Home

Courage Chibiya BSc Computer Science (Industry)

1. Distribution Parameters Id Marker Distribution Equation Weight Equation Normalisation Parameters Id Marker Normalisation Equation Risk Factors Id Name Disorder Factor The design of the database has 8 relation variables tables User Pregnancy Dating Markers Disorders Marker Set Distribution Parameters Normalisation Parameters and Risk Factors The bold underlined fields are primary candidate keys and the non bold underlined fields are foreign keys A primary key 1s a unique record identifier and a foreign key points to a primary key of another database table Foreign keys in this database will be governed by integrity constraints The integrity constraints 33 are database rules that are executed on record update to maintain referential which means that a foreign key in a given table can only contain values in the primary key column of the referenced table 6 5 Development Plan The solution will be developed using a mixed evolutionary and waterfall model When the requirements for a given component are clear and stable that component is implemented using a Waterfall based model When requirements are not clear or unspecified an exploratory approach will be used to develop a prototype and user feedback will be used to improve the component until it 1s satisfactory The user interface will be developed using an evolutionary approach A skeleton graphical user interface will be initially developed to ge
2. 8 2 Testing Results The primary objective of the code testing process was to uncover as many errors as possible and fix them before the acceptance testing was done White box testing enabled the developer to uncover and correct most logic or program structure based errors As described by Hetzel 1988 white box tests are typically applied to small program components and in complement black box testing validates the functionality provided by the system All errors discovered during unit testing were resolved before integration testing was initiated System testing test cases evaluated the functionality of the system and its performance The evaluation chapter explains in detail the results of testing the final solution After completing code testing user acceptance testing was conducted by Mrs C Wilson MI Project Supervisor The system met acceptance standards in almost all cases The one requirement that the system did not meet was the batch calculation functionality but in this case both the developer and the user agreed that providing this function was infeasible under the current project constraints Therefore both the developer and Media Innovations Ltd agreed that the system had passed the acceptance testing As experienced developers say testing never ends it just gets transferred from the developer to the user and every time they use it they will be performing a test Klein 2005 so in essence our aim was to design test case
3. Analysability Replaceability Changeability Stability Testability Maintainabili ty Figure 15 ISO Software Evaluation Criteria The evaluation criteria for the solution are Functionality How well does the system satisfy user requirements aims and objectives Reliability The results of testing will be used to demonstrate the accuracy of functionality fault tolerance and recovery Efficiency The performance and speed of response of the final solution Usability and Accessibility Usability is the effectiveness efficiency and satisfaction with which users achieve specified goals Accessibility is the opportunity for everyone regardless of technical or physical inclination such as colour blind people to access software products 43 Portability and Maintainability Portability is the relative ease with which the system can be installed across platforms Maintainability 1s the ease of making modifications and additions to a software system User evaluation results In addition the solution was compared to existing systems and subjected to user acceptance testing and user evaluation was used to evaluate the system 9 2 Functionality Functionality refers to the services provided by the system to satisfy user requirements The solution s requirements were specified through project aims objectives and requirements specifications It is logical to assume that the success of the system is judged on how well and t
4. yr end if if letter decimalSep num fDhecS5epocr brrend if it num fDecS5ep 1 validMumber false l end if Py fend far if validMumber false Control focus alert Value must be a valid number gt 0 return else EnteredS5tring Entered5tring replace decimalSep 7 Entered tring parseFloat Entered5tring T Control value Enteredstring replace decimalSep fend if l end function validate numbers 87 Reporting System The reporting system uses a fillable PDF form to create a patient report Figure 1 shows the form in design view The program reads in this file into memory and fills the fields of the template with the current patient s results ose PrenatalLinkfyrteening Calculator PaticewiiAporte M murs P ate usai atio Tibtsurname t urrame Be stala EI 20068 B x DatingiGestorv EN lblForenameL Mitirorename Ba B etusg Sereen E E ae i Xm sli xtherDetailz RUNE 4 SS REESE xtotherDetails Burun Tune AblMoMHeadinaL KAIRI Er ert biom bee BL m rm IbIMarker Btn oni PE xtGest tT inbiat xtTvvinGest1 A BR NU a xtGest2 xt Tino 2 B IU NES BR 2 xtGest3 MUN EE RIPE DE E INST a xtGest pt TeinMoM4z B Test rvinGesta lbivarker E mE s xtGest5 AxtTwinMoMS Bl Jet T winGest5 IblMarkert xt ome BE xtioestb xtTwvinhlohB xtTwvinGestb Theo B ns xtGest tat TwinMoM7 b TeinGest blResu Suits sHeadi adir Ha 1 1H EINE jbiFina
5. AN Value must be a valid number 0 If an invalid number or date is entered an error message will be displayed describing the cause of the error and the corresponding field with an invalid value will be refocused to enable you to correct the entered value 113 25 3 Calculating results Instructions for calculating results e In the data entry screen form complete the necessary fields in each section as explained in the data entry section below e Select the disorders for which a result is required e Click Calculate e From the results screen click Print if required 25 4 Data Entry Sections 25 4 1 Patient Information Surname Forename Id These fields are optional but appear on the report DOB Required Ethnic Optional Serum markers can have ethnic specific median corrections a Enter the patient s surname forename and id in the appropriate fields Patient surname Forename Id DOB v Ethnic Default v b Use the data picker to select the patient s date of birth c Select the ethnicity of the patient from the ethnic drop down list Pregnancy 25 4 2 Pregnancy Dating Number of Fetuses Number of Fetuses If twin pregnancy select this first M onazygnus Dating Type IRE Dating Type Required Select the type of measurement to be used to calculate gestation Gestation means that the Measurement calculated gestation will be entered in the Measurement field Dating Date Dating Da
6. UR SOU2 UR SOU3 UR USEI UR USE2 UR USE3 UR USE4 Test Cases 102 Requirement Test Cases UR USES5 UR USE6 UR USE7 UR VIEI UR VIE2 UR VIE3 UR VIEA UR VIES UR REPI 20 5 Testing Results 103 21 Appendix F Evaluation 21 1 Media Innovations Ltd Project Evaluation 21 1 1 Project Elipse PrenatalLink Beta test version 21 1 2 Project summary Elipse PrenatalLink is the latest item in the Elipse suite of products It is a web based version of Elipse using the same calculation engine DLLs as the Win32 version that is in use worldwide The aim of Courage s project was to design and build a Web application for the existing calculation engine developing it to the Beta test stage with the intention that it would not require much modification for commercial release The Elipse PrenatalLink project is a commercial development and Courage was required to work to professional standards in all areas of his work This included e Taking primary responsibility for system design Adhering to the user requirements Adhering to Media Innovations software development standards Producing high quality fault free code Producing documentation in accordance with Media Innovations systems Working to strict deadlines 21 1 3 Product specification The following functionality was required e Asecurity and login system that prevents unauthorised access and allows customisation Data entry and validation Calculation and p
7. University supervisor to ensure such a problem is discovered early in the term During unit testing time will be allocated for checking that the system meets the legal requirements Intellectual property conflict Loss of main product litigation etc Plan to discuss problems with the Media Innovations project supervisor The test plan will have to be scaled down Only the major units would be tested Improve communication so that goals are understood by all stakeholders The developer is prepared to spend a bit of their own money to support the project if necessary Another project will have to be chosen to meet the needs of the University Corrections will have to be done to ensure the system works correctly and that data 1s secure Action gt Risk gt Effects Low Customer satisfaction levels Copyrights and patents might be used to protect the system This can provide bad publicity seriously damage the business s reputation in the market and contribute to losing customers Moderate Reputation The system must be highly usable and efficient 64 Good usability and appearance will improve customer satisfaction If the main server fails then the website would still be accessible gt Effects Anything stored on the physical server at offices will be inaccessible and customer service would suffer as staff will not be able to run website Moderate High Moderate Technical gt Action Backup
8. rendering it unsuitable for the purposes of this system Jscript is a Microsoft implementation of the ECMA script an international standard scripting specification Jscript is a dynamic object oriented language similar to Java Its object framework is very similar to that of JavaScript JavaScript is a Netscape client side scripting language It is the de facto industry standard for scripting It is object based and certainly it 1s superior at transforming a lifeless page into a fully interactive page with real time response The availability of control flow like case statements gives JavaScript a considerable advantage It also provides a comprehensive assortment of built in functions for formatting mathematics and many other mundane programming functions The advantages and versatility of JavaScript led to it being chosen as the client side scripting language of the final solution JavaScript is highly documented and can run on most browsers making it a practically viable choice 26 6 Design This chapter describes the conversion of the system requirements into design specifications that define the overall system architecture It models the system components modules and data required to satisfy requirements from an implementation perspective Before system design commenced a design pattern was chosen based on suitability and fitness for purpose to provide a guiding framework in which the final solution was designed 6 1 Design
