two years Other hormones in blood can also be monitored for determining how the estrus cycle works and at what state it is in For example lutenising hormone peaks regularly during the follicle wave phase spiking markedly around day 18 of the cycle a Multi Variate As stated conductivity temperature and other effects are influenced by more than one cause so any system based on these factors alone will be ineffective One answer is to use a combined analysis of a variety of measurable quantities can form the basis of a reliable detection system Mitchell Smith and Sherlock 1995 42 Such a multi variate approach using activity concentrate intake milk yield temperature and conductivity has met with some success but always with a variable number of false positives and negatives de Mol 2001 126 99 103 de Mol and Ouweltjes 2001 49 71 82 de Mol and Woldt 2001 84 400 410 In one trial sophisticated filtering of an accumulated database with two years worth of data true positives and true negatives for estrus and subclinical mastitis ranged from 65 to 99 4 de Mol et al 1997 48 219 227 Other Experimental Techniques Other less obvious estrus detection techniques could also prove useful These novel ideas in the experimental stage include breath ketone urine and pheromonal analysis Applicability of On line Milk Sensing Methods In future more efficient estrus d
161. lectronics package is installed the act of sealing the syringe increases the pressure to about 50 mbar above ambient The gascell is switched on with a 100 duty cycle to pressurise the A Syringe as quickly as possible Initially consumption of oxygen by the gascell reduces the pressure to below ambient Hydrogen production commences once all oxygen is consumed and the pressure increases until the piston starts to move All syringes were controlled with the pressure start override command via the wireless link so that B they were synchronised with each other At this point drug delivery profile time has commenced time equals zero on all graphs On insertion into the cow temperature of the vehicle increases There is a corresponding drop in C vehicle viscosity The resulting decrease in backpressure from the piston gives rise to a lower average pressure for the remaining part of this graph D Pressure spikes from individual active gascell periods are clearly evident Figure 37 Downloaded Results Pressure Temperature and Estimated Volume Delivered Close Up 135 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Pressure Temperature and Estimated Volume Delivered Full Duration Day of Trial E 0 1 2 3 4 5 6 7 1150 r 160 1100 N D aro L 23 ne SL 0X qu Fo Au NY w 9 be Te WS Ne a SAM 120 T TOS AC t MARCO i 5 E l E Pressure i 400 A Q 1000
162. ligned with the observation times since observations occurred at 0815 45 minutes Exact observation times are shown as triangles along the bottom of the graph and are given in Table 8 of Appendix A Motion Day of Trial 1 0 1 2 3 4 5 6 7 100 i 5 i i iv v 5 i i l l lu i S T4 o Oo l I T 8 Qu o g D i gt o 5 60 13 amp ES O a i o E o UNUM 1 5 5 3 Oo o 20 14 I 0 L 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes A Observation time Temperature From Monitor deg C Motion Closed 96 Motion Cycles cycles Figure 42 Downloaded Results Motion 139 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 140 Discussion Some peaks in motion cycle activity always correspond with the times that the cows were herded towards the yard for observations triangles There is a marked lack of activity in the hours just before dawn The fact that motion cycle peaks coincide predominately with the time of herding the animals toward the pens for daily inspection provides confidence that the motion cycle data is a good indication of general activity level In contrast the motion percent closed data needs to be compared with a comprehensive study of cow stance and behaviour over a number of hours before it can be considered useful No in depth obser
163. literature review section of this thesis lead to a detailed specification of what was required in a full system The need to demonstrate original results for this thesis was also a major consideration The system requirements are deemed to be Able to be located in the cow vagina for up to two weeks Be sized and shaped for comfort and retention Must withstand the physical environment Ingress of moisture mechanical stresses chemical attack from urine and faeces Precise delivery of 40mL of pharmaceutical gel vehicle for controlled release Easily design a volume rate profile on a PC and download it to the device Short or extended periods of delivery time from one day up to three weeks Delivery in accordance with a complex profile with a continuous range of possible rates between 1 mL day up to 40 mL day 80mL day peak Accuracy of 10 in real world conditions in vivo Safety limit overrides for temperature pressure motion and volume External control start stop override Development of a Practical Drug Delivery Communication and Monitoring System Machine readable ID number Human readable ID number Interface with outside world via digital data wireless network Two way 30mrange Transmit through any matter that might cling to the device Data rate sufficient to transfer a week s worth of data in five minutes Inter cow communication Environmental sensors for m
164. livery a drug was to administer it in a high dose by pill or injection at a given time Repeat doses followed when the initial effect started to wear off This is inconvenient uneconomic and sometimes results in damaging side effects Controlled drug release aims to overcome these issues by giving drugs continually for prolonged time periods in a controlled fashion and to the preferred site Rathbone and Gurny 2000 1 The primary method of 33 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 34 accomplishing this has been through incorporating the chemicals within polymers The advantages of controlled release are stated as having the potential to Useless drug Be more effective Have fewer side effects Increase patient compliance in humans One application for controlled drug release is to synchronise cattle for artificial insemination and reduce the time to breed In designing such products there are a number of factors to consider These range from raw effectiveness through to animal welfare and economics Influencing these factors is the ability of the underlying drug and delivery method to facilitate administration provide adequate absorption and bioavailability and to control the time and place of release 2 5 2 Advantages of Electronically Controlled Drug Delivery Electronically controlled drug release has the potential to ease delivery in a number of areas including chem
165. loped as it becomes clearer how drugs are transported across the skin barrier Magnusson Walters and Roberts 2001 50 205 227 Not all routes allow the transport and absorption of the chosen active biological material so this restricts the route choice significantly To address this issue iontophoresis electroporation and sonophoresis techniques are being developed These techniques are explained later on in this chapter starting on page 41 2 5 4 Commercial Products for Estrous Control Commercial products are based around the three main types of programmes Prostaglandins e g Lutalyse Estrumate IN SYNC Progestins e g Syncro Mate B Melengestrol Acetate CIDR PRID Gonadotropins e g Cystrolrelin Factrel Fertagyl Most programmes are designed to control the corpus luteum while more recent products control ovulation follicular development or a combination of these They work mainly by artificially controlling levels of hormones in the blood which act as messages to regulate the activity of the tissues involved in the reproductive cycle Rasby and Deutcher 2001 Each product requires its own protocol and often involves the use of one or more injections at specific times Controlled release products include P which is an intramuscular injection of microspheres to control the estrous cycle in heifers SYNCRO MATE B and Crestar are silicone subcutaneous ear implants for cattle CUE MATE PRID and CIDR also
166. ls of 5 minutes and integrated current accumulator is only logged every 15 minutes E External control via a wireless command overrides the internal programme telling the gascell to stay off After external control is relinquished by another wireless command the programme catches up to the F target again as quickly as possible G Progress continues as before H Again a wireless command overrides the internal programme Again it catches up as quickly as possible when the release command is sent J The target rate is zero from this point onwards so there is no further increase in volume dispensed 132 Figure 35 Flat Profile Target Integrated Current and Target Integrated Current Rate Experimentation 4 3 6 Gascell Voltage and Current Results Gascell Voltage and Current Day of Trial 1 0 1 2 3 4 5 6 7 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes Gascell Voltage V x Gascell Peak Current mA A At first the gascell is on a 100 duty cycle to make the syringe pressurise as quickly as possible The delivery programme has commenced Every 5 minutes the following decision is made If gas B production is below target the gascell is switched on else it is switched off C During the external control periods the gascell is never on Shortly after external control is released there is a short period of 100 duty cycle while the delivery D programme rapidly makes
167. luded that peaks of physical activity coincided with estrus in 75 of the cases Lewis and Newman 1984 67 146 152 Nevertheless motion sensing does not detect cows exhibiting silent heat as they display very little increase in activity The simplistic form of motion sensing uses a collar or leg mounted pedometer typically incorporating a mercury tilt switch to measure activity Step counting 23 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 24 by a leg mounted pedometer is quoted as being 15 more accurate than collar mounted pedometers The Westfalia pedometer transmits both the stored activity information and the cow s identification each time the cow enters the milking parlour The parlour computer then calculates an average activity level for the previous 10 days If the value for that day is greater than 7096 of the average an estrus prediction is output Afikim produces the Afitag pedometer which is a similar device while CowTrakker from Bou Matic is a combination passive ID and motion sensing estrus detector mounted on a collar The more sophisticated of these products use single or multi axis accelerometer MEMS devices Overall pedometery systems are moderately successful with up to 70 of estrual periods and 99 of nonestrual periods being predicted depending on the exact type of system and the conditions in which the animals live Koelsch Aneshansley and Butler 1994 58
168. ming Cable IC DMU Schemat Figure 54 183 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 164 Appendix C DMU PCB Layout Appendix C DMU PCB Layout Table 10 DMU PCB Specifications Item Description Six conductive layers Top and bottom signal layers inner layers 1 Layers and 2 power and ground planes Solder mask both sides no silk screen Substrate Core FR4 fibreglass 1 8mm Pre preg 1080 and 7628 grain Design Rules Minimum clearance 7 thou Minimum track width 7 thou Dimensions Circular 28 6mm diameter Fabrication Solder mask over bare copper with plated through hole vias and holes Hot air solder levelled finish Full net test Core 63mil Ts Prepreg 12 6mil Power Plane VDD gt p 4 Core 12 5mil MidLayerl gt E d n Prepreg 12 5mil MidLayer2 gt Core 12 6mil Ground Plane GND gt BottomLayer gt Figure 55 DMU PCB Layer Stack Up 185 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Progr am Run Mid Layer 1 Figure 57 DMU PCB Layout Digital Board Side 2 Power and Ground Planes 186 Appendix C DMU PCB Layout Side 1 Actual Size Power Plane Ground Plane ae Side 2 Actual Size oe e Wr 6 fe see D V TEN Af Ees Figure 59 DMU PCB Layout Transceiver and Batteries Board Side 2 Power and Ground Planes 187 Appendix
169. mity Self Diagnostics Battery Voltage V Battery Current mA Vdd V AVdd Vref V Power Supply Voltage Regulator Unregulated Voltage Regulator Shutdown Status Flags Any Reset Power Fail Reset Watch Dog Reset Judicious use of bit packing and a simple form of memory compression enable the storage of these items into 20 bytes of memory This allows 720 data sets 7 days worth of data to be stored before the data begins to be overwritten 97 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 9 98 starting with the oldest record This allows indefinite uninterrupted data logging as long as a download is accomplished once a week At any time a snapshot can be requested manually over the wireless link In that case a full set of measurements is immediately taken and the results transmitted The measuring process takes 0 6 seconds allowing manual capture of data to be observed at a much higher rate than the datalogging frequency 3 8 5 Event Scheduler The event scheduler is driven from the real time clock tick For example on every 5 tick it calls the gascell switch decision routine on every 15 tick it calls the data logging routine to measure and store the next set of results Other events are handled in the same way Current analogue to digital converter calibration TC1073 voltage regulator enable and ID transmission Environmenta