9. Chichester 1990 pp 111 17 28 Microsoft 2005 Overview of Web Application Security Threats http msdn microsoft com library security 29 MSDN 2006 MSDN home http msdn microsoft com netframework gettingstarted default aspx 30 Oncology 1999 Molecular Biology of Neuroblastoma Journal of Clinical Oncology Vol 17 Issue 7 July American Society for Clinical Oncology 52 31 Pressman R S 1992 Software Enginerring A Practitioner s Approach third edition McGraw Hill New York 1992 pp 66 32 Python 2006 Python home http www python org doc faq 33 Royce 1987 Managing the Development of Large Software Systems IEEE Computer Society 1987 Vol 9 13 37 34 Schaum 2001 Schaum s Outline of UML McGraw Hill ISBN 0077096738 35 Schwartz 1997 Learning Perl A Nutshell Handbook Publisher O Reilly ISBN 1565922840 36 Scott 2005 Wal Mart chief executive Wal Mart to expand health cover http news bbc co uk 2 hi business 4745464 stm 37 Shaffig 2004 Elipse Homepage http www elipse org uk 38 Smith Bindman 2001 Second trimester ultrasound to detect foetuses with Down syndrome pg 1044 1055 39 Somerville 2004 Software Engineering Pearson and Addison Wesley 7th edition 2004 40 Stellman and Greene 2005 Applied Software Project Management O Reilly Media ISBN 0596009488 pp 44 49 41 Stepanek 2005 Why Software Projects Fail ISBN 1590595
10. VORD Viewpoint oriented requirements definition is a service oriented method for requirements elicitation It is concerned with identifying different stakeholders or system viewpoints such as users data sources or any other receivers of services Identifying the requirements of these viewpoints enables system requirements to be easily gathered VORD is particularly valuable in discovering requirements conflicts Easterbrook and Nuseibeh 1996 A key strength of VORD is that it identifies the diversity of requirements sources and therefore imparts a sense of completeness However it is less appropriate for structuring requirements as there is no connection between some internal system viewpoints and system end users Interviews of key stakeholders are crucial to the requirements gathering process Structured interviews involve asking prepared question to stakeholders of the system to find out what they require in the new system Pressman R S 1992 Interviews are a straightforward means of gathering requirements The specifications produced are more acceptable to users since they are based on their 18 opinions However considerable time 1s required to prepare for the interview and study background information extensively Kotonya 1998 After considering the advantages and drawbacks of the requirements elicitation techniques discussed above the developer decided to use stakeholder interviews and VORD for gathering requirements The s
11. 24 Mar 2006 08 Apr 2006 Appendix and Glossary Compilation 16 Apr 2006 20 Apr 2006 Reflection and Project Review 15 Mar 2006 21 Mar 2006 Table 3 Schedule Overview 1 7 1 Revised schedule Based on advice from the project supervisor and the feedback from the mid project report the developer decided to increase the time allocated for background reading to ensure he gained extensive knowledge of available solutions to problems before embarking on a task In addition the developer increased the time allocated for writing up appendices and for testing in order to spare time for resolving errors and for contingency It became apparent that more time was vital during implementation to write up quality implementation documents Table 4 shows the revised schedule Rescheduled Tasks Implementation Period Relevant research and background reading 27 Sep 2005 01 Jan 2005 Report Planning and Write Up Incremental 12 Nov 2006 22 Apr 2006 Methodology and Design Specification 27 Nov 2005 15 Dec 2005 Implementation 15 Jan 2005 24 Mar 2006 Verification and Validation 05 Mar 2006 18 Mar 2006 Release Packaging amp Report Finalisation 24 Mar 2006 08 Apr 2006 Appendix and Glossary Compilation 16 Apr 2006 25 Apr 2006 Reflection and Project Review 15 Mar 2006 21 Mar 2006 Table 4 Revised Schedule 1 8 Relevance to Degree This project will require the developer to utilise a wide range of knowledge gained throughout his
12. Pattern Design patterns describe classical models of solving recurring problems in software design ASP Net provides a classical model for implementing web applications called the Model View Controller MVC pattern The main objective of this model is to decouple the system s presentation layer View from the data processing layer Model so that the same model layer could be used for various views it also enables better organisation and code reuse Design Patterns 2004 The MVC pattern was selected as the overall framework in which further architectural design of system components was done using design models such as the object oriented model Model The model layer handles data processing and storage it also encapsulates most of the code functions and result calculation logic View The View is the graphical user interface it handles the presentation of results produced by the model and is separated from the complex data and calculation operations Controller The controller handles events from user interactions and in turn calls some methods of the model layer which eventually update the view MVC 3 Tier Cross Section 3 lier Application Model Html Files Presentation Web Pages Layer Event Handlers Application Layer Calculation amp Data Access Classes View Model Controller pee Leer Figure 4 Design Pattern Overview E tored data The system was dissected into MVC
13. There is no overriding factor for project failure however there are some fundamental problems that regularly face projects that fail Lack of planning may lead to project failure because developers may fail to foresee likely problems and may lead to things being done in an unorganised manner Long schedules or a lack of them may be life threatening to the project because it leads to systems being delivered late or when their functionality is out of date Poor requirements specification might mean that users will not be satisfied by the application and if it delivers the wrong functionality then the project might fail Another common cause of project failure 1s scope creep a situation whereby the project grows insidiously during the project lifecycle User involvement is crucial to this project lack of it may be fatal to the project and result in the system being unacceptable or fails to meet expectations User involvement is essential for users to feel as contributors to the final solution and might make it easier for them to accept the new system Stellman and Greene 2005 In view of these problems the developer sought to identify ways of maximising the success of this project There is a clear need for planning it enables problems to be put into perspective and for the developer to choose the right solution Clearly defined and measurable results are very important for scheduling user acceptance testing purposes and to the success of the pro
14. UES 26 Apr 2006 v 26 Apr 2006 Y 06 06 06066665 Default Serum Date Optional If provided this date is automatically copied to each serum date field but can be overwritten in individual fields if there are samples from different dates Ultrasound Site Optional If results are received from more than one ultrasound centre the appropriate one must be selected as centres can have different calculation parameters Ultrasound markers Complete all enabled fields For a twin pregnancy a result must be entered for each fetus Is MoM means that the value entered is a MoM not a raw level 115 Serum Markers Complete all enabled fields Results from two different serum samples may be entered for each marker Is MoM means that the value entered is a MoM not a raw level 25 4 4 Additional Information Additional Information Date 1 value 1 Date 2 value 2 Weight Kg Diabetes smokingL ivFL Prevrz1 PrevT18 Weight Optional Enter a weight measurement taken as close as possible to the serum sample date A second weight and date can be entered if there is more than one serum sample Covariables Optional To select Covariables check mark the Covariables to be included in the risk calculation 25 4 5 Disorders fests T21 O T18 HR 11813 E Select all the disorders required by check marking the appropriate checkbox at most one disorder is required for a calculation to su
15. Za COMS CANIS ENTE TT T TTE 14 3 3 LAK CHOICES bu DI o 14 34A AA A 14 Dee Riskanalys a Lm 16 2 07 Existe GY SUSI cepe sete pa obe Ee QUERN E PIaM E eM RU MINE PRIUS PAMa M dI ODE 17 4 System Requirements inci 18 4 Elicitation Techniques ada 18 Z2 Regumreme ms Crate EID tete id 19 4 3 Puncttonmaland Data Requirements eie ee ien HR PE t eR te a SEE PUER dts 20 AA JNon dunctonal Bequite uelis ties 20 4 5 External Interace Requirements ouais qi e e n ER ee ben RR ADU eee ne bad tes 21 AO dqR gumements BvaludblOftus titi 22 5 Methodology and Development Technologies eee ee eee eee eee eee e eee ee eeSe 23 5 21 Development Technologie AAA 23 3 2 Prosramuime I AA eei edid eddie uide edd d 25 5 3 ao arta e ul EC RT DE 26 EE clu i A ESIP E E E E TE E A EA T N EO A E TE EEO ETA E 27 6 1 BEER AAA O o A T TEN 27 5 2 Arehitecutal O uoo oe RTI epu mI ee 28 6 3 EPS ODE A AAA Se 30 54 Weserinteriace View Desio ieinu e ees es cede elas A 30 6 5 Controller BIS SITAE TOT DT Rete ein ner ne aed he ee ner Nene Renin TE 31 6 6 O M E MER I EM M NEM 3 5 Databaso Desi il DI I Ma UM DI NEU 32 65 Development blandas 34 oS Des Du a AAA er PSOE eter Peer TSE eT tet ee Aree nr 35 E SO C TAC TU ATION HTH 36 7 1 IDLE ops S NNUS een NN REDE EE T T TM 36 a OSE T NCH se terete dise 36 7 3 A EE TD DUE 34 7A Data ryanna Nalda Midis 37 7 5 Development Sunma toas 38 7340 Howdhe Systm WO a TOTE E 3
16. a data store properties only then it is probably unnecessary o For public properties use module variables and property Set Get Let not public variables o If a group of input data must be set before calling any methods general class data not specific to the method then consider using module variables instead of public properties and setting them all in a single data passing method DCOM elipse Markers Covarlnfo o Don t return results functions or public properties as user defined types they can t be accessed by VBScript Using obj ects o Declare object variable as specific class type not As Object wherever possible early binding Providing collections o Create your own specific collection class instead of using the generic Collection class elipse Markers ActiveX dll interface o Where useful to client programmer and sparingly provide component global variable Declare as public in standard module read write using property Set Let Get in GlobalData global multi use class used for this purpose only Database connection for each database have DatabaseName Connect class that just handles database operations reads writes etc Use ADO for any type of database elipse elModelConnect o f connection information shouldn t change during running of client program receive it in component global variable s o In Class Initialize set module variable to open ADODB Connection using connec
17. ardPrices flexOrders hsbVolume imglcon imacboProduct IsAlllcons IbIHelpMessaae linVertical IstPolicyCodes IvwHeadinas mpmSentMessage mpsSession mciVideo mnuFileOpen mvwPeriod chSalesbyReqion msaClients mstFirst oleWorksheet optGender picVGA cloToolbar praLoadFile rdTitles rtfReport tabOptions txtLastName Timer tmr tmrAlarm Toolbar tlb tlbActions UpDown upd updDirection Vertical scroll bar vsb vsbRate For controls not listed above you should try to standardize on a unique two or three character prefix for consistency Use more than three characters only if needed for clarity For derived or modified controls for example extend the prefixes above so that there is no confusion over which control is really being used For third party controls a lower case abbreviation for the manufacturer could be added to the prefix For example a control instance created from the Visual Basic Professional 3D frame could uses a prefix of fra3d 4 Suggested Prefixes for ADO Obect Prefix Example Command cmd cmdRecordSource Connection con conPatientsDb Database db dbAccounts Field fid fldAddress Parameter orm orm obCode Recordset rst rstForecast 5 Suggested Prefixes for Menus Menu caption sequence Menu handler name File Open mnuFileOpen File Send Email mnuFileSendEmail File Send Fax mnuFileSendFax Format Character mnuFormatCharacter Help Contents mnuHelpContents 6 User Defined Types In a large project with many
18. corresponds to the displayed results PI 43 2 All fields are positioned correctly and filled with the right values do labels match their values PI 43 2 The PDF file is generated with default A4 dimensions PI 43 1 Manual comparison of the patient report shows that the patient report is generated correctly PI 43 2 comparison of the patient report to the template PDF file showed that the positions of control were not being changed when the template was filled to create the patient report PI 43 2 The PDF file is generated with default A4 dimensions 210 x 94 297 mm 20 2 8 Exception handling Test Case Specification Notes PI 43 1 All errors messages are cleaned of brackets and punctuation to prevent conflict with JavaScript PI 43 2 The correct error messages are displayed for every exception PI 43 2 No post backs occur when an error is raised on the client side PI 43 1 All errors messages contain strings only verified through almost test cases PI 43 2 The correct error messages are displayed for every exception on every test case the displayed error messages are checked for correctness PI 43 2 Visually checked by raising errors and checking that no post backs occurred when an error had been raised on the client side 20 2 9 Installation Test Case Specification Notes PI 43 1 The system works correctly on an internet information services server see Appendix XX Product Content for installation inst
19. date not supplied O Error on calculate Missing pregnancy dating information Dating Measurement O The input measurement is outside the gestation measurement range specified acceptable for the Dating type Error on calculate X value out of range Measurement supplied when not required LMP Assisted Reproduction LMP calculations must be successful but an error on calculation must occur for Assisted Reproduction calculations O Measurement not supplied when required Gestation CRL BPD HC O Error on calculate Missing pregnancy dating information Monozygous O Set to true when the pregnancy 1s not a twin pregnancy O Error on calculate Invalid pregnancy dating information AR Event O Invalid AR Event not Fertilisation or Implantation O Error on calculate Invalid pregnancy dating type Egg Extraction date O Invalid Egg Extraction date When dating type Assisted Reproduction O Error on calculate Invalid pregnancy dating information Age on EDD O Invalid Age on EDD 1 e gt 60 or lt 1 O Error on calculate Invalid pregnancy dating information Donor DOB O Invalid Donor DOB when Age on EDD is not given Resulting age outside the acceptable age 1 e gt 60 O Error on calculate Invalid pregnancy dating information O Invalid Donor DOB Resulting age gt 60 when Age on EDD is given O Error on calculate Invalid pregnancy dating information Correct calculation the Donor DOB is not used
20. disorder Malformation of the spine resulting in a spinal opening Effects include paralysis nerve damage and bowel bladder complications Pregnancy is divided into three trimesters The first trimester is 0 13 weeks gestation The second trimester is 14 26 weeks gestation A trisomy of chromosome 18 effects include mental and serious Edward s syndrome health problems Trisomy 21 Down syndrome Down syndrome is a variable combination of congenital malformations caused by Trisomy 21 It is the most commonly recognized genetic cause of mental retardation 60 I5Appendix C Introduction 15 1 Project Schedule Month November December January February March April Wekm BE aik L 01 Pe reparation ua S E DLL A ee Research and objectives specification Requirements gathering Requirements specification Requirements evaluation Ew NM Background reading Feasibility Analysis Risk analysis High level design Low level design of components a Object design User interface design W mW Database design Design review and evaluation MINS Documentation planning 4 u Mid term project report MINS E E Eu Ll EN pt L LL MN LLL NENNEN EN L LL L LL LL EN lp LEES L LL SIS il Noni re Denier c Dima o Fun Wah Dini marane EE s aa B Project document write up INI iain L p
21. have a white background and a black fore colour This means the application will be usable on most systems and accessible to most users including colour blind people Buttons will be accompanied with small icons that illustrate what the button does The icons will improve the usability and appearance of the system 6 5 Controller Design In the MVC pattern the controller is responsible for handling events raised from user interactions to initiate appropriate system processes ASP Net hides the complexity of handling events that happen on the server or client from the developer such that code the required to implement the controller is relatively simple and straightforward To receive notification of an event the application is required to declare an event handling method for that event When the event occurs the runtime environment invokes and executes the declared event handler If the event occurred on the client side ASP Net will capture the event information transmit the event message to the server and call the appropriate event handling method As a result all the application has to do is declare event handling code for the events that it wants to be notified of This made it unnecessary to specifically design the system controller 6 6 Security Threats faced by the system can be categorised based on the goals and purposes of the attacks Understanding threats helps in developing security and mitigation strategies for those threats T
22. inheritance and encapsulation Liang 2003 The caching mechanism provided in the framework greatly improves performance Business logic is totally separated from presentation through the code behind development model It provides library classes and controls that enable easy access to databases and rapid application development ASP Net can maintain session variables enabling a web server to recover lost data when an exception happens and to provide state to the inherently stateless web ASP Net is primarily restricted to Microsoft Windows platforms because it uses platform specific objects The use of scripting languages for client side code does not scale properly for large solutions Memory usage is not very efficient due to the large amount of code it generates resulting in longer execution time ASP Net is expensive in comparison to other web technologies 5 1 3 Perl Perl is a remarkably powerful scripting language and is highly efficient in terms of string processing memory management and data typing It is a procedural programming language that derives mainly from C and has one of the biggest collections of third party modules Wall 2000 Perl is powerful very expressive and provides an infrastructure for handling common tasks efficiently It is widely used and supported database drivers are easily accessible Perl can be implemented on many platforms consequently Perl code is highly portable Many tasks such as memory manageme
23. large complex site na Easy to use Search function is provided as needed na Site accommodates novice to expert users Functions are clearly labelled y Essential functions are available without leaving the site r Plug ins are used only if they add value Y Site reflects user s workflow y User can cancel any operation r Clear exit point is provided on every page X Per page size 19 less than 50K to accommodate slow connections All appropriate browsers are supported T Important information and tasks are given prominence Information of low relevance or rarely used information is not included P Related information or tasks are grouped on the same page or menu y in the same area within a page r Language is simple without jargon y Paragraphs are brief y Links are concise expressive and visible not buried in text Terms are defined v Site is designed to require minimal help and instructions T Help and instructions if needed are easily accessible 82 System and user feedback It is always clear what 1s happening on the site visual hints etc T Users can receive email feedback if necessary x Users can give feedback via email or a feedback form X Confirmation screen is provided for form submittal y All system feedback is timely P Users are informed if a plug in or browser version is required and a link is provided y to the necessary plug in or browser site Each page include
24. locales and will automatically detect user s locales and display month names in the user s language and validate input using the user s regional formats The system s code was written in an object oriented manner with classes that can potentially be extended and inherited without having to modify existing code Should the need arise to modify the solution the code is well documented and implementation documents were created to ensure the system is maintainable These software and technical documents were reviewed by the MI project supervisor to check that they complied with the MI software engineering guidelines 9 7 Comparison to existing systems The Analysis chapter gave an analysis of currently available systems and described their advantages and disadvantages It seems fitting to evaluate the current system to ensure it does not suffer from the same drawbacks and that it delivers more than the existing systems The previous system developed by MI did not offer the latest marker technology it was not network enabled and was only Windows based The final solution incorporates some of the latest research findings in prenatal screening technology it is configurable and can be hosted on a web server making it usable over a network The final solution is more efficient and provides a better reporting 47 system that produces multi stage results so that users can see the results of every important calculation stage The application hides t