170. more effectively in the global marketplace Other advantages are increased marketing spending and customer priority According to the Boston Consulting Group this restructuring is predicted to form a giant vertically integrated dairy organisation capable of lifting dairy earnings by NZ 300 million annually Promar International 2001 Fonterra accounts for 7 of New Zealand s GDP Although significant by New Background Zealand standards Fonterra is only a mid sized player in a global market where scale is increasingly important US production is predicted to gradually shift towards the larger and more cost efficient units in the West and Southwest of the United States Such a shift is also likely in Europe and Canada These huge companies are substantially bigger than Fonterra and may prove more efficient due to larger economies of scale However with the right structure linked to the industry s new strategic plan projections are that in 10 years the New Zealand industry will be earning around five times present levels or NZ 40 billion To attain and maintain the higher herd sizes envisioned by an expanding dairy production base requires a dedication to advanced breeding technology There is a fixed supply of skilled labour in New Zealand that is only just keeping pace with current demand Developments in automated breeding technology will have to maintain pace if they are to enable those few with the specialised knowledge to
171. n implemented to make the best out of the given channel characteristics In the protocol designed for the DMU the ARQ process was not implemented due 89 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 90 to time constraints This means that the cycling packet number in the packet header is redundant but it can be used if automatic request repeat is added in the future There is still an acknowledgement from the receiver which can be checked by the user to see if a command was successfully received but the transmitter does not check for the absence of this to resend automatically Actual Implementation of Channel Encoding Due to time constraints only a simplified form of the entire channel encoding process was implemented This resulted in a reduced but acceptable range of up to 60 metres outdoors For a given range the more palatable bit patterns and greater redundancy for error detection would have increased the probability that decoded packets arrived in an error free or at least recoverable state The coding gain introduced by a particular channel encoding scheme is equal to the extra signal to noise ratio that would be required to provide the same raw bit error probability using the uncoded pattern at a given range The more sophisticated channel encoding scheme would have introduced a greater coding gain and hence range 3 7 5 LED Status Indicator As well as the wireless link t
172. nal milk production system relies predominantly on highly productive rotationally grazed pasture and herds of high genetic merit The favourable climatic conditions eliminate the need for indoor housing and expensive feed supplements This system enables farmers to produce milk substantially below the average world costs giving New Zealand its advantage over competitors worldwide On farm technologies including artificial insemination for genetic improvement newly improved pasture species and milk harvesting technologies have enabled New Zealand to remain a world leader in low cost milk production 2 1 4 Current New Zealand Dairy Efficiency New Zealand has a completely unsubsidised industry whose efficiency is commonly regarded as the highest in the world The New Zealand dairy industry provides 1 3 billion kg of relatively low cost product annually On farm cost of milk is estimated at NZ 2 80 per kg of milk solids compared to Australia where it is NZ 4 00 and Europe and the US where costs are at least NZ 6 50 In 1986 87 the average full time farm worker produced NZ 40 000 worth of farm produce In 1995 96 this had risen to NZ 55 000 of produce per worker This represents an increase of over 34 in real terms in labour productivity in agriculture compared with a productivity gain of 14 in real terms for the total New Zealand economy over that period While New Zealand dairy farmers are extremely efficient producers of milk their
173. natomy of the Cow Webster 1993 2 13 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 16 Estrus Occurring over a period of between 18 to 24 days 21 days on average numbered 0 through 20 the cycle starts with the release of follicle stimulating hormone from the pituitary gland at the base of the brain This triggers follicle growth in the ovaries over the next 12 to 36 hours 14 hours on average Waves of follicular development occur throughout the cycle following pulses of lutenising hormone which is required to make them active After two or three such waves a single follicle becomes dominant and ovulates Ovulation occurs after the end of standing heat and is defined as day zero in the cycle The duration of heat can range from 2 30 hours with an average of 15 hours This means that the state of ovulation can be predicted with reasonable accurately once estrus is detected This allows Artificial Insemination AI or natural insemination to proceed with the most probability of success shortly after this time A delay of several hours is recommended for AI since it is usually delivered directly into the uterus in contrast to naturally deposited semen which must transit the cervix Background granulosa celis antrum adult follicle odd ature degeneration follicle atresia Oestrogen Progesterone LT LEB 7 Corpus 77 we 7 AA luteum IL P Ly Yi the 77 Prosta
174. nch spacing Header replaced by SPST SMD DIP switch Connector made from 1 socket and 1 pin of a SIL strip Connector made from 1 socket and 3 pins of a SIL strip Connector made from 1 socket and 1 pin of a SIL strip Ferrite Bead Inductor chip Impedance 120 DCR OHM 0 8 MA 200 Ferrite Bead Inductor chip Impedance 120 DCR OHM 0 8 MA 200 Inductor 100nH 5 SMD 0805 package Inductor 56nH 5 SMD 0805 package Connector 1 pin SIL header pin strip single row straight 32 way circular cross section Connector 1 pin SIL header pin strip single row straight 32 way circular cross section GASCELL HEADER DIP1 SIL2 SIL4 SIL2 Ferrite Bead Ferrite Bead 100nH 56nH PIN Vdd PIN GND 11 5mm Diameter Button Cell CON SIL2 NARRO CON SIL2 CON SIL2 CON SIL4 CON SIL2 0603 0603 0805 0805 CON PIN1FP CON PIN1FP Manf Part Number gascell 11 6mm diameter 5 4mm height 150mL CHS 01B 250603121 7YOD 250603121 7YOD 0805CS 101XJBB 0805CS 560XJBB Manufacturer Supplier Simatec NIDEC Copal Electronics Fair Rite Fair Rite Coilcraft Coilcraft RS RS RS RS RS Allied Electronics Allied Electronics Coilcraft Coilcraft RS RS Supplier Order Code gascell 11 6mm diameter 5 4mm height 150mL 227 2639 352 4818 267 7416 227 2639
175. nd countries that wish to export there Establishing cow ID automatically is also advantageous if sampling milk from individual cows and compulsory when a robotic milker is used since there are no humans there to make the identification Once identified on line sensors pedometery devices and other sources of information can predict for cows on heat cows that are giving colostrum or cows having health problems The milk from these cows can then be routed to an alternative destination if required and the individual feeding and drafting actions carried out automatically for each cow There are many examples of useful applications in which it is not economically feasible to apply machine readable tagging for that purpose alone However if used for a range of activities the cost benefit threshold can be met Examples of these ancillary applications include being able to automatically prevent a mastitically identified cow from being accidentally milked in a conventional milking parlour Even activating an audible alarm would be enough to save a complete bulk tank being ruined by milk contaminated with somatic cells or antibiotic treatment from an mastitic cow Food Safety There are some issues with food safety for the consumer when any device is attached to an animal intended for human consumption Although RFID tags in small beads injected under the skin are less prone to tampering they can sometimes migrate of their own accord making t
176. ne enough time to store important data to non volatile memory and write an error log entry before the system is reset The first task for the power fail interrupt service routine is to turn off any peripherals The reduced current results in the reset delay extending to 5ms before hardware reset occurs 3 6 5 Watchdog There is a watchdog timer on board the microcontroller which can be used to recover from code crashes Important data such as unexpected reset events elapsed time and delivery profile progress markers can be saved to non volatile memory during the watchdog interrupt service routine before a reset is issued Watchdog resets are logged for diagnostic purposes Development of a Practical Drug Delivery Communication and Monitoring System 3 6 6 System Voltage Measure Battery voltage is logged for diagnostic purposes 3 6 7 System Current Measure Battery current is logged for diagnostic purposes 3 6 8 Voltage Regulators The 3 0V voltage regulators provide a stable supply voltage from the batteries as they discharge Normally the batteries are well above 3 0V but to obtain the longest possible battery life the regulators operate in drop out out mode when the batteries drop below this level supplying as much voltage as possible In retrospect it would have been prudent to power the RF board off a separate voltage regulator to reduce noise into the receiver from the switching in the digital board Fortunately
177. ns in a manual or semi automated fashion As herd size increases and labour becomes more expensive there will be a greater adoption of some of these technological aids Diskin and Sreenan 2000 40 481 491 Manual and automated methods for estrus detection both experimental and commercially available are listed below Manual Observation Estrus detector animals vasectomized or surgically altered bulls or androgenized females are allowed to roam among the herd Social grouping mounting and other behaviour is entered on a recording of observations and estrus expectancy chart Tail paint is the most common aid to the process of manual observation The 21 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 22 water soluble latex based paint is rubbed off when another cow mounts allowing the inference of mounting behaviour in the absence of direct observation Correct use of tail painting identifies almost 90 of cows on heat Dexcel Ltd 2001c Tail painting also picks up cows that are only on heat for a short time and that might otherwise be missed 1 e between milkings or overnight Kamar is a similar product Patches are glued to the coccygeal vertebrae of the tail head Red dye is released after a specified time and pressure of mounting Another variation is Bovine Beacon a light stick that emits light for 18 to 24 hours or more after the stick is broken by mounting pressure Altho
178. nting activity for estrus control via repeaters in 300m hops to the farmer s computer Integrated Estrus Detection No single method of estrus detection is 100 reliable Exhaustive visual observation ultra sound rectal palpation and blood sampling for hormone levels are the best methods in terms of detection and rate but are not practical for the average dairy production unit due to their significant cost You have to go a long way to beat tail paint backed up with a quick knowledgeable observation and a notebook To automate the estrus detection problem for production units with low skilled workers and a high cow worker ratio other methods must be employed Conclusion However there are no commercial products exist that have met wide spread adoption as being adequately accurate and economical Single sensor solutions are prone to error Temperature increases occur with estrus but other influences also affect temperature to such a degree that it makes temperature sensing by itself unreliable Tail head pressure sensing devices are susceptible to false triggering when the cow lies against a fence The use of motion sensing detectors is cow specific varied with footing conditions and does not detect silent heat Electrical resistance has a high variability between cows and is not reliable by itself In short no one method has been successful enough to make the wide spread use of estrus detection de
179. o penetrate thick flesh with little extra attenuation This method of communication has now been packaged neatly for rapid application to the general task of short range high data rate two way wireless communications in a product called LibertyLink from Aura Communications Aura Communications Inc 2001 This takes the form of small chip and antenna ready to be connected to an existing microcontroller Having negligible electric field component licensing by 75 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 76 electromagnetic spectrum management agencies is easy to obtain worldwide A variation on inductive coupling is a method termed oad shift keying where the implanted circuit gains power from an inductively coupled source The implant alters the impedance of its receiving coil which varies the voltage across the RF coil in the external power source data reader As expected this method has a range of only a few centimetres but has been successfully demonstrated in a patient based application where the hand grasp function was restored to a partial quadriplegic Tang et al 1995 42 524 528 Because the range of inductive and capacitance effects are not far enough these methods of communication were ruled out for this particular application although it is perhaps the best solution for a device situated in the rumen Electromagnetic Radio Frequency Waves Electromagnetic waves in the Radio Frequ