25. object and relationships capture how two or more entities relate to one 22 another Attributes are properties of entities for example the height attribute of a person entity Entities must also have identifiers which name or uniquely identify instances of entities Elmasri and Navathe 2003 Customer Mame Identity A 208 M M Pregnancy Username User ICA TE Dating ce I M Passvvord CTE s M t Mame Pe Disorder M M Marker l Uses Firial Risk Equation Assigned Has Has M S Distribution Marker Risk P et Sel Factors SS Figure Entity Relationship Diagram Gestation Equation Measurement Limits Age Risk Equation LA Mame Correlation values Haz Normalization Parameters The relational model enables the design of a consistent logical representation of information The logical representation of the database was normalised to third normal form Appendix E to improve performance avoid data duplication and inconsistency problems The eventual relational schema to be implemented in SQL Server is shown below User Customer Id Username Password Pregnancy Dating Id Name Gestation Equation Min Measurement Max Measurement Markers Id Name Disorder Min Value Max Value Gestation Range Disorder Id Name Age Risk Equation Final Risk Equation Marker Set Id Disorder Markers List of between 1 and 5 markers
26. system results and the level of user satisfaction with the final solution 4 How does the system compare to existing systems The system is judged by how well it compares to existing system it must bring new benefits to stakeholders to justify 1ts creation 5 Does the system meet the usability requirements and development guidelines Although it is important to give priority to system functionality its usability is of great importance because it determines acceptance and adoption levels This criterion is a heuristic evaluation of the application and checks its usability accessibility and conformance properties 6 Has the project been delivered on time and passed user acceptance testing and evaluation The project must meet the schedule requirements specified in the project plan to be successful If a project 1s delivered late the required functionality may have changed and this might lead to loss of customers and project failure In addition the system will be evaluated according to the ISO 9126 software evaluation criteria section 9 1 provides a detailed explanation of this criteria 1 10 Report Structure The structure of this report mirrors the project development process itself Most chapters begin with a summarisation of the background research carried out The next chapter explains the problem domain in detail and justifies the methodology used in solving the problem It is followed by an analysis of current systems projec
27. test version before submission for clinical validation by external validators The system shall validate weight values so that they are within acceptable ranges The system shall validate the patient date of birth so that maternal age values so that they are within acceptable ranges The system shall validate marker test results values to ensure that they are within the acceptable range of values The system shall not store cookies on the users machine since most windows system have cookies disabled The system must be able to accept uploaded files import them to a database and format data correctly and calculate results The system shall be able to interface correctly with the existing dynamic link library and when it calls methods of that dll it must do so using the correct number and type of parameters The system shall calculate marker mom results from the input marker values The system shall calculate pregnancy dating results from given pregnancy dating clinical testing results The system shall provide an export function that will allow users to print results and save them in some easy to retrieve format The system shall provide a web form navigation function on every page in the form of a control that is reused on all web forms of the application The system shall provide a date entry function for entering dates into date fields This function must also validate dates before accepting user input The system shall display dates and numbers in
28. the same format as the users locale The system shall display numbers to a precision of 3 decimal places Truncation and rounding of values can only be done at the end of a calculation the system must not truncate values during a calculation All operations that can be executed on the client side must be scripted for client side execution whenever possible to minimise post back times Security SE The system shall provide access to authorised users The system shall protect encrypt sensitive information while it is being transferred over the internet The system shall prevent access to unauthorised users The system must provide a user identification functionality without having to store cookies on the users machine External users must not be able to access internal server data through the system 78 SE 6 SE 7 SE 8 SE 9 SE 10 SE 11 SE 12 UI 1 UI 4 UI 5 UI 8 UI 9 UI 10 UI 11 UI 12 UI 13 UI 14 The system shall abide to the data protection act of 1994 The system shall prevent access to users once they have logged out or once their session has expired until identification has been done The system must not allow users to view other users data The system should be protected from security risk as much as possible The system must be capable of identifying users using a customer identity username and password The system must be accessible to registered clinicians only The system must display a standard
29. transcript 4 2 Requirements Gathering In addition to interviews the MI project supervisor would review the application on a regular basis give suggestions and clarify requirements Figure 3 shows that system requirements gathering was an iterative process with continual feedback from each task 19 Requirements checking Requirement prioritisation Specification d Documentation Interviews Requirement evaluatian WORD Mewpoints Requirements collection Classification Figure 3 Requirements Specification Process 4 3 Functional and Data Requirements Functional requirements describe the services that the system 1s expected to provide They describe all external and programming interfaces of the system and 1ts functionality requirements such as results calculation data manipulation and processing Appendix F User Requirements gives a full listing of the system s functional requirements 4 4 Non functional Requirements Non functional requirements define specific system quality attributes properties and behaviour They are just as important as functional requirements since they determine how usable a system is to users The paragraphs below describe the non functional requirements of the final solution Appearance The system shall comply with the existing colour scheme used on existing MI applications It should use similar images and banners to those already in us
30. user defined types it is often useful to give each such type a three character prefix of its own If these prefixes begin with u they will still be easy to recognize quickly when you are working with a user defined type For example ucli could be used as the prefix for variables of a user defined Client type 74 17 Appendix D System Requirements 17 1 Interviews Interview Information Interview Title Project Conception Interview Interviewee s Carol Wilson MI Project supervisor Daniel Taylor mente Que A O Could you explain the problem domain of the project and how the proposed system would be of use What will be the main purposes of the system Who will be the user of this system What are the benefits that you or the users might expect from this system When do you want the system do be delivered What are the constraints of the system from a business point of view In order of priority what are the main things you require of this new system How is the system going to be used J N How would you like the system to be developed in order to ensure minimum disturbance to you whilst maximising your involvement What are the main or likely rival systems to this product What are the likely legal requirements that this system is required to satisfy How will the system be deployed What security mechanisms would the system be expected to provide Follow up questions and additional answers Wh
31. will describe mitigation strategies and allocate project resources for managing risks In order to manage risks effectively the process of identifying risks and developing strategies for avoiding transferring or reducing the negative effect of the risk has to be documented clearly Boehm 1991 The developer compiled a risk register as part of the project risk management The risk register is a log that lists all the identified risks their analysis and mitigation strategies Contingency plans were also drawn up for every risk to provide alternative solutions should the mitigation strategies fail This document was used to track and monitor risks throughout the life cycle of the project The risk register can be found in Appendix C Analysis Provided the strategies specified in the risk register are followed with a little fortune the likelihood of the risks identified in the risk register happening is minimal The risk register showed that the risks most likely to affect the project would not be fatal for the project if mitigation strategies were implemented early enough Throughout the lifecycle of the project the developer consciously tried to make informed decisions to prevent risks and involved stakeholders with the intention of stipulating preventive actions 3 6 Existing systems Existing systems that are used in prenatal screening were analysed to aid the design of the system The analysis evaluated a Screening Calculator previously
32. 505 Publisher Apress 42 Sun 2005 JavaServer Pages tm Technology Comparison with ASP http java sun com products jsp jsp asp html 43 Thompson SG 1997 Estimating a woman s risk of having a pregnancy associated with Down s syndrome using her age and serum pg 387 402 44 UKGTN 2005 ACOG Practice Bulletin Prenatal diagnosis of fetal chromosomal abnormalities Obstet Gynecol Vol 97 Part 1 45 Wald N J Rodeck C Hackshaw A K Walters J Chitty L Mackinson A M 2004 Journal of Medical Screening Volume 10 Number 2 pp 56 57 Publisher Royal Society of Medicine Press 46 Wall 2000 Programming Perl O Reilly Pages 1104 47 Westerway 2000 Ultrasonic fetal measurements new Australian standards for the new millennium Obstet Gynaecology 2000 Aug Vol 40 part 3 97 102 48 Wilcox 1993 Estimating a woman s risk of having a pregnancy associated with Down s syndrome using her age and serum 1997 pg 387 402 49 Wilson 2006 Discussions with Mrs C J Wilson The Media Innovations Ltd Scientific Software Department Manager 53 54 12Appendix Table of Contents t Appendix A RETO cion 57 2 Appendix b GIOSSAEY mi ao OO RR cR EUREN 59 S APpendix C Introduction c uen I RE PER ERI EI SEO E OE NEIN IE M 61 3 AAA 61 4 Appendi C Analysts iO RE ia ENNIO NEIN IEEE TIME 63 4 ALGAS Lus iratis ists diese ease deu ossu iM qaid a tetra Ee 63 4 2 MI Software Engineering Guidelines 68 5 Appen