180. o be kept waiting in receive mode Instead the transceiver is placed in receive mode for 8ms out of every 600ms It is asleep for the remaining part of the receive polling cycle When DMU or base station wants to send a message it transmits a 800ms preamble before the first packet This tickle wakes up all DMU s in range and keeps them awake and in full time receive mode until they detect 2 5 seconds of continuous silence They then re enter sleep mode to continue receiver polling see Figure 11 The potential problem with this scheme is that internally generated receiver noise periodically generates a spike about once every second even with a moderate squelch threshold Other electromagnetic radiators in the area such as electric fences and cell phones can also cause spikes Noise spikes could keep the DMU permanently awake depleting the batteries within an hour or so The solution was to implement a noise spike rejection algorithm This was simply a matter of sampling the receive data line a few times in a row and making sure that the data stream consisted of zeroes and ones instead of just spikes With the noise spike rejection algorithm in place a transmitting digital cell phone did not keep the transceiver awake unless brought within 5cm of the DMU antenna Before noise spike rejection this distance had been up to 1m 3 7 4 Channel Encoding The wireless channel is a hostile environment Multi path interference and noi
181. ocket of wire wrap SIL socket pin strip single row straight 32 way circular cross section SOCKET103 Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section SOCKET104 Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section SOCKET105 Connector 1 socket of wire wrap SIL socket pin strip single row straight 32 way circular cross section 200K 270K 270K 100K not present not present not present not present RS 235 7639 Manual Reset SOCKET Vdd SOCKET GND SOCKET Battery SOCKET Battery SOCKET Gascell 0805 0603 0603 0603 0603 0603 0603 0603 TILT SWITCH SPST MOMEN TARY ACTION CON PIN1FP CON PIN1FP CON PIN1FP CON PIN1FP CON PIN1FP Manf Part Manufacturer Number MultiComp CRGO603 Neoohm Series CRG0603 Neoohm Series CRGO603 Neoohm Series 7914J 001 Bourns 000 Supplier Farnell RS RS RS RS Farnell RS RS RS RS RS Supplier Order Code 771 612 213 2597 213 2597 213 2531 235 7639 535 930 267 7444 267 7444 267 7444 267 7444 267 7444 Top mark Comment 204 271 271 104 Farnell Feb 2000 catalogue page 1715 price is for 1 socket of the 32 way break off strip price is for 1 socket of the 32 way
182. ocol are described more fully It is useful to examine these in closer detail for the purpose of assessing what kind of drug delivery capability is required of the forthcoming experimental system 37 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals CIDR The CIDR device from InterAg is a T shaped intravaginal insert made of silicone with a nylon core The silicone matrix outer is loaded with 1 38g of progesterone to prolong follicle development Elution of the drug follows the standard square root of time curve over nine days Manual injections of a estradiol benzoate by the trade name of CIDIROL are then required as in Figure 3 4 wu ESTROUS SYNCHRONY Cattle V4 TREATMENT PROGRAM SYNCHRONY A synchrony system should aim to precisely concentrate estrous onset while optimising fertility The time line below shows the latest recommended synchrony treatment for all classes of cattle Alternate programs using other hormonal products in different sequences may be marketed depending on the marketing authorisation in each country Day 0 Day 8 Day 9 Day 10 11 12 Administer CIDR emove CIDR Cattle Inject CIDIROL Inseminate at the fixed Cattle insert insert 0 75mg in maiden heifers ime of 24 hours after img in parcus cows CIDIROL injection or Inject CIDIROL nject prostaglandin inseminate to Img in maiden heifers detection on days 10 to 12 2mg in parous cows e Express
183. of conformity to level 1 under Schedule 1A of RFS 29 Specification for Restricted Radiation Radio Apparatus Radio Spectrum Management Department of the Ministry of Commerce 1998 Falling under the general license this means that no individual licensing is required from the end user s point of view In particular the DMU falls under level 1 77 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 78 consumer appliances not level 2 radio products because it is covered by the definition of low power apparatus To meet the licensing requirements an independent facility carries out tests in accordance with the regulations and prepares a report A declaration of conformity is lodged with the Ministry of Commerce and the applicant may be required to prove the conformity in a more rigorous manner if randomly selected as part of the quality control auditing process Complying with the Restricted Radiation Device format sets limits on frequency of use power output and duty cycle of transmission Matching the restrictions in Part 1 of Schedule 1A against desired technical performance of otherwise suitable devices defined the allowable operational limits Frequency 433 700 to 434 100 MHz Modulation Any Usage Telemetry Telecontrol Maximum Power output 1mW effective isotropic radiant power Many other combinations of operational limits would have been suitable but the best transcei
184. off override switched gascell off As per 29 05 2001 Download accomplished 9555 9364 programmed 0845 Override made inactive delivery profile Download accomplished As per Out for good Piston position inaccurate 30 05 2001 geis 10985 10794 102 26 0 30 0 66 programmed because DMU 3 fell out of the cow at 1800 the delivery profile previous day and may have been subject to rough treatment 174 mm to mL conversion factor mL mm 0 6667 Time at pressurisation min 191 Appendix A Additional Experimental Results Table 9 User Requesting Setup Information From DMU 4 Over Wireless Link a 1 User types command 2 Base Station confirms message is sent input gt 1104 3 DMU 4 replies with confirmation 4 DMU 4 replies with Setup Information command action output f output Transmitted message ii04 input 4 output Message from base station received 4 output gt Get setup info command initiated 4 output Replying on both comm port and transceiver Identification DMU ID 4 4 Gascell monitor serial number 0xB645 Gascell Management Active delivery profile is test5 DMU profile Recalibrate gascell current ADC period 40 minutes Gascell on off decision period 5 minutes Gascell safety check period 40 minutes Full gascell ICA register value 31 93 mAhr Integrated current to volume dispensed ratio 3 10 mAhr mL Cal
185. onitoring the animal and the device Temperature Pressure Motion Light Cow Proximity Data logging All environmental data above Drug delivery progress Device self diagnostics 15 minute resolution Sufficient capacity to hold a week s worth of results before result sets are overwritten starting with the oldest record 49 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 3 4 50 Snap shot of information available over wireless link A manually initiated process to get the current status of the device rather than having to wait until the end of the data logging period Overview of DMU Operation To satisfy these requirements the Delivery and Monitor Unit DMU is designed as a vaginally insertable device that allows a variable rate drug delivery profile to be carried out The profile is defined on a rate of volume versus time graph on a personal computer and is downloaded to the DMU before insertion Sensors monitor the state of the device and its environment and can inform a rule based intelligence to dynamically alter the rate of delivery The miniaturised electronics package occupies a volume of 16mL including batteries It includes a two way wireless data link for manual override device interrogation and data log download Data logging of 27 different variables occurs every 15 minutes for a two week period Once inserted regular transmissions f
186. ons The connection to the telephone network enables an exchange of information from the farm to a database located remotely Results are sent each day to the drug programme administrator via this link It also allows changes to the drug programme to occur without requiring a visit to the farm For example it might be that a trend develops in the results that are collected over the first week and the drug programme administrator wishes to change the rule based system present in the DMU within the cow This can be accomplished by the same two way link The secure and robust drug delivery process has many parallels with financial transaction processing networks used in such applications as automatic teller Conclusion machines In such cases reliable communication of authentication requested action and acknowledgement of completed tasks must be carried out with high reliability System reliability security and robustness would be key features of further development If we add more farms to the drug trial they can all communicate via the Internet to the sponsor s local area network for centralised recording feedback control and monitoring Site2 Ay mE b Site 1 aly Hed B i i Ste3 lt Internet 4 EM ay Es M Herd Station 7a Company Premises i o ps E E d Ld 5 p ae E E Ss Centralised Drug Program Design T ae Results Database and Status Monitoring E e Es mc a KES Company LAN
187. ontaining blood would be diverted from the bulk tank This is part of a continuing trend which will see more accurate physiological based testing carried out routinely at each milking and for each cow The accumulated results will be stored by sophisticated database software for greatly enhanced herd management and automated response activities 2 4 3 The Role Of Machine Readable ID Tags in Estrus Detection Definition There are many ways of uniquely identifying animals The printed or laser etched ear tag provides a human readable number which is easily discernable Laser printing is more durable Occasionally the ear tag gets caught on something and is accidentally ripped out If tagging is especially important the tag may be applied to both ears Incorporation of a bar code promotes automating identification but this still requires the presence of a human to position the scanning device Radio Frequency IDentification RFID lends itself to a more automated approach RFID works on the exchange of information between a reader and the tag using electric fields magnetic fields or both More sophisticated versions include the ability to store information on the tag for later retrieval The tag itself may exist in the form of an ear tag be mounted on a collar or strapped to the cow s leg A tag in an injectable glass pellet form is widely used to uniquely identify pets Active tags have batteries that provide them with a read
188. opment of a Practical Drug Delivery Communication and Monitoring System Although this code sounds as though it would require significant memory and processing requirements in practice it can be implemented as a finite state machine in just a few hundred gates This coding is 80 efficient Scrambling Scrambling is a technique which has a low overhead and a reasonably low processing requirement By scrambling the data it effectively randomises it Scrambling for the purposes of removing patterns in the source data is termed data whitening One technique is to use a pseudo random repeating sequence that modifies the data stream through the exclusive or operator The same process at the receiver recovers the original data assuming that the pseudo random sequence is identical and synchronised The other main approach uses a linear shift register A stream of patterned data even long strings of zeros or ones is transformed into what looks like a random bit pattern The inverse operation at the receiver recovers the original data There is a minute possibility that data to be coded will end up being scrambled into a long string of ones or a long string of zeros If that is the case the transmission fails but it will be encoded differently on the next re try The previous codes guaranteed a short minimum run length and DC balance The scrambling technique provides a long run approximation to these ideal characteristics
189. or determining the six calibration coefficients These are read and used by the software running on the microcontroller to correct for both the gain and offset errors of the pressure sensor temperature sensor and analogue to digital converter Second order compensation is used to get the most accuracy for temperature conversion 3 9 3 Motion Sensor There is no literature on motion sensors being located in the vagina but commercially available sensors described in the introduction use pedometers placed on legs and collars Tilt Switch TILT 5 6 RS 235 7639 35 Open when cow mounting A oscillates otherwise 8 7 Physical Dimensions in mm DMU Schematic RS Components Ltd 2000 Figure 18 Motion Sensor Physical and Schematic The most sophisticated of the commercial motion detectors utilise a MEMS multi axis accelerometer The motion detector used on the DMU is much simpler consisting of a ball in cage tilt switch S101 in Figure 18 There are two contacts that form the sides of the cage if the ball touches both sides of the cage the switch is closed and the input to the microcontroller goes high The input lines to the microcontroller are configured as standard input pins with no internal pull up resistor The full schematic is presented in Appendix B DMU Schematics 100 Development of a Practical Drug Delivery Communication and Monitoring System Being a non mercury mechanism it is completely