33. 66 it is not an urgent priority to protect it 67 16 2 MI Software Engineering Guidelines 16 2 1 Project Organisation Development documentation as a minimum always have the following separate documents to hand and keep them up to date o Specification formal if it exists otherwise informal as complete as possible and updated as required o High level list of additions modifications and corrections to make add to this list immediately when an item comes to your attention o Current task s expanded information o Information required decisions to be made o Design decisions decisions taken about functionality and implementation with brief reasons o Essential information for project handover development packages used components and their locations version information passwords lists of finished and unfinished tasks Product documentation o User manual use AMIE Beta manual as template o CD label and cover use Template CDLabel as template Code storage and version control o Folder structure C 1MIDew lt ProjectName gt Code or Debug etc contains current version Released versions are stored on MI SRV o In general don t keep previous versions If there is a particular reason for keeping a version then ensure that folder file names make it clear that the version is not current o Version numbering Major Minor Revision Start project with 0 0 0 First release version to customer should be 1 0 0 I
34. 9 ye RPO O ea A e TAEAE EA D E ELI L C du M Le Stet 40 MEE A aee E EEEE N 41 8 1 Ed DE a Oe da 41 Bo esto RES aiii 42 9 Eva T i Ress E EEE 43 9 Evaluation ete pitones 43 A e A e set E E E II DS eA st Mame er ee T eae eee 44 C A URDU ERU A ONERE O ACROSS 46 CE MEE Ducit EE 46 95 Usability and Ace ssIDIITVC oe RR Ue mmu 47 2 6 Portability aud Manta al A idu edo 47 9 7 Comparsoinnto existmb OS 47 To GUSOPeValudbHODcst dio tuesotiid dos 48 TH cBEyaluauon FSCS UU 225 oin ad AS 48 OO SUpsestedimpto Vene AA atop dde Pedido 48 10 LOTO c 50 11 A pee rH 51 12 Appendix Table ol Contents e CET ESSEN ONERE airoso RR EUER 55 vi 1 Introduction Prenatal screening covers the procedures that are available to pregnant women in order to find out those at risk of having a child that is affected by congenital disorders especially chromosomal abnormalities such as Down syndrome Screening is mainly carried out using biochemical markers hormones or ultrasound sound waves techniques Recent developments in prenatal screening such as new ultrasound and biochemical markers and the generation of likelihood values at different stages of the calculation process are not available in existing software Wilson 2006 In addition as screening becomes more widespread screening centres increasingly wish to be able to use highly advanced systems that allow them to calculate risk results from wherever they may be in the world
35. CSng 1 234 Displays 1 234 Input From User and Database e When reading in a numeric string use the Parse method of the target variable e g Mylnt Parse myNumsString Format Provider Named Format Precision to convert the string before storing it in a variable of the correct data type Passing To Other methods SQL Server 2000 Access 2000 Conversion e Use local aware conversion functions to convert strings to numbers and numbers to other numbers CSng Clnt etc e Before conversion make sure the expression passed to the conversion function is within the range of the datatype to which it is being converted otherwise you will get an error or unwanted truncation Display Use the ToString method and the correct NumberFormatinfo s FormatSpecifiers e g MyDouble ToString dx where d is the format specifier and x is the precision specifier e g no of decimal places 71 o Web applications Set the current culture system locale in VB Net of every user s thread on form load using the Culture information supplied by the Browser object Follow the localisation instructions given for VB Net dates and numbers l Variable Scope Prefixes Scope Prefix Global q Module level m Local to procedure None 2 Suggested Prefixes for Variables Data type Prefix Boolean bin Byte byt Collection object col Currency cur Date Time dtm Double dbl Error err Integer int Lona Ing Obje
36. E HEALTH SYSTEM FOR PRENATAL SCREENING PrenatalLink Courage Chibiya BSc Computer Science Industry The candidate confirms that the work submitted is their own and the appropriate credit has been given where reference has been made to the work of others I understand that failure to attribute material that 1s obtained from another source may be considered as plagiarism Signature of student Summary Prenatal Screening covers the range of procedures that are available to pregnant women in order to find out those at risk of having a child that 1s affected by congenital disorders Prenatal screening software serves to calculate the likelihood of such a risk based on clinical tests results and pregnancy related factors Although it is possible to do such calculations manually the calculations involved are too complex to be done routinely by a physician Advances in prenatal screening have now reached a level where physicians would like to carry out prenatal screening using highly efficient internet enabled software systems Up until now prenatal screening has not been fully available on the web Shaffiq 2004 and current systems do not meet user needs such as being e Configurable and incorporating the latest research findings on biochemical and ultrasound markers e Scalable for networked applications and internet enabled e Able to give multi stage results and interpret them for the user This project aims to develop
37. Ensure that all information conveyed with color is also available without color for example from context or markup 4 Clearly identify changes in the natural language of a document s text and any text equivalents e g captions 6 1 Organize documents so they may be read without style sheets For example when an HTML document is rendered without associated style sheets it must still be possible to read the document 6 2 Ensure that equivalents for dynamic content are updated when the dynamic content changes 7 Until user agents allow users to control flickering avoid causing the screen to flicker 14 1 Use the clearest and simplest language appropriate for a site s content And if you use images and image maps Priority 1 YesNoN A 1 2 Provide redundant text links for each active region of a server side image map 9 Provide client side image maps instead of server side image maps except where the regions cannot be defined with an available geometric shape And if you use applets and scripts Priority 1 YesNoN A 6 3 Ensure that pages are usable when scripts applets or other programmatic objects are turned off or not supported If this is not possible provide equivalent information on an alternative accessible page And if all else fails Priority 1 YesNoN A 11 4 If after best efforts you cannot create an accessible page provide a link to an alternative page that uses W3C technologies is accessible has equivalent
38. Markers and Disorders objects that calculate results from the input data in conjunction with the Elipse object It uses the information provided by the user to create a patient object handles results and report generation It forms the code behind the calculation page which is described in section 6 4 29 6 3 Elipse Object The Elipse COM dynamic link library encapsulates the rules for calculating pregnancy dating markers and disorders results This component was developed by MI The final system uses this component when calculating results The Elipse object receives values and setting data from the system classes and calculates the required risk values The Elipse object has already been validated and tested by MI 6 4 User Interface View Design The purpose of the user interface is to enable users to interact with the application supply information for calculating results and initiate various application processes Consequently the user interface has to provide controls that enable users to input data calculate and view results and perform various operations Figure 7 shows the fields that each proposed page was required to provide for the underlying classes to work Security Interface Fields Customer Id Username Password Operations Login Reset Calculation Interface Fields Forename Num of Foetuses Marker Name Disorder Surname Monozygous Assay Date Prior Risk Id Dating type Marker Value Risk factors Maternal DOB M
39. O Calculating the correct results for valid inputs O Results O Calculating correct results for all dating types and parameters O Compare with manually calculated values 97 Pregnancy dating results Expected date of delivery EDD supplied or calculated Gestation returns calculated gestation at SampleDate MAEDD calculated age 20 3 2 Markers MarkerName All required elModel records must exist for the given Marker AssayDate Must be supplied in all calculations Value amp Value must be must be supplied and Normalisation Cannot be lt 0 except when the marker normalisation is Delta a mmm Gestation Type amp Both must be supplied if the value is not normalised the Mau value must be a valid gestation value TwinValue Must be 0 or not supplied 1f it s a twin pregnancy but the marker 1s not fetus specific Must be supplied if it s a twin pregnancy and the marker is fetus specific TwinGestationValue Both must be supplied if the Twin Value is not normalised MoM The gestation value must be a valid gestation MaternalWeight If supplied it must be a positive value Covariables Checkbox based if supplied they will be used in the calculation of Additional marker moms Must be a string the same key cannot be used more than once information Invalid inputs 13 MarkerName O Two markers with the same gestation added to the Markers Collection O Error on calculate Unable to add element to collection e
40. The aim of this project is to develop such a system for calculating the risk likelihood value 1 1 The Problem Domain Over the past two decades prenatal screening for fetal defects has become a standard part of prenatal care Kristol 1993 Tests conducted during the first trimester of pregnancy are designed to detect a wide range of genetic and other disorders and to give prospective parents the option of obtaining abortions if defects are diagnosed Cuckle 1997 This development has been heralded as a breakthrough in the age old battle against illness Others consider it as more than that an instrument to improve society There are invaluable benefits to screening lives can be saved or improved and major cost savings incurred by the health system and families Like any major medical development likely to occur in our age prenatal screening has its own share of controversial consequences The issues surrounding the morality of screening are the stuff of moral philosophy Kristol 1993 and agreement on the acceptable social consequences is difficult to achieve however that should not obscure the benefits that can be gained from prenatal screening The most common form of prenatal screening is ultrasound screening about 8096 of all pregnant women receive ultrasound screening Jones 2005 This method uses sound waves to generate a sonogram or image of the fetus in the womb The main purpose of this test is to establish the gestational age of