190. otherapy insulin delivery pain management antibiotic treatments and AIDS therapy This is because electronic control provides various abilities not otherwise easily obtainable The advantages of controlled release over uncontrolled release have just been described The finer control provided by electronic means will enhance these existing benefits Electronic control can enable an otherwise simple mechanism to exhibit an arbitrary delivery profile rather than being restricted to the standard exponential decay or square root of time responses To be sure non electronic alternatives can deviate from these classical delivery profiles Varying the thickness of dissolvable coatings on beads is one way to accomplish this However these techniques are somewhat dependant on Background the environment they operate in They are often affected by changes in temperature and pH etc In built intelligence coupled with electronic sensors make the device self aware These sensory inputs can affect program execution the result of which can be used to alter the delivery profile Delivery rate could be decreased proportionally with increase in temperature An emergency response bolus could be activated if heart rate is out of normal limits The existence of electronic circuitry can facilitate incorporation of additional electronic functions Once an electronic system is present there is relatively little effort in providing addit
191. otor or a piezoelectric diaphragm pump Both pump types have an inherent safe guard built in They only pump when a pulsed voltage of approximately the right frequency is applied This acts as a mechanical check in case the electronic driver circuitry fails in one particular state which is the most common failure mode There are many such commercial devices Medtronic Inc make the SynchroMed model which is percutaneously refillable and includes a programmer that can communicate and program the embedded device over an RF link Electrolytic Gas Powered Pumps The general method of propelling a liquid vehicle from electrolytic gas production is well established Many patents have been granted for variations on this method Bae and Kwon 1993 Bae and Kwon 1994 Gross and Zucker 1993 Herres 1994 Winsel 1993 Wyssmann 1998 Most of these mechanisms rely on a constant gas production to provide a relatively constant rate of drug delivery In such a system there is generally a battery to supply current and a fixed resistor connected in series with the gas generating cell The resistor primarily determines the current flow and hence rate of delivery is constant By selection of appropriate battery chemistry the variation in voltage during discharge does not unduly affect delivery rate An extension of the constant gas production technique can provide pulsed delivery This is achieved by making the drug reservoir a thin long tube The drug in l
192. oyment with a few minutes notice 3 5 3 Over Pressure Release Mechanism The blue cap seen in Figure 22 on page 110 backs off and vents the internal cavity of the syringe should differential pressure exceed 700 200 mbar If that fails to work the entire release cap holder pops at a differential of 1100 100 mbar 3 5 4 Safety Limit Checks If body temperature or animal activity were to deviate from normal during delivery it would be prudent to stop as soon as possible If a blockage occurred in the outlet of the delivery device and the over pressure release mechanisms failed this could have serious consequences If the blockage frees itself after some time the pent up pressure could deliver a bolus of drug that might prove dangerous If the blockage did not free itself and the syringe mechanically failed despite the two over pressure release mechanisms sharp plastic edges might harm the animal during rupture If the person designing the profile made a mistake and tried to deliver more payload than is physically possible the piston would press with increasing force against the end of the syringe at the end of the actual payload Again pressure would build up as if there was a blockage For these reasons the gascell decision check is modified to check for out of limit readings for temperature pressure activity and volume delivered see Figure 9 on page 60 If any of these factors are out of range delivery is stopped
193. perimental Results Summary sse 144 5 DISCUSSIQN A sepe us ecbepede bes dasoeue sexe sspe seus uewo bukege Pese S udES paE EH ERS EUER 145 5 1 Performance Assessment cee cece ee ceccccccccceccccececececececececececeeeeees 145 52 Defect Analysis and Suggested Remedies sssusse 146 D Critique of Design Process icin ehe ei eu deed dene deci 148 5 4 Suggested Enhancements to Current System s 148 6 CONGCEUSION 5 co seen cea saos euentu aed eu Va gene tees cues Pues su ee do uud ea sa suvute age dee augue 155 6 1 Future System Sene eec qeu eee huc LM MEL E 155 6 2 Extending the Control Loop to the Organism 159 6 3 Remote Data Gathering From Many Sites sss 163 64 Integrated Estrus Detection iiis ea seco Eten rare ida 166 6 5 COoric usion 32 o erede iter E eae ere tested 169 APPENDIX A ADDITIONAL EXPERIMENTAL RESULTS eese eee eee eee 173 APPENDIX B DMU SCHEMATICS 2 o eere bero tu pe oco e repo aepo rae oda e reos aero ape peR eua 177 APPENDIX C DMU PCB LAYOUT e eee ener enne tnt ente ttes eae en aca 185 APPENDIX D DMU BILL OF MATERIALS ee eee eere erento eonun 189 APPENDIX E BASE STATION DEVELOPMENT DMU ee eene enne nnne 199 APPENDIX F REFERENCES ix Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure
194. pin Sense resistor for measuring gascell current Part of low pass filter for gascell current accumulator measurements Facilitates gascell current measurements when bench testing Gascell load dissipates 100mW max Current limit for red LED Current limit for green LED Cost Ex GST NZ 0 05 0 05 0 05 0 00 0 05 0 05 0 05 0 05 0 05 0 00 0 05 1 11 0 05 0 05 0 05 0 05 0 05 193 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Designator R128 R129 R130 R131 R132 R133 R134 R135 R136 R201 R202 R203 R204 R205 R206 194 Description Light Dependent Resistor Miniature cadmium sulphide TO18 package Resistor 1M SMT Format 196 tol 0 1W Resistor 15K SMT Format 196 tol 0 1W Resistor Resistor ORO SMT Format 196 tol 0 1W Resistor 1M SMT Format 196 tol 0 1W Resistor 1M SMT Format 196 tol 0 1W Resistor 2R2 SMT 0805 Format 2 tol 0 1W Resistor 1M SMT Format 196 tol 0 1W Resistor 100R SMT Format 196 tol 0 1W Resistor ORO SMT Format 196 tol 0 1W Resistor 4K7 SMT Format 196 tol 0 1W Resistor 30K SMT 0805 Format 596 tol 0 1W Resistor 30K SMT 0805 Format 596 tol 0 1W Resistor Part Type LDR 1M 15K not present ORO 2R2 1M 100R ORO 4K7 30K
195. position from a datum using a steel rule as in Figure 27 Discussion As soon as external control was relinquished the closed loop system made up the error as quickly as possible This simplistic behaviour may be undesirable but 130 Experimentation demonstrates the basic closed loop control An alternative behaviour would be to suspend target volume growth while the override is active If it was important to keep the overall delivery programme on schedule a sliding window delivery algorithm could be easily added instead This would look at the difference in estimated and target volumes over a certain previous window period rather than all the way back to the start of delivery The last actual volume delivered point is the most in error because the DMU fell out of the animal shortly before this time It would have been subject to greater error due to cooling and the rough treatment it experienced when it was dislodged from the cow and lay in the field for a few hours When the DMU is not in the cow cooling of the hydrogen gas makes it contract and the piston is sucked backwards a small distance There is some decrease in accuracy compared to the bench test The loss in accuracy is attributable to the ever changing environment that the DMU is subjected to inside the cow This includes physical squeezing and other interactions produced by the cow as well as other more minor effects produced by the temperature and pressure of the day
196. producing pressure pump but there is less that 0 5mL error during the start of delivery Constant rate delivery till 480 minutes is very acceptable and the gradual decrease in target rate after that accommodates the fact that the syringe does not cope well with having to stop quickly The 24 hour period of zero delivery between 960 and 2400 minutes is designed to test the pressure leak compensation mechanism The small bumps in the pressure trace during this time are due to the gascell maintaining constant pressure by switching on occasionally At 2400 minutes the syringe is asked to start delivering at an extremely low rate after a long period of inactivity Hs External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 116 This should cause it some problems as the static friction on the piston is higher than the moving friction It sticks for a while but catches up again by 2880 minutes The remaining short fall is due to the fact that the leak compensation holding pressure while not delivering must be a certain amount below that required to make the piston move This is to make sure that no volume is delivered during this period Due to this a small amount of gas supposed to start the piston moving is actually used simply to increase the pressure again to that point This results in a small delivered volume deficit from that point onwards At 3360 the rate undergoes another step change this time to zero The syr
197. ptance of any one method Kyle et al 1998 found that the prediction of estrus based on vaginal temperature was excellent when an estral peak in vaginal temperature was defined as an increase of at least 0 4 C for 3 or 22 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 26 more consecutive hours detection sensitivity of 89 4 and prediction power positive of 85 7 Corresponding detection sensitivity and prediction power positive for visual observations of standing estrus were 53 2 and 96 2 respectively The mean maximal increase in vaginal temperature at estrus was 0 9 0 3 C Redden Kennedy et al 1993 found that vaginal temperature increased by 0 6 0 3 degree C at estrus and remained elevated by at least 0 3 C for 6 8 4 6 hours When increased vaginal temperature was used on an individual basis 17 of 21 estruses were detected with 3 false positives Although these results look promising by themselves temperature cannot be taken by itself as a reliable estrus indictor in isolation Body temperature is also affected by circadian rhythm ambient temperature and the health of the animal de Mol 2001 126 99 103 Lefcourt and Adams 1998 76 1830 1837 Acute mastitis for example also increases body temperature To complicate matters these effects have a different influence dependant on the site chosen to be monitored Fordham McCarthy and Rowlinson 1987 11 367 379 No commercial
198. r Code of Ethical Conduct for the Use of Animals for Teaching and Research and section 83 of the Animal Welfare Act 1999 4 2 3 Delivery Programme Two different profiles were employed Two of the devices were set to deliver at a Experimentation constant rate of 4 29 mL day delivering 30 0mL over 7 days The remaining two had the same variable rate profile used in the bench test in the previous chapter but scaled in rate and time to deliver over a week instead of three days The maximum rate in the variable rate profile was 10mL day also resulting in 30 0mL total delivery at the end of that time This variable rate profile can be seen in Figure 30 To ensure the devices were operating correctly they were pressurised and activated before insertion Approximately 3 mL of the vehicle was delivered prior to insertion to confirm correct operation Target Volume and Rate of Delivery Day of Trial 0 1 2 3 4 5 6 7 50 40 Volume 30 Volume mL Volume Rate mL Day YNSeOKRNON AD 10 Volume Rate 0 0 1440 2880 4320 5760 7200 8640 10080 Time Since Start of Delivery Minutes Figure 30 Variable Rate Profile Used In DMU s 1 and 2 4 2 4 Control Monitoring and Results Collection The program in the DMU took a reading from each of its environmental sensors every 15 minutes which was logged to memory The power supply and drug delivery readings were also logged at this time In theory it was
199. range of a few metres Passive tags do not require batteries but have a range of less than a metre They Background obtain their power from the coded interrogation pulse Anti theft labels in clothing and music stores use this technology Machine Readable ID Tags in Controlled Drug Release Machine readable ID tagging is an essential part of many automated herd management functions This tagging has many additional uses and it will become increasingly cost effective as the uses of ID tagging expand For this reason a machine readable tag should be included in any long term electronic drug delivery device e g rumen implant To prove this point an inter cow communication scheme using cow ID is added to the in vivo drug delivery device developed in later chapters In fact animal identification is important in many applications ranging from pet security to salmon migratory studies Many machine readable ID tag products exist for these markets Automated estrus detection can also make great use of animal ID the identity of the cow must be available with estrus information for it to be of any use Cost Effectiveness in Multiple Applications The CowTrakker product mentioned earlier is a combination passive ID and motion sensing estrus detector mounted on a collar It has been proven to detect when cows are in heat or are unwell The manufacturers of such devices state that they are cost effective even considering their hig