41. The schema below shows the database in non normalised form and in third normal form 18 1 1 Un normalised form User Customer Id Username Password S Pregnancy Dating Id Name Gestation Equation Min Measurement Max Measurement Markers Id Name Disorder Min Value Max Value Gestation Range Markers List of between 1 and 5 markers Distribution Equation Weight Equation Normalisation Equation Disorder Id Name Age Risk Equation Final Risk Equation List of Risk Factors 18 1 2 Third normal form User Customer Id Username Password Pregnancy Dating Id Name Gestation Equation Min Measurement Max Measurement Markers Id Name Disorder Min Value Max Value Gestation Range Disorder Id Name Age Risk Equation Final Risk Equation Marker Set Id Disorder Markers List of between 1 and 5 markers Distribution Parameters Id Marker Distribution Equation Weight Equation Normalisation Parameters Id Marker Normalisation Equation Risk Factors Id Name Disorder Factor 81 18 2 Usability Guidelines Massachusetts Institute of Technology IS amp T Usability Guidelines Courtesy of MIT http web mit edu 1s web reference guidelines usability html c Current location within the site is shown clearly T Link to the site s main page is clearly identified na Major important parts of the site are directly accessible from the main page y Site map is provided for a
42. a web system for calculating the risk of a pregnancy being affected by a fetal anomaly The system will use clinical test results and pregnancy related information to calculate the risk of a pregnancy being affected by a selection of chromosomal disorders Based on the calculated risk value the system will interpret whether a pregnancy is affected or not using population risk cut off parameters This project aims to produce a tool that automates the risk calculation process for physicians Derived objectives of the project are to produce a solution that fulfils the current needs of physicians to have a system that is internet enabled highly configurable and that produces multi stage results The final solution would also relieve physicians from carrying out of complex analysis and interpretation of results Acknowledgements This project was proposed by Media Innovations Ltd a company that specialises in developing medical diagnosis software Without their involvement and support this project would not have been a success I would like to express my utmost gratitude to the company and their representatives for their support and for enabling me to develop the final solution I would also like to give thanks to the following people for their suggestions and input during the compilation of the system requirements and the project lifecycle First and foremost I would like to thank Mrs Carol Wilson this project s supervisor at Media Innova
43. aediatrics 2004 Sep Volume 114 pp 889 94 7 Design Patterns 2004 The Model View Controller Design Pattern http www adobe com devnet flash articles mv controller html 8 Easterbrook and Nuseibeh 1996 Using Viewpoints for Inconsistency Management Software Engineering Journal Volume 11 No 1 Jan 1996 9 Elmasri R Navathe S 2003 Fundamentals of Database Systems fourth edition Addison Wesley pp 177 182 10 GM Stocks 2005 Related Articles Links Spina bifida tethered cord and regional anaesthesia Anaesthesia PubMed volume 60 2005 Nov 60 pp 1114 9 11 Haddow J E 1999 Screening of maternal serum for fetal Down s syndrome in the first trimester 1998 pg 955 961 12 Hernandez D and Fisher EMC 1996 Down syndrome genetics unravelling a multi factorial disorder Publisher Planet papers pages 1411 1416 1996 Medline Plus 1996 13 Herndon 1997 Software Productivity Consortium Evolutionary Rapid Development SPC document SPC 97057 CMC version 01 00 04 June 1997 Page 9 14 Hetzel 1988 The Complete Guide to Software Testing QED Information Sciences Wellesley Massachusetts second edition 1988 15 Hook EB 2000 Rates of chromosome abnormalities at different maternal ages Obstet Gynaecology Report 2000 Volume77 pg 282 5 16 Horton 2003 Beginning Java 2 JDK 5 Edition ISBN 0 7645 6874 4 Dec 2004 pp 99 111 5 17 IEEE 1998 Human centred design processes f
44. all server and workstation data in a remote location Add ability to use a workstation away from the network as well as within 1t to enable the site to be managed gt Rationale for Backing up all data will enable remote access and the website will be gt evaluation managed as normal Customers will also be able to have access to the entire database as opposed to just what is stored on the web server gt Risk Failure to follow future usability and pricing trends gt Effects Customers may turn elsewhere 1f other services are more accessible and user friendly Will lose business if opportunities are not taken Impact Likelihood Type Medium Low Operational gt Action Specify job tasks and objectives when hiring staff commercial advisor and support will need to stay in touch with changes to such businesses gt Rationale for Keeping ahead of the times and providing fresh changes updates to gt evaluation the website and competitive prices will keep customers interested and ensure their business continues to come our way gt Risk Software failure gt Effects Poor service low sales customer dissatisfaction Impact Likelihood Type 65 Medium Action gt Rationale for gt evaluation gt Risk gt Effects Impact medium gt Action for gt Rationale gt evaluation gt Risk gt Effects Impact medium gt Action gt Risk gt Effects Impact High gt Action gt Rationale for gt evaluation gt Risk g
45. ally in order to satisfy requests from remote browsers clients sent via the Common Gateway Interface CGI CGI is a standard that enables a web server to execute applications generate and send content to browsers for display The development technology used to implement a system normally determines the web server on which it is capable of efficiently running on for example Microsoft based systems are more suited for running on a Microsoft web server Five potential technologies for implementing the final solution were considered with the aim of selecting the most suitable they were JavaServer pages ASP Net Perl Python and PHP 5 1 1 Java Server Pages JavaServer pages JSP technology was developed by Sun Microsystems and enables the development of interactive pages for web applications It is part of the Java architecture and has access to all its components such as JavaBeans and servlets JSP was designed to be both platform and server independent and adheres to the write once and run anywhere philosophy of the Java architecture Sun 2005 JSP technology is platform independent and can run on any web server without having to be design for portability The use of Java for scripting provides many benefits over basic scripting languages such as superior performance and scalability Because JSP is developed by the java community an open source software development process it has undoubtedly gained from the robustness of that extended communi
46. at define culture related information the calendars in use and the format patterns for dates and numbers The NET Framework also defines standard format providers and specifiers e g d for ShortDatePattern for formatting and parsing dates see the DateFormatinfo class Input From User e Use Date Parse inverse of formatting to create a date from a date string e Always store the resulting date from parsing in a date variable of the correct type Date DateTime etc Passing To e SQL Server 2000 4 e Access 2000 4 Conversion e Use the IsDate function to determine if a value can be converted to a date before converting Display e Format using the overloaded ToString method when displaying dates e g Date toString Date Format Provider When displaying don t use explicit formatting e g Format 3 0 00 Instead use the format function e g Format 3 0 00 without any literal characters characters are displayed exactly as typed Use standard numeric format strings when formatting common numeric types for display They take the form Axx where A is the format specifier e g D for decimal and xx an optional integer is the precision specifier e g Number of decimal places E g Double ToString CDbl 17 3344 D3 VB Net does not convert strings to numbers for you so make sure you convert numerical strings to numbers first before formatting thus Format 1 234 2 3 Displays 4 Format
47. at security features do you normally provide for your software What contractual requirements is the developer required to abide to What technologies do you use for developing your software What browser technology is mostly used in the health industry Plans for future meetings 76 17 2 Functional Requirements ID RC 1 RC 2 RC 3 RC 4 RC 5 RC 6 RC 7 RC 8 RC 9 RC 10 RC 11 RC 12 RC 13 RC 14 RC 15 RC 16 RC 17 Calculation RC Requirement The system shall provide a data input form that enables users to input biochemical markers pregnancy dating and disorder information The system shall validate user input numerical and dates according to user locales formats The system shall give appropriate error messages in order to enable users to correct errors before submitting a form The system shall provide a results form for displaying the calculated pregnancy dating results marker moms and disorder results The system shall provide a data import form that will enable users to upload files that contain records for result calculation The system shall provide a data export function that will enable users to download calculated results that will be stored in an access file or text file The system must calculate results based on the settings equations and parameters stored in a settings database the settings database must be designed to meet the data requirements of the pregnancy dating markers and disor
48. ation could begin The manager s review evaluated system designs to ensure they correctly implement all system requirements and that requirements traceability is maintained in documentation 35 7 Implementation This chapter describes how the system was developed based on the design decisions and explains the code behind the system s functionality Due to document scope reasons not all aspects of the system could be explained in detail in this chapter Consequently only a limited range of fundamental or advanced functions of the system were described Appendix H Implementation explains in detail the code of a broader range of the final solution s components 7 1 Database The database was implemented in SQL Server 2000 Appendix G explain how to install the database 8 tables were created corresponding to section 6 6 The structure of all tables 1s identical to the structure defined during design In addition all fields were assigned appropriate data types for example numerical fields were given an Integer or double data type The relationship between tables was defined using table constraints to ensure referential integrity was maintained 7 2 User Interface The user interface web pages were created using ASP Net Usability and Accessibility guidelines were followed to improve the system s usability accessibility and appearance The look of the application is similar to that of other software products developed by MI Figu