200. riginal analogue signal generated from an input transducer In a digital cell phone the analogue output from the microphone is translated into digital numbers before being transmitted It is assumed here that source coding and data compression have already been performed Each stage of channel encoding is now discussed in the order in which it is performed at the sending end The inverse of these operations is performed at the receiving end in reverse order to recover the original data The data is best treated as blocks of bits called packets By limiting the packet size the amount of information that needs to be resent when an unrecoverable error occurs is minimized If the packets are too small however there is unnecessary overhead in the percentage of bits used to manage the packetisation process Long messages are split up into numbered packets to be stitched together again at the other end The packet structure and encoding described below applies to each packet individually Cyclic Redundancy Check It is essential to know if the data has been received without corruption In the case of the Cyclic Redundancy Check CRC this is achieved by appending redundant bits to the end of the payload A linear feedback shift register implements a polynomial algorithm to generate a 16 bit field that is sent along with the data The polynomial used in the hardware CRC unit on the microcontroller in the DMU is the CCITT standard polynomial whic
201. rom each DMU are logged by other nearby DMU s providing a simple form of inter animal communication These transmissions are recorded allowing the pattern of social grouping to be determined Although the DMU s act autonomously a base station attached to a laptop allows manual override of the device and the downloading of logged data via wireless while in vivo A snapshot of the current state of the device is also available in case more immediate results are required An overview of the entire system is presented in Figure 5 Development of a Practical Drug Delivery Communication and Monitoring System Digital Data Wireless Network j J Uu ET Figure 5 System Overview Base Station ec The remaining discussion on DMU operation is structured in a similar way to the hierarchal DMU block diagram in Figure 6 51 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Digital Data Wireless Network Environmental Monitoring Temperature Communication Pressure Drug Delivery Control Motion Transceiver Light Wired Communication Drug Delivery ET Profile Cow Proximity Management Programming N and Debugging m Port Safety Limit Checks Message Over Pressure Handling Release Mechanism LED Status Indicator Drug Delivery Actuator Gascell Piston Gascell Load Control Gascell Voltage Measure Gascell Curren
202. roughout the day Pressure ranged between 1016 and 1028 mbar 4 3 2 Explanation of Graph Data The following graphs from the experiment are derived from data downloaded over the wireless link from DMU 3 DMU 3 had a constant 4 29mL day delivery profile Experimentation Figure 31 is provided purely for familiarisation with the generic graph layout which exhibits some commonality with all remaining graphs The trace in Figure 31 shows temperature purely for the purpose of providing a concrete example The actual temperature results are presented later Sample Graph for Familiarisation Temperature Day of Trial 1 0 1 2 3 4 5 6 7 40 38 36 34 i 32 30 o 28 gt 26 o 1 5 24 T 22 2 E 20 18 16 14 12 10 8 i 4 ii iii iv v 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes 191 minutes The DMU is turned on and the syringe begins to pressurise on the bench i The delivery program does not start until O Minutes ii 714 minutes DMU insertion into animal iii 3604 minutes DMU removed for inspection and replaced after 10 minutes iv 7904 minutes DMU removed for inspection and replaced after 10 minutes V 9874 minutes The DMU fell out of the cow Figure 31 Sample Graph for Familiarisation Each trace is made up of 720 data points sampled 15 minutes apart and joined with straight line segments This provides 7 days worth of data The ver
203. rs were important but hard to predict during initial design so they had been given a generous safety factor early on An hypothesis was that animal ID could be easily incorporated into a drug delivery device if basic electronics were already present Animal ID was added to the DMU with little effort 145 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 5 2 146 Defect Analysis and Suggested Remedies 5 2 1 Antenna Positioning and DMU Retention The environment is a potentially harsh one for the electronics It was certainly not in the original design plan that cows would want to chew on the antenna Fortunately chewing of the antenna was not harmful for either cow or DMU After one bite the novelty seemed to wear off More of a concern was the fact that the antenna occasionally slipped inside the vulva meaning that communication with the devices failed at that point This happened about 50 of the time and required a Dexcel technician to flick the antenna out before the data could be downloaded This problem would be easily rectified with a stiffer antenna possibly by attaching the antenna to the stiff plastic CIDR removal cord Alternatively the antenna could be tethered to the animal through the use of a Kamar patch described below The DMU which fell out was found to have been inadequately taped Separation of the syringe body and retention mechanism followed This problem is easily
204. rship New Zealand farmers own co operative processing companies By far the biggest of these is the newly created Fonterra company that accounts for 95 of dairy production 2 1 2 Exporting New Zealand with four million people does not have a large domestic market and is forced to export the bulk of its production 96 Because of this New Zealand accounts for over 30 of the total volume of internationally traded dairy products while it accounts for only 2 of the total world dairy production 2 1 3 Farms In 1999 New Zealand had a total of 14 400 dairy farms In the last twenty years the number of dairy farms has fallen but the average farm and herd size has increased Productivity per hectare and per cow have both been substantially improved The majority of New Zealand dairy farms are owned and operated by farming families In addition there is an established share milking system under 5 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals which a sharemilker is contracted to milk the herd and carry out a range of farm duties for a share of the milk income This system promotes trained eager young people into the industry and allows them to build equity before proceeding to farm ownership Most herds are milked in spring summer and autumn but dried off in winter when pasture production is lower The rest of the herds supply milk year round for the domestic liquid milk industry The seaso
205. s higher This can be taken account of by considering it as part of the integrated current to volume delivered ratio At 38 6 C for every 3 mAhr of integrated current produced there is about 1 mL of drug delivered The assumption here is that the gas leak rate is constant for different internal external differential pressures This is untrue but turns out to be a good enough approximation in practice A more accurate method would be to maintain another integrating current accumulator representing estimated total lost gas This leak compensation accumulator would be incremented by the full amount during the zero rate delivery period Le it would accumulate when the pressure is being topped up to keep it at the pressure just below which the piston would move as in the rudimentary scheme In addition it would be incremented to some extent during the non zero rates of delivery The actual pressure readings are read and stored every 15 minutes so this pressure reading could then be turned into an instantaneous leak rate prediction through a look up table Integration of this would result in a better estimate of the hydrogen gas leakage for any profile The leak compensation accumulator is subtracted from the gas produced integrated current accumulator to provide a more accurate estimate of amount delivered 61 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 62 This more complicated method was not
206. s one of only two part numbers to have a frequency that falls inside one of the regulatory New Zealand bandwidths This device uses amplitude shift keying or on off keying at a carrier of 433 92MHz On off keying was selected because the amplitude shift keying mode is only available at higher data rates which have correspondingly reduced range These transceivers have been used in products to track climbers on Mt Everest and runners in the Boston Marathon Amplifier Sequenced Hybrid Design Theory The Amplifier Sequenced Hybrid ASH design employed in the TR3000 contains 79 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals an amplifier time sequenced receiver section which provides over 90 dB of RF gain without the use of frequency down conversion Unlike a superheterodyne receiver which achieves stability by distributing RF gain over multiple frequencies the ASH transceiver distributes the total RF gain over time TX CN CN IN TRL1 TRLO VCC1 Pin2 VCC2 Pin 16 Power D GND1 Pin1 Control GND2 Pin 10 GND3 Pin 19 RREF Pin 11 CMPIN Pin 6 BBOUT Low Pass Peak Filter p 5 I Detector EL Crno LPFADJ 9 PKDET 4 Con Re J AGC Reset Pulse Generator amp RF Amp Bias za 0 rx LAE T AGCCAP 3 13 n 42 db ond GCC Pla Tubi Tani AGC Control Ron Row Ron ne Figure 13 Amplifier Sequenced Hybrid Transceiver Block Diagram RF Monolithics Inc
207. se contribute to degrade signal quality at the receiver The multi path problem lies in 87 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 82 the fact that the signal can take more than one path from transmitter to receiver Different paths have different path lengths so these signal components arrive out of phase with one another sometimes cancelling sometimes re enforcing The result is a sporadic fast fading attenuation that can generate erroneous bits in the data stream even when the average signal strength is relatively high Amplified semiconductor and thermal noise noise from man made or artificial sources and other transmitters in the area can also cause bit errors In the dairy farm environment there are many industrial electric motors for powering everything from vacuum pumps to drench mixers A rotary milking platform has a set of rotating glands with slip rings to transfer the power to the moving platform which is a another potential problem area Each electric fence and its associated controller is a very effective transmitter of electromagnetic radiation The wiring in the fence acts as an antenna extending a great distance around the farm Even though most of the interference is comparatively low in frequency the switching noise is audible on telephone circuits and radios in the vicinity Another consideration is that the reception process introduces a limitation in the patterns of dat
208. sis in Word for Windows and PDF format high resolution for printing thesis web_ page This thesis as a web page thesis refs References in Reference Manager 9 database format cad_file Schematic and PCB layout files in Protel 99 SE format data_sht Data sheets and application notes for all components used in the DMU results Results downloaded from each cow into Excel format This includes all raw data processed data graphs and explanations source The source code for the DMU and base station written in C for the M16C microcontroller using the Mitsubishi KNC30WA compiler 207 References References Aura Communications Inc LibertyLink Wireless Voice and Data ASIC CL 0040 N3 2001 Wilmington MA USA Aura Communications Inc Bae Y H and I C Kwon Gas pressure driven infusion system by hydrogel electrolysis WO 9307920 1993 Bae Y H and I C Kwon Gas pressure driven infusion system by hydrogel electrolysis US 5354264 1994 Beach R D F Kuster and F Moussy 1999 Subminiature Implantable Potentiostat and Modified Commercial Telemetry Device for Remote Glucose Monitoring IEEE Transactions of Instrumentation and Measurement 48 no 6 1239 1245 Bergen M T W N Tapp and S Reisman 1997 Telemetry Recovery Antenna Primate Body Temperature Bioengineering Conference 1990 Proceedings of the 1990 Sixteenth Annual Northeast 125 126 Berner B and S M Dinh 1998 Electronic
209. solved by correct taping A third band of tape in the middle is recommended in future for added security To help prevent against accidental loss of a DMU the antenna could be tethered to the tail head of the animal The Kamar heat detection patches were mentioned in the section on estrus detection page 22 and use a proven bio compatible glue to fix a capsule to the coccygeal vertebrae of the tail head Using this glue it would be possible to tether the free end of the DMU antenna The glue is strong enough to hold the weight of the DMU but would pull free in the event that the animal got caught in a fence Tethering by Kamar patch would keep the device from straying too far into the vagina keep the antenna in a position where it could receive and transmit properly Discussion and in the worst case retain the DMU should it escape the vagina 5 2 2 Moisture Resistance Being such a low power circuit there are many high impedance pins that are susceptible to moisture This problem was identified on the bench and an attempt at thwarting it by the use of an acrylic conformal coating was only partially successful This could have been the reason why some of the DMU s stopped working after insertion The next step would be to use a silicone conformal coating possibly with a Paralyne C moisture barrier Parylene C is a polymer which is used in aerospace and medical applications It is deposited in vapour form under vacu
210. sophy but some compromise was necessary due to time and resource constraints The DMU is not independent of delivery route but can be adapted to different drug types period of delivery and animal within limitations Many drug types that can be used in a vehicle of viscous fluid could be used Period of delivery can range from a few hours to several weeks The animal must necessarily be quite large due to the size of the DMU Perhaps it could be used in horses without modification although the retention mechanism might need modification Due to time constraints the bench testing was carried out within a narrow range of temperature and pressure It was assumed that temperature would remain within a degree of 38 6 C and that pressure would not change more 5 mbar during delivery During the field trial recorded temperature inside the cow stayed within the assumed range Likewise the pressure of the day stayed within the expected range however the external pressure experienced at the syringe outlet would have varied from this as it was subjected to movement within the vagina Suggested Enhancements to Current System 5 4 1 Further Testing and Field Trials To properly quantify the effects of changes in temperature and pressure during delivery a barometric chamber testing programme would be required This would establish the robustness of the system under the widest expected range of external pressures and temperatures If ano
211. strates the variable delivery rate ability 113 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Aeq qw a ey eunjo A pe1eAjeg jenjoy 4eq 1u ejes eunjo A OBIE Jequi eansseJd qu wnjo A pasanraq enjoy Tui ysew g 0 g xew yy m wnjo A 198161 5 se3nuiry A18Ad og JO eS e2uris eun Ovse 009 09 OCLE 0887 0Ov9c 0O0vc 09LC 0O0c6L 0894 Otri OOCL 096 024 08v Ove 0 Otz OL gi 0 wot 0k 0L 9 e euinjioA Pd g 62 7 9 0 T D v Ov f ere dd i ee z oS TN Lii if ft tb L Lt prritrit jH LLIILII WH tit Lt pruriit LLIILIE j piiit Lt LI LII ii LILILIE j lIi Lit LLIILIE L LIILILIE jt LILLLIL j i LLILILL d MOLD 0 09 4 z A 04 T 4 l 3 08 T E T 9 2 oet 4 T 8 et Me H T OT d T i E OLL T xer je p zi S oT AAA 1 vi lt tee e eee eee tene me eseseem eee eee eee eee en 2 Bey 5 T 94 opk 5 euin OA d gl Ost 091 OLL ost 061 ooz oLz v9 9S 8v ov ce ve 9L 8 0 SanoH As aatjag Jo J1e S BOUIS eun Jequi 00L O1 OZOL eJnssaJg PUAU Y 9 9 8E eunje1eduie Jaze M Buisf 81e 3 pue euin oA jeryoy pue 3eBje sa Youeg Bench Test ing Rw euinjoA Volume and Volume Rate Dur of Variable Rate Delivery 25 igure F 114 Development of a Practical Drug Delivery Communication and Monitoring System The actual volume delivered was obtained by placing the syringe on scales inside an oven maintained at
212. subishi Electric Corporation 200D dosi ecol ete e reatu e ete leid 96 Pressure Temperature Sensor Physical and Schematic 99 Motion Sensor Physical and Schematic sss 100 Light Sensor Physical and Schematic 45 5 uc dero du c tt 102 Base Station DMU Communication Flow ess 107 DMU Electronics Construction S etu qut pus tte Dato n ied d Ss 109 DMU Electronics Package Fitted to Modified Theratron Syringe Sectioned for Display and Shown Actual Size 110 Complete DMU Ready for Deployment sss 110 xi External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals xii Figure 24 Figure 25 Figure 26 Figure 27 Figure 28 Figure 29 Figure 30 Figure 31 Figure 32 Figure 33 Figure 34 Figure 35 Figure 36 Figure 37 Figure 38 Figure 39 Figure 40 Figure 41 Figure 42 Figure 43 Figure 44 Figure 45 Figure 46 Figure 47 Figure 48 Figure 49 Figure 50 Volume and Pressure During Bench Test of Variable Rate Delivery 113 Volume and Volume Rate During Bench Test of Variable Rate Deliver dedecore Ctt Ce a e Rr Ru RS RE Su UN 114 In vitro L and In vivo W Rate of Delivery Bunt et al 2001 28 116 Removal For Inspection and Piston Position Check 121 DIMI In ViVOQ 2i iin tkt hari IR Ing ER EHI A AAA XR EUER ERRASSE ERA ER
213. suspects increased risk of exposure or evidence of infection Conclusion 6 1 4 Extension of Cow Communication Ideas Extension of Social Grouping Propagation Algorithms A New Zealand patent is pending on the inter animal communication ideas developed during the course of this thesis Bunt and Cross 2000 NZ428552 The grouping information can be enhanced with a number of different techniques three of which are now described Broadcast One Hop Passive At regular intervals each animal asynchronously transmits its ID number to all those around it using a broadcast packet Those receptions are logged with the payload which contains the time and ID number of the sender Received signal strength is also logged for each packet When the animals are next in range of the base station or portable data gathering unit a central database is updated which provides a master record of all transmissions The base station has graphing capabilities to visualise which animals were grouped together Polling One Hop Interrogation As for Broadcast One Hop Passive but transmissions are addressed individually to each animal which has not been heard from recently within the last 20 minutes for example Receivers reply to the sender with their ID number If an interrogation reply is received they are not interrogated again for another 20 minutes The sender notes the replies that it gets in the standard manner The receiver logs the results of al
214. t System Current Measure Regulator Control Measure Executive Control Medos Power megraie Real Time Current Management Aaa Manual Override Unique ID Via Wireless MCU Mode q Control Retention Power Fail Event Scheduler Mechanism Warning CIDR System Voltage Data Logging and Measure Retrieval Figure 6 Block Diagram of Delivery and Monitor Unit DMU 52 Development of a Practical Drug Delivery Communication and Monitoring System 3 5 The function of each item in the block diagram is now discussed in detail For reference photos of the device start on page 109 Complete schematics are provided as Appendix B starting on page 177 A description of each component and its purpose is given in the DMU Bill of Materials in Appendix D starting on page 189 The PCB layout is provided in Appendix C starting on page 185 Drug Delivery Control 3 5 1 Drug Delivery Actuator Gascell Piston The gascell needs no battery in order to produce gas the act of drawing current is enough Assuming other factors are constant gas production is directly proportional to current drawn Winsel 1993 The electrolysis action produces a constant supply of gas behind the syringe piston which propels controlled delivery of the pharmaceutical vehicle Since th
215. ted detection of estrus and mastitis in dairy cows Tijdschr Diergeneeskd 126 no 4 99 103 de Mol R M G H Kroeze Achten J F M H K Maatje and W Rossing 1997 Results of a multivariate approach to automated oestrus and mastitis detection Livestock Production Science 48 219 227 de Mol R M and W Ouweltjes 2001 Detection model for mastitis in cows milked in an automatic milking system Prev Vet Med 49 no 1 2 71 82 de Mol R M and W E Woldt 2001 Application of fuzzy logic in automated cow status monitoring J Dairy Sci 84 no 2 400 410 Dexcel Ltd DRC Field Day Meeting the 4 Challenge 2001a Ruakura Hamilton New Zealand Dairying Research Corporation Limited References 21 22 23 24 25 26 25 28 29 30 31 32 Dexcel Ltd 2001b Farm Facts Heat Detection Available from http www dexcel co nz farmfacts view farmfact cfm id 1 9 Dexcel Ltd 2001c Farm Facts Tail Painting Available from http www dexcel co nz farmfacts view farmfact cfm id 1 10 Diskin M G and J M Sreenan 2000 Expression and detection of oestrus in cattle Reprod Nutr Dev 40 no 5 481 491 Edwards D F and R J Levin 1974 An electrical method of detecting the optimum time to inseminate cattle sheep and pigs Vet Rec 95 no 18 416 420 Ellison R S 2001 Diagnostic Tests The Future Society of Dairy Cattle Veterinarians NZVA Conference 2001 Publica
216. the estrous cycle can help in attaining increased efficiencies in the following ways 2 3 1 Increased Spread of Processing Demands Throughout the Year New Zealand dairy processors receive volumes of milk between 1 and 100 of peak milk flow In comparison with Northern Hemisphere processing throughput where processing throughput varies between 82 and 100 of the peak milk flow this is a large seasonal variation The global demand for milk products is spread more or less evenly throughout the year As a result the New Zealand dairy industry s processing facilities operate below optimum efficiency for large parts of the 12 month cycle In addition the requirement for increased storage to match supply with demand further adds to overall costs The ability to spread processing volume more evenly throughout the year to achieve greater efficiencies relies partly on a regionally co ordinated calving cycle Controlled drug release for the purposes of timing exactly when cows are inseminated throughout the season can pay dividends in this area if spreading were to be actively pursued Unfortunately being pasturally based there is a hard limit to which this spread of milk solids output can occur in New Zealand External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 10 2 3 2 Compact Calving Compact calving aims to maximise spring pasture and cow days in milk before pasture quality and quantity declines in s
217. ther field trial were to be carried out with the opportunity to study two more Discussion variables external syringe pressure and acceleration vector would be included in the data logging ability Logging of external syringe pressure would be used with the barometric chamber results to enable a clearer picture of how the in vivo environment affects delivery accuracy An algorithm could then be designed to maximize delivery accuracy under these conditions The use of a multi axis accelerometer would allow the logging of acceleration vectors This would be useful in inferring activity level mounting behaviour and posture By validating initial data with human observations it is hoped that this would enable more accurate heat detection and give an indication of stress levels that the animal feels in different circumstances 5 4 2 Simple Software Changes Alteration of the drug delivery profile is straightforward and can be accomplished without specialist training using a text editor Likewise the following parameters can be altered by modifying a single line of source code and recompiling Data logging period ID transmission interval Safety check pressure volume activity and volume limits Resolution of volume dispensed Integrated current to volume dispensed conversion factor Without further modification the system s best use in the medium term is purely as a tool for further research It can help in understand
218. they could contain dissimilar drugs or combinations of both types of layers Such a system would need careful evaluation in terms of shelf life and storage conditions due to the potential for mixing over time Add bolus capsules to support complex profiles Spare input output pads on the DMU printed circuit board IO1 and IO2 allow the connection of switches to trigger fast acting electrolytic bolus capsules Bolus shots in parallel with the main controlled release payload can easily be added A separate and smaller syringe or syringes can be glued inside or outside the body of the Theratron syringe A 10 w w solution of sodium bicarbonate and water is subjected to electrolysis to produce enough gas to deliver 3mL of drug over a period of six minutes This has been successfully tested in real world conditions Add or modify existing sensors Multi axis solid state accelerometers to allow sophisticated motion detection posture estimation path length by inertial navigation and inclination during mounting Add an external pressure sensor to take into account variations in ambient pressure within the vagina Global positioning system to track the individual paths of cows These receivers are now truly postage stamp sized Digital still camera A pill sized ingestible capsule which records colour video images as it travels the GI tract is currently undergoing clinical trials for capsule endoscopy Given Imaging Ltd 200
219. thics Supplier RS RS RS RS RS Mitsubishi Electric Australia Memec Power Technology Component s Ltd Intersema Sensoric SA Maxim RS Farnell Megtech Supplier Order Code 267 7444 267 7444 267 7444 267 7444 267 7444 M30624FGL GP DS2438 MAX4628 MS5535AM MAX6308 UKOOD3 T 246 7843 119 301 TR3000 Top mark Comment price is for 1 socket of the 32 way break off strip price is for 1 socket of the 32 way break off strip price is for 1 socket of the 32 way break off strip price is for 1 socket of the 32 way break off strip price is for 1 socket of the 32 way break off strip M30624F GLGP DS2438 AADN blank Do not exceed 225C Do not heat for longer than 30 seconds ABIQ Approximate price given OK GT F3FA Alternate supplier Insight Electronics Cost Ex GST NZ Function Gascell PCB 0 25 interconnect Rx Data PCB 0 25 interconnect Tx Data PCB 0 25 interconnect CNTRLO PCB 0 25 interconnect CNTRL1 PCB 0 25 interconnect Controls all functions 35 00 son DMU Integrates current from 5 11 gascell measures pre reg and Vdd voltage Digital 1 wire serial interface Switches low rate load 3 51 for gascell on and off Pressure and 63 18 temperature sensor Provides power fail and 3 00 reset signal when Vdd drops too low 3 0V regulator with low 4 34 1 1uA quiescent current 3 0
220. tical grid lines are equally spaced at 0815 each day They are approximately aligned with the observation times since observations occurred at 0815 45 minutes Exact 795 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 126 times are given in Table 8 of Appendix A Additional Experimental Results The roman numerals across the bottom of the time axis signify specific events during the experiment and are repeated on all subsequent graphs as readily identifiable reference points i At 191 minutes the DMU is turned on and the syringe begins to pressurise on the bench 191 minutes later at time 0 the drug delivery program commences but the device has not yet been inserted into the cow The DMU continues to deliver on the bench for another 12 hours During this time approximately 2 mL of vehicle was delivered from DMU 3 ii At 714 minutes after start of the delivery program DMU 3 is inserted into the animal This meant insertion occurred at 714 191 915 minutes after power on reset or 0900 on day zero iii At 3604 minutes the DMU is removed for inspection and replaced after 10 minutes iv At 9874 minutes the DMU is removed for inspection and replaced after 10 minutes V The DMU fell out of the cow of its own accord Experimentation 4 3 3 Temperature Results Temperature Day of Trial 4 0 1 2 3 4 5 6 7 40 38 j 1 papas nies c 36 J 34