49. ation is adequate for most purposes All the tasks that a user might want to carry out on the system are explained in detail The user guide was written in a concise and straightforward manner that enables the user to quickly find the information they need Date input The HTML based date picker was implemented to enable efficient entry of date system values It prevents users from entering invalid dates and speeds up the date entry process This control is cached on the client such that the control needs to be downloaded only once every sixty days greatly improving bandwidth usage Page caching All images and static parts of web pages are cached on the client side to reduce functionality response times and reduce bandwidth usage The system s pages take less than 5 seconds to download on a broadband internet connection Internationalisati The application displays dates in the user s language and validates both dates and on numbers according to the user s locale formats The application automatically detects user s date and number formats and displays month names in the correct language However the system could be improved so that it displays text in labels using the language of the current user Table7 Requirements vs Implemented Features 9 2 2 Aims and objectives This aim of the project was to develop a web application that will accept maternal biochemical and ultrasound test results from clinicians and calculate the likelihood of the gi
50. base class for retrieving user details from the database 4 Logging in functionality and the security features described in section 8 5 1 Improvement of the login page based on user feedback evolutionary process and interface design guidelines 6 Skeleton calculation page containing the fields required to perform calculations without 34 10 11 12 13 14 15 any styling or colours Skeleton results reporting page containing the fields required to display calculated results for unit testing purposes Pregnancy Dating class implemented Markers class implemented Disorders class implemented Improvement of the calculation and results page based on user feedback evolutionary process and interface design guidelines Patient report page implemented and improved based on user feedback Help system implemented Testing and integration of Pregnancy Dating Markers and Disorders classes Final verification and validation Table 6 System Functionality 6 9 Design Review A design review was carried out to ensure that the system can proceed into implementation and meet the stated requirements in the available time and other system constraints The review assessed whether the implementation platform provides the required capabilities and services needed for the implementation to succeed In addition the MI project supervisor analysed the functional graphical interface designs to determine their completeness before implement
51. cceed Notes For twin pregnancies with ultrasound marker measurements fetus specific risks are provided It is essential to ensure that all fetus specific information is entered in the same fetus order in every place on the data entry screen 116 26 1 Results Results Patient Markers Mora Delta Gestation Mor Delta Gestation LED t C 1L 1 LIE EDD d Woy 2006 MoM Delta Gestation MoM Delta 2 Gestation 2 MAEDD 23 yrs mnths 11 00 i2wksadas L ES Download Reader adobe Reader Twin 2 Prior Risk Final Risk Result Final Risk Result 1 in 1447 1 in 2771 Screen Negative Risks 26 2 Patient Report a Click on create report to generate a PDF patient s report b A dialog box will be displayed asking you to save or open the patient report file File Download Do you want to open or save this File POF i Mame 0060426 1 122 pdf Type Adobe Acrobat 7 0 Document 256 KB From 213 237 225 18 117
52. ct Maternal Serum hormones that are present in a pregnant woman s blood such as AFP and UE3 A small amount of villi hair like fringes of the placenta is removed using either a needle inserted into the abdomen A disorder caused by the abnormal distribution of chromosomes that carry genes A component of a cell that contains genetic information DNA and proteins The expression of the genes present in an individual e g blood group An image of a persons chromosomes used to screen for disorders Disorders that are present at birth The formation development and variation of cells A Database management system it handles the creation of databases their management data storage and access Examples are SQL Server Oracle and MySQL Procedure that involve using a probe with a camera to inspect inside the uterus A malformation of the fetus during its development An unborn baby after 8 weeks from conception to the delivery date Gene mutation is a permanent change in the DNA sequence that makes up a gene An alteration of a gene it may have hazardous consequences on the 59 Genetics Gestational Age GA Human Chorionic Gonadotropin hCG Intellisense Klinefelter s syndrome Maternal Serum Menstrual history Metabolic disorders MIT Neural tube defect Patient cohorts Prenatal diagnosis Risk Factors Risk likelihood Screening Spina bifida Trimester Trimester 18 affected fetus A topic related
53. ct obj Single sng String str User defined type udt Variant vnt 3 Suggested Prefixes for Controls Control type Prefix 3D Panel pnl ADO Data ado Animated button ani Check box chk Combo box drop down list box cmb Command button cmd Control specific tvpe unknown ctl Data dat Data combo dbc Data arid dad 42 Example astrUserName mblnCalcinProgress dblVelocity Example blnFound bytRasterData colWidaets curRevenue dtmStart dblTolerance errOrderNum intQuantity InaDistance obi Current sngAverade strFName udtEmplovee vntCheckSum Example pnlGroup adoBiblio aniMailBox chkReadOnlv cmbEnalish cmdExit ctlCurrent datBiblio dbcAuthor dadTitles Data list Data repeater Date picker Directory list box Drive list box File list box Flat scroll bar Form Frame Gauge Graph Grid Hierarchical flexarid Horizontal scroll bar Image Image combo ImageList Label Line List box ListView MAP message MAPI session MCI Menu Month view MS Chart MS Flex arid MS Tab OLE container Option button Picture box Picture clip ProaressBar Remote Data RichTextBox TabStrip Text box dbl drp dtp dir drv fil fsb frm fra qau gra ard flex hsb Ima imacbo ils Ibl lin Ist vw mpm mps mci mnu mvw ch msq mst ole opt pic clp rd rtf tab txt 73 dblPublisher drpLocation dtpPublished dirSource drvTaraet filSource fsbMove frmEntry fraLanquade gauStatus graRevenue
54. ct the final solution In addition the proliferation of Windows based systems in the health industry made ASP Net a more attractive choice 5 2 Programming Languages The chosen development technology ASP Net provides three main programming languages VB Net CZ and J J was not considered as a viable option due to its lack of widespread use and support C and VB Net perform largely equivalently since they have access to the same framework libraries However VB Net is simpler and more productive due to better intellisense optional parameters and background compilation Consequently VB Net was chosen for writing the final solution due to its relative simplicity and natural language case insensitive approach It also has key strengths in its support for optional parameters in method invocation and good exception handling 25 5 3 Scripting languages Client scripting is used to make web pages dynamic and interactive for example it might be used to validate a field so that a user gets immediate feedback when they enter invalid data The main scripting languages considered for client side scripting were Jscript VBScript and JavaScript The main features affecting the choice of scripting language are target platforms and language features VBScript is a subset of the Visual Basic language It is flexible in that it can be used in a huge variety of applications and provides good error handling However it does not work on Netscape browsers
55. ctional or non functional requirements 4 1 Elicitation Techniques There is no perfect or universally applicable requirement gathering technique several approaches were considered for this project Four main techniques were considered with the aim of choosing the most suitable ones they were ethnography scenarios view point based elicitation and interviews Ethnography is an observational technique in which an analyst observes the work environment where the system will be used Agar 1996 From this social or organisational context the analyst can discover implicit requirements which reflect actual work processes It can be particularly useful at discovering process details that are often missed by other techniques However due to its focus on the end user it wont discover other more specific or new requirements As a consequence of its incompleteness ethnography has to be used in conjunction with other elicitation techniques Scenarios are a requirements gathering techniques in which users describe how they might interact with a software system in a given real life scenario The scenario descriptions can then be used to formulate system requirements This technique is particularly useful for adding detail or clarifying outline requirements However it requires the developer to have considerable understanding of how the final system might work if they are to present the user with enough scenarios to comprehensively capture requirements
56. cuted every time the page 1s loaded it mainly initialises the login vars Private Sub btnReset Click sender event Handles btnReset Click Clears the text in the login fields only called if the client h javcript disabled Private Function ChkUserDetails Boolean Checks a users login details and returns true if they are correct Private Sub SetInitialFocus ctrl Control Sets focus to the given control on this page Private Sub SaveLoginInfo Writes the current username used to login in the cookie username 107 P re natal PIS Web Solution L IM S User Guide Contents TY K 11T lt s m A 108 ZEB Pip e 111 S MEE AICI ATION O 108 3 IB CAT o E E EA M 112 D MEME Cnt 113 3 3 Calculatine TOSUILS uerioanatensrire E E EEEE 114 SU Data SLT Se ST erat atar deas 114 AEN uu asia A 117 4 cui M 108 LP MEME curn Cr CNN 117 109 This document contains instructions on how to use the Prenatal Link beta test system Although this system has been subjected to preliminary testing it is yet to undergo extensive clinical evaluation and testing by third parties It 1s expected that the user will have read the accompanying report or has background knowledge on prenatal screening Knowledge of Prenatal screening theory is required to be able to use the system The PrenatalLink system was developed b