221. tion No 211 41 45 Eveready Battery Company Inc Energizer NO EPX76 Battery 2001 FitSense Technology 2001 FitSense FS 1 Human Accelerometer RF Based Pedometer 2001 Available from http www fitsense com Flick B B and R Orglmeister 2000 A Portable Microsystem Based Telemetric Pressure and Temperature Measurement Unit JEEE Transactions on Biomedical Engineering 47 no 1 12 16 Fordham D P T T McCarthy and P Rowlinson 1987 An evaluation of milk temperature measurement for detecting oestrus in dairy cattle I Factors affecting measurement of milk temperature Vet Res Commun 11 no 4 367 379 Franaszek P A A Peter and A X Widmer Byte Oriented DC Balanced 0 4 8B 10B Partitioned Block Transmission Code International Business Machines Corporation 1984 Armonk NY USA Franaszek P A and A X Widmer 1983 A DC Balanced Partitioned Block 8B 10B Transmission Code JBM Journal of Research and Development 27 no 5 440 451 Fullwood Ltd Merlin Automatic Milking System and Herd Management 2001a Shropshire 2H External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 33 34 35 36 33 38 39 40 41 42 43 212 England Fullwood Ltd Fullwood Ltd 2001b Merlin Robotic Milking Machine Brochure 2001b Available from http www fullwood com merlinbrochure startmerlinbrochure htm Gartland P J Schiavo C E Hall R H
222. to assist in a self diagnosis role for the drug delivery hardware 141 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 4 3 3 8 Voltage V 3 3 Light varied greatly in vivo It is suggested that this is caused by the LED light being subject to total internal reflection from the syringe wall as the device moves around inside the animal If confirmed this might lead to a simple method of detecting what kind of material is in contact with the syringe near the LED be it gas flesh or a thick layer of clear or opaque mucus 4 3 11 Power Supply Voltages Results DMU Power Supply Voltages Day of Trial 1 0 1 2 3 4 5 6 7 ul cu Battery V y Voltage V Temperature C AVdd V vere w v I Mw vr Vv wr Vdd v 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes 60 10 40 90 Temperature C Battery voltage closely follows the temperature trace with an overall decline as time progresses Even the sharp decrease in temperature on removal for inspection on days 3 and 5 has been captured by the battery voltage level Both t he Vdd and AVdd regulated voltages remain constant as expected 142 Figure 44 DMU Power Supply Voltages Experimentation Discussion Starting at a no load battery voltage of 4 65V on installation on load battery voltage stabilised at 4 50V after ten hours of op
223. tory This means that this type of data for each individual cow is done only once or twice a year herd testing Somatic cell count and other factors are measured at each milking but only for the herd as a whole from a bulk tank sample To address these issues and improve the quality of life for dairy farmers robot milking systems have been developed In these systems the cows attend a robotic arm which places the cups on the cows teats automatically There are a few hundred of these systems operating worldwide mostly in Europe Some robots also carry out a range of other cow management activities The Merlin robotic milking system from Fullwood Fusion cleans milks and feeds the cows on an individual basis It also measures stores and displays data useful for estrus detection through the use of sophisticated herd management software 2 Sensortec has a conductivity sensor for Merlin in the final stages of pre production Somatic cell counting blood contamination and fat measuring devices are also being developed 29 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 30 Fullwood Ltd 2001b Fullwood Ltd 2001a Currently the machine makes predictions of certain events such as lameness and estrus Such a cow will find itself drafted to a separate holding pen for the farmer s attention Strongly abnormal results trigger the machine to carry out immediate actions by itself For example milk c
224. ts collection network could be developed This system would integrate these individual inserts into a tool capable of drug delivery and monitoring where behaviour of the device could be changed autonomously or externally via wireless control as delivery progresses This chapter documents the development and bench testing of such a system The following chapter describes its use in an experiment in vivo Contemporary Controlled Drug Release Design Philosophy Historically products have come about from animal need After an academic or market research study a perceived animal need leads to a drug choice and then to a route and method of delivering it But this approach is no longer good enough Instead a technology approach is required Delivery platforms capable of being designed in advance of a particular use can provide a wider range of function and hence applicability Rathbone 2000 As much as possible they should be independent of Delivery route e g intra vaginal subcutaneous ruminal Drugtype Required activation period Required range of flow rates Animal to which drug delivery is applied Only the various parameter settings should need to be tailored to match the technology with the specific application Development of a Practical Drug Delivery Communication and Monitoring System Additional desirable attributes are Biodegradable No effort required for removal No environmenta
225. tting the Program Run switch to run and pressing reset executes the downloaded program The same port outputs debugging statements if the debug option has been defined during conditional compilation The emulator also uses this port so if the emulator is resident the debug statements are sent to the wired comms port instead As a space and power saving measure no RS232 level conversion is performed on the DMU A special programming cable contains the level conversion IC which is powered from the DMU when connected to it The DMU programming cable schematic is given in Figure 54 on page 183 3 7 8 Message Handling Incoming messages from the transceiver are fed into a finite state machine Valid messages are passed to the message handling routine which calls the appropriate functions on behalf of the message The first task for any message is to reply to the base station with text to let the user know what command has been received and what action will be carried out 9 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 92 A help screen is available from the base station and is presented here to show the available commands Table 4 Base Station Help Screen N 00 V UDDER Base Station pO r Aio eN Ver 1 00 c 2001 InterAg Ltd E n output Format is command sub command from ID to ID output E g it05 or type for help input gt
226. ty to both transmit and receive the cows have the ability to communicate with one another over a short range 30m Current behavioural recording systems such as tail paint MountCount and HeatWatch rely on the physical interaction between cows in heat i e contact during mounting behaviour They miss cows that prefer to associate with others in estrus but who prefer not to be mounted themselves With a short range inter cow communication system it is possible to record the grouping of animals throughout the day This will detect all types of grouping behaviour including cows that prefer not to be mounted while in estrus More subtle behavioural patterns can be established by combining the data from the other on board sensors with the grouping data For example if a cow does not have an increase in vaginal temperature but begins socialising with the group of cows which are in heat and participating in mounting behaviour this would allow a study to target cows of that nature The absolute position of the animals cannot be determined from the data set nor is any individual data element a reliable indication of the range between animals involved in a data exchange However by analysing enough exchanges over time an accurate statistical representation of animal grouping will be discerned over a time period of several hours 103 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals The inter cow communic
227. uation Board PCB Layout m o 202 Appendix E Base Station Development DMU Power connector 32 768kHz and 10 0000MHz oscillators Programming port M16C microcontroller Modified M16C microcontroller evaluation board Light sensor Gascell switch Battery holder Gascell connector Gascell monitor Reset supervisor Reset switch Modified Transceiver evaluation board pe po EG pu 09000000 eocooccoco LT T 1 1 1 1 1 3 y n 0606000000 6 Nie Biebie el opti ONTO o LU i Inv A MSA0654 MEAUST Voltage regulators Pressure sensor RS232 level converter PC comms connector Auxiliary comms connector Figure 63 Base Station Development DMU 203 Table 12 Development Board Cross Reference Wiring Table External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Daughter Board Tilt Switch UARTO TX Uld 3ulod S L ojsjojo ele uonounj oBexoed asenbs uld NOW d9194v290 IN abeyoed auez uid NOW dd 194vz290 IN uonduoseg Uld NOIN uld 10 99UU0D NIG m aBeyoed asenbs uid NOW d9194v290 IN abeyoed albue o uid NOW dd 19dvz90 IN uonduoseg Uld NOW i uld 10 99UU0D NIG Lt oBexoed asenbs uld NOW d9194v290 IN abexoed oj6ue32e3 uid NOW dd 19dv290 IN uonduoseg uld NOW TT LED Green LED Red VReg Error 102 DI
228. ugh primary and secondary indicators of behavioural estrus are the best methods for determining timing of AI breeding to an activated Kamar device can yield acceptable results Nebel et al 1994 77 3185 There is little evidence for more frequent occurrence of estrus activity during nocturnal periods The most marked decrease in mounting and total sexual activities occur at milking and feeding times In spite of this proficient dairy farmers using manual observation with tail paint should be able to adequately detect estrus in non silent cycling cows even if they can t be physically present around the clock Pennington Albright and Callahan 1986 69 2925 2934 The limiting factor for the manual observer is the degree of skill required and the time available to observe If one or both of these are missing automated forms of estrus detection will prove more cost effective a Ultrasound Active follicles can be detected by ultrasound Previously these systems were difficult to use routinely in the field but now self contained head mounted units are available These are very effective but are used mainly by specialists due to the expense and level of training required Palpitation Palpitation via the rectum is a routine method for veterinarians to check the state of estrus by tactile detection of an active follicle This is recommended practice Background during the AI procedure Electronic Pressure Switches The
229. um The gaseous dimer molecules are split into monomers at temperature When allowed to contact the item at room temperature the gaseous monomers then polymerise to form a very tough clear impermeable membrane only microns thick 5 2 3 Battery Mounting The battery clip solder joints were prone to cracking where they met the PCB When cracked this had the effect of reducing the clip pressure onto the battery This sometimes resulted in unreliable battery contact and intermittent failure of the DMU The short term remedy is to re solder the battery clips after about five or so battery changes The longer term solution is to change the battery mounting and or battery type One possible replacement battery is the CR2032 which is a 3V lithium manganese dioxide battery The CR indicates that it can cope with high load pulse currents The Panasonic brand CR2032 has a 235mAhr capacity The nominal voltage is 3V so two would be required in series 5 The new taping procedure and use of the Kamar patch has proved successful in subsequent experiments with Sensortec 147 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 5 3 5 4 148 Critique of Design Process It was stated previously that contemporary controlled release systems should be designed in advance of a particular use to provide a wider range of function and applicability Rathbone 2000 Attention was paid to this design philo
230. ummer Each extra cow day in milk before mid summer is equivalent to an extra 1 5 kg of milk solids This focus on increased milk solids output is at odds with the need to spread production throughout the year so a compromise between these philosophies is necessary A tight calving pattern has been shown to be a common factor in successful farming systems in general Burke 1999 Major threats to achieving this are cows which have prolonged periods of anestrum late calving cows and poor estrus detection These potential problems and by definition the requirement for a tight calving pattern required to make use of this potential gain relies on controlled breeding with an effective controlled drug release programme 2 3 3 Increased Scale of Operation To counter the increased size and globalisation advantage of overseas companies the decision to merge two of New Zealand s two largest dairy companies has recently been completed Over the last 80 years the number of processing companies has fallen from over 400 to just seven Although three companies were responsible for 94 of production until recently now New Zealand Dairy Group and Kiwi Dairies along with the marketing resources of the New Zealand Dairy Board have merged to form a globally oriented dairy company named Fonterra Fonterra is the world s 9 biggest dairy processing company Attaining this scale of economies should lead to unit cost savings and the ability to compete
231. umptions about the direct applicability of any system to work equally as well in other farm animals however this should be avoided For example the method of detecting estrus in cows by measuring peak vaginal temperature works well Kyle Kennedy and Small 1998 49 1437 1449 but does not apply to sows Soede Hazeleger Broos and Kemp 1997 47 245 252 Any applicability to other animals would require careful consideration 6 1 2 Simplification for Commercial Use The DMU was designed purely as a research tool without much regard for cost per unit Various aspects of the DMU could be used as a starting point for development of a cost sensitive device incorporating limited functionality The gascell monitor IC could easily be replaced by incorporating its function into the microcontroller The integrating current analogue to digital converter would be replaced by the analogue to digital converter on the microcontroller A pressure sensor would not be required and the M16C microcontroller could be substituted for a much less expensive one running at a lower speed Most of the other peripherals would likewise be discarded The reduced hardware requirement would allow for a much bigger and cheaper battery in the same physical space There is a strong argument for eliminating electronics altogether Complexity should be reduced as mush as possible for cost and reliability reasons An open 155 External Control and Monitoring of Intravaginal
232. vations were attempted in this first experiment so it remains unknown how helpful the motion percent closed data is All motion stopped when the DMU fell out This proves that the motion sensor can assist in device self diagnostics Motion cycles has an arbitrary scaling for data compression which will be increased to provide better resolution in future now that an idea of what the upper limit of the motion cycles inside a cow is Experimentation 4 3 10 Light Intensity Results Light Intensity Day of Trial Temperature Light Intensity Temperature C 20 T T T T T 726 0 714 2154 3594 5034 6474 7914 9354 10794 Time Since Start of Delivery Minutes A The light reading takes a while to settle down after insertion When removed for inspection during daylight hours the light reading shows a marked increase as B expected After the DMU fell out the light readings varied less and were below that obtained during the C previous few hours The value is below that obtained when removed for inspection because it fell out during the night Figure 43 Downloaded Results Light Intensity Discussion Unfortunately a programming error meant that the LED was kept on while light readings were taken This explains why the light readings are unexpectedly high when inside the animal Even so the light intensity readings are of some value and indicate that a simple light sensor can be used
233. ver judged on technical merit is only available at set frequencies so this factor defined exactly how the regulations applied to the rest of the parameters 3 7 3 Transceiver Hybrid Transceiver Selection The TR3000 amplifier sequenced hybrid transceiver IC from RF Monolithics was chosen as the hardware to carry out the transceiver task This is designated U201 and shown by itself in Figure 12 immediately below The full transceiver board incorporating the transceiver hybrid is shown in Figure 53 on page 182 The TR3000 was the only device that complied with the Restricted Radiation Device regulations fitted in the restricted physical space and was integrated Development of a Practical Drug Delivery Communication and Monitoring System enough to easily form part of a circuit with the minimum amount of components It also meets the minimum requirements for data rate power consumption availability of technical support and ease of development CNTRLO p CNTRLI 028 J 4 oor oor 7 06 Qo 1 80 PWIDTH PRATE 002 051 Oo 3 m TXMOD THLDI ty THLD2 L E 040 m 102 l TR3000 d 0 04 1 02 Physical Dimensions in inches and mm DMU Schematic RF Monolithics Inc 2001 Figure 12 Transceiver Hybrid Physical and Schematic Being of fixed carrier frequency the part number determines the frequency of operation The TR3000 i
234. very Time Since Start of Delivery Hours 0 8 16 24 32 40 48 56 64 Volume mL a Volume Rate mL Day i i i i i i t i t 4 30 20 Volume Rate 10 0 10 240 0 240 480 720 960 1200 1440 1680 1920 2160 2400 2640 2880 3120 3360 3600 3840 Time Since Start of Delivery Minutes Figure 8 Sample Delivery Profile Defined on PC for Download into DMU The volume rate trace in Figure 8 is a typical target rate profile and would be entered in the following format 0 0 1 30 480 30 960 0 2400 0 2880 30 3360 30 3361 0 INFINITY MINUTES 0 55 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 56 The list of tuples represent the vertices where the first number in the bracket is the time in minutes and the second number is the rate in units of mL Day During system initialisation the unique ID is used to determine which profile graph in flash memory should be pointed to for subsequent access Routines are provided to integrate the rate profile between the current time and the start of the delivery resulting in the instantaneous target volume to be delivered Start of Delivery Assuming that all DMU s are reset and installed in syringes at the same time the user can choose how delivery will commence By relying on a set and forget trigger pressure to activate the delivery for each syringe the syringes will deliver th
235. very in farm animals is possible and that future systems may extend the control feedback loop outside the drug delivery device the animal itself and perhaps even the farm at which the animal resides Thesis Objectives Drug delivery is a huge field Even within the veterinary field of application controlled drug release is a major focus of journals regulatory bodies discussion groups and an even larger number of commercial products This introduction establishes a need for pursuing the thesis objective of investigating external control and monitoring of intravaginal drug delivery in farm animals A significant commercial application of the intravaginal route in dairy cows is for estrus control Estrus detection and control are covered in detail because those topics motivate the development of experimental objectives in later chapters This scope provides an active topic large enough by itself while allowing a focus on specific issues of application Reference is made to the more general problem of drug delivery in humans and animals where appropriate Current demands from animal scientists include the need to deliver multiple drugs at different rates and amounts over extended time periods Such demands are difficult to meet using conventional polymeric formulations The use of electronically controlled drug delivery systems provides the pharmaceutical scientist with the opportunity to achieve these demands however much fundamental an
236. vices economically viable There is a gap in the market for an integrated form of estrus detector with an integrated package of sensors that can look at all the indicators to produce a more reliable result The challenge is one of cost manufactures must be able to convince dairy farmers that these systems will pay for themselves in terms of initial outlay and maintenance The attributes of a successful future system would be Integrated animal ID temperature conductivity motion and pressure switch mounting sensors Intelligent enough to process the combined data for that cow Low cost low maintenance Other attributes would be desirable but might not be worthwhile in the cost benefit stakes This includes the ability to communicate to the remote 167 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 168 database for herd management To sense mounting the device must be located at the tail head To sense mucus conductivity probes must be inserted into the vagina A system could be attached to the tail HeatWatch style with probes lodged in the vagina There are many practical difficulties however not the least of which is the attaching line being snagged or pulled out by other curious cows A short range radio link would work and has worked well to solve a similar problem for a human pedometer system called FitSense FS 1 This system has a shoe watch wireless link and uses an accelerometer
237. vivo for a distance of 7m Beach Kuster and Moussy 1999 48 1239 1245 This experiment used a digital form of transmission known as on off keying A frequency of 4MHz was used successfully in an implantable intracorporal pressure and temperature application in humans with dimensions of only 6x28x2mm Range was only 5mm due mostly to the restricted space available for the antenna Flick and Orglmeister 2000 47 12 16 A frequency of 86 to 88 MHz was used in an implantable transmitter for the purposes of subcutaneously monitoring glucose in dogs Shults et al 1994 41 937 942 This worked well when implanted in thoracic subcutaneous tissue Both frequency and amplitude modulation were trailed with amplitude modulation giving the best range of 10m RF is the only proven method of communication established in the literature for data transmission from within body cavities with a range in excess of 5m and so was chosen for this application 3 7 2 Licensing There are restrictions in New Zealand and other countries on the use of the electromagnetic spectrum with the aim of avoiding interference with other users This is achieved by licensing which is a potentially complicated and expensive process To alleviate this the DMU is designed to comply with the description for a Restricted Radiation Device as defined by the Radio Spectrum Management Department of the Ministry of Commerce in New Zealand These devices require a declaration
238. w is proportional to current flow allowing a constant or varying delivery profile Zhao Hung and Choy 1997 23 Suppl 1 S27 S29 The resistance of skin varies greatly Sensation of current flow can be felt by most humans at around a threshold of 150uA cm however skin can begin to be damaged at a level ten times lower than this A direct current source has higher efficiency but has some side effects The pulsed direct current technique has little side effects but the efficiency is lower 4 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals 42 Examples of commercial iontophoresis drug delivery systems are the Phoresor from Iomed Salt Lake City USA and the DUPEL from Empi St Paul USA Electroporation Also called electropermeabilisation electroporation is when the cell membrane permeability and electrical conductance drops due an applied voltage Berner and Dinh 1998 When a short pulse is suddenly applied to the membrane conductance is relatively low allowing it to charge up as the electrical field is established The current is then relatively stable for a short time period on the order of one microsecond Breakdown then occurs and the membrane becomes unstable During this time the membrane is permeable to substances it would otherwise present a barrier to In some cases the action causes permanent damage but in most cells and tissues it is reversible if short pulses of about
239. xhibiting silent heat ovulation occurs without bulling or coming on heat Since the increase in temperature is not present it is interesting to speculate whether the drop in temperate due to ovulation might be used to indicate silent heat instead The answer is probably no since even with a highly accurate thermometer the small temperature change of less than one degree would be masked by other factors which also cause fluctuation in body temperature These other factors include circadian rhythm ambient temperature and the health status of the animal Even if the slight variation could be used to pinpoint ovulation it would be of little use At this late stage there would be too little time to arrange for artificial insemination to coincide with the viable egg Remote monitoring of vaginal temperature in cows via a radio link has been accomplished more than one once Kyle Kennedy and Small 1998 49 1437 1449 Redden Kennedy Ingalls and Gilson 1993 76 713 721 and has also been applied to humans McCreesh and Evans 1994 904 905 Remote temperature sensing via radio from other parts of the body has also been accomplished many times e g in poultry Kettlewell Mitchell and Meeks 1997 17 161 175 primates Bergen Tapp and Reisman 1997 125 126 and cows Lefcourt et al 1993 54 798 804 Lefcourt and Adams 1998 76 1830 1837 Methods utilising temperature to improve detection of estrus in dairy cows have been without wide acce
240. ys have the highest priority Figure 9 is a simplified flowchart The actual decision making process is complicated by the routines for managing the start of programmed delivery and implementing various diagnostics modes 59 External Control and Monitoring of Intravaginal Drug Delivery in Farm Animals Gascell override active Estimated volume dispensed target volume Gascell override value on Safety check pressure temperature volume dispensed and activity level within limits Yes Gascell switch on Gascell switch off End Figure 9 DMU Gascell Switch Decision 60 Development of a Practical Drug Delivery Communication and Monitoring System Pressure Leak Compensation Hydrogen gas molecules are extremely small and given time can even penetrate thick glass It is next to impossible to stop the hydrogen gas leaking slowly from the O ring sealed plastic syringe Escaping hydrogen gas would lead to a shortfall of delivered drug volume To counter this a rudimentary form of pressure leak compensation is employed If the pressure drops too low during periods of zero rate delivery the gascell is turned on until the pressure increases to the value just short of moving the piston During this compensation behaviour the integrating current accumulator is not incremented Unfortunately the syringe leaks at an even greater rate at non zero rates of delivery because the pressure i

Thesis - Peter Cross - screen viewing

Contents

Download Pdf Manuals

Related Search

Related Contents