57. d ltest Elipse Version 3 0 0 8 PrenatalLink Version 0 1 Beta Test Disclaimer This report was produced by Elipse PrenatalLink v0 1 which is an unvalidated Beta Test version THESE RESULTS MUST NOT BE USED FOR CLINICAL PURPOSES Figure 17 Sample Patient Report 89 20 Appendix E Testing 20 1 Test Plan Project no and name Developer Courage Chibiya PrenatalLink Date 20 April 2006 Document name Author Courage Chibiya PrenatalLink Test Plan Revision 4 Date 20 April 2006 This document describes the test cases created for the PrenatalLink application to test the system up to the beta test version This version would be evaluated and subjected to further testing and validation by Media Innovations and eventually by external validators This test plan is based on the test requirements in section 16 1 above 20 1 1 Definitions and acronyms Term Acronym Description PI Program Input PI the test executed on the program PO Program Output PO the out put produced by the program when the test was executed Calculate A button on a page of the system as shown here Calculate 20 1 2 Features to be tested Implemented Feature Implemented Feature 1 Calculation Pregnancy Dating 7 Input Validation Disorders and Markers 2 Database Data Access and retrieval 8 Caching 3 User Interface 9 Reporting system 4 Security 10 Internationalisation 5 Help 11 Exception han
58. d for a covariable O Error on calculate Invalid maternal weight O A value is not supplied for a boolean covariable CovarID gt 5 O Correct calculation O An invalid value is supplied for a non boolean covariables gt upper limit O Error on calculate HHHH O Covariables used correctly in calculating the Normalised median O Correct median Covar factor applied or to raw median 20 Calculating the correct results O Results O Calculating correct results for markers O PrenatalLink results must be equal to manually calculated results Markers results Median Value used in calculating MoM Normalised value Normalised for gestation and corrected by covariable info Raw MoM Property Normalised for gestation but uncorrected by covariable info Twin Normalised value If the marker is fetus specific Twin Raw MoM If the marker 1s fetus specific 20 3 3 Disorders Parameter Data requirements Disorderld The required elModel records must exist for the given disorder For Combined disorder gt 2 disorders must be supplied Risk Factor A RiskFactorT record must exist for the given risk factor Marker Set amp The markers that are required in the given MarkerSet must be supplied Mes See the Markers test proposals above for the Markers data requirements Collection Collection not required when calculating prior risk only 100 MAEDD Valid age within the Disorder s age range Valid integer g
59. degree program The proposed solution involves considerable planning programming and report writing Furthermore extensive application of technologies studied will be required examples include databases internet distributed systems software project development and many others The developer will also use skills acquired during his degree such as team working and research techniques This assessment shows that the project is of substantial relevance to his degree program and will enhance or enable them to learn new skills and technologies which may be useful later in his career 1 9 Evaluation Criteria The success of this project will be judged on completion To ensure this can be done successfully a set of evaluation criteria was drawn up and these are 1 Does the system fulfil the project objectives and fulfil the project aim This criterion evaluates how well the system meets its objectives and aims The success of the system is based on how well the delivered features satisfy the project objectives 2 Have the specified minimum requirements been met and extended The second criterion judges how well the system satisfies the user requirements and how these were expanded upon MI have to benefit from the business value of the system 3 Does the system calculate correct results and function efficiently The third criterion will be assessed during the testing and evaluation processes Testing aims to validate the correctness of the
60. ders calculations The equations and parameters required to perform these calculations will be specified in due course The system shall provide a file download function that will enable users to download informative leaflets that Media Innovations publish The system must display informative message boxes informing the user of the progress of time consuming operations errors and system malfunction The system shall validate input data for clinical validity before calculation The system shall provide a response time of less than 30 seconds on average The system must provide the functionality to automatically generate the patient report in memory using a template PDF form and the calculated results The system shall only perform a post back when the user initiates a calculation The system shall generate friendly and useful error messages as explained in the usability guidelines The system shall hide the database access functionality and code so that it is not specifically stated in code The system shall cache all static objects such as images and non changing parts of the user interface The system shall follow the accessibility and usability guidelines of MIT and Nielsen respectively TI RC 18 RC 19 RC 20 RC 21 RC 22 RC 23 RC 24 RC 25 RC 26 RC 27 RC 28 RC 29 RC 30 RC 31 RC 32 RC 33 SE 1 SE 2 SE 3 SE 4 SE 5 The system must be tested for implementation errors and must be validated to beta
61. developed by MI and the Benetech PRA system The MI Screening calculator 1s still in use however this system is Windows based and cannot be used over networks It has notable batch calculation functions and has been clinically validated however it is not widely used by physicians It has disadvantages due to a lack of configurability and is quite complex to use because it lacks a results interpretation mechanism Benetech PRA 2006 is a widely distributed system in North America that uses the latest database technology and has a good user interface It does not include the latest markers and heavily relies on one marker alpha fetoprotein This makes its usefulness rather limited since results are more accurate depending on the breath of input information These systems are not network internet enabled and do not yet include the latest ultrasound markers Due to the drawbacks of the above systems MI felt that a new robust configurable and internet enabled system was needed to deliver the latest developments in prenatal screening Wilson 2006 17 4 System Requirements This section defines the high level functions and services that the system was expected to provide User requirements analysis produces descriptions of content functionality and quality required of the final solution The software specification defined here served as the basis for more detailed design and implementation Software system requirements are often classified as fun
62. dix D System Requirements scccccccssssssssscscccccssssssssssccccssccssssssscccssseeees 75 5 A Tm 75 5 2 Punctronal Requiretients iere pee ete o ce rts TI 6 IA E E RS 81 6 1 Database NO rita ISA OD TT 8l 62 Usability Gude NES e i ooi eer e did 82 6 3 A Do NS UM otaveua UA c pOMA IC MU 83 7 Appendix F Implementation ccce ecce ee e eee ee eee eee eee ettet eee eese esos et tete ees ssssssaoe 86 TZ Database 86 1 2 Son 86 7 3 NAM ON PETRO 87 7 5 Reporte TTT 88 5 Appendix E TesUlio nieta scarico dI 90 8 1 SS T O 90 02 West eT KT EL ALOIS cereo Und a 9 8 3 Calculator Test C as unida 95 8 4 Requirement vs Test Case Traceability Matrix ccccccsesssssssssessssssseseeeeeees 102 8 5 TES oo 103 9 Appendix E EV alain siii iniciaron Seele a deve Ep e Ca e Yeu Fa ERE SAUL Ee EE ERE EE aaa 104 9 Media Innovations Ltd Project Evaluation ccccccccssecceeeceeeeeeeeeeeeeeeeeeeeeees 104 10 Appendix G Product R C 2 2 2 2 eter eer er dre eee se Coetu t oppo dte repe e rena Vo eee pec saEpace 106 55 10 1 10 2 10 3 Installation A T TR RITE TA Implementation DOCU lE ie NE E E E RAE e User Guide 56 13Appendix A Reflection Like every other major project my final year project was both enjoyable and demanding at times I gained a lot from the experience from developing a big application and the responsibility that came with the solution I was implementing Dealing wit
63. dling system 6 Date picker 12 Installation 20 1 3 Features not to be tested Reason for non implementation 90 Batch calculation 2 File importation and exportation 20 1 4 Environmental needs The environmental needs are the software and data required to execute these tests on the system They are A settings database created in Microsoft SOL server database as explained in Appendix XX Product Content A web server with an installation of Microsoft Internet Information Services 20 2 Test case specifications 20 2 1 Data access Test Case Specification Database connection PI 6 1 Change connection string for settings database to an incorrect one before start up PO 6 1 Error on login Missing current data No current record PI 7 1 Try calculations with a required settings record missing PO 7 1 Error on Calculate 20 2 2 Security Test case specification Validate username and password login PI 2 1 On login user enters known correct login details PI 2 2 On login user enters incorrect details PO 2 1 Main screen displayed PO 2 2 User not allowed to proceed to main screen error message displayed Notes Shows the correct database is being used Notes All login details have to be correct otherwise login fails Missing login detail s login All login details have 9 PI 3 1 At login some login details are not provided or entered in the are compulsory prov
64. e cut down to a minimum and on the server to cater for non JavaScript enabled browsers In addition static data must be cached to improve performance If the batch calculation feature is implemented the system must provide a timely less than 24hrs response to batch transactions Operational Requirements The system cannot store patient data on the server due to the stringent data confidentiality and protection rules that have to be followed Non registered users must not be able to access the system It must validate user input in order to ensure that valid data is used in calculations although the user takes responsibility for the values that are used in calculations Maintainability and Support Requirements A document must be written to explain how the system was implemented The system must be written in a proper object oriented approach and reusable functions and modules must be created whenever possible Security Requirements The application must have some security features that will prevent unauthorised access and prevent access to internal resources Login details must be protected during transmission over the internet The system shall prevent access to users when they have logged out and shall not store confidential data on the users machine in accordance to confidentiality guidelines Project documentation The application s implementation documents must be written according to Media Innovations document writing guideli
65. e m SensibleName e g mTransactionDate For global variables use g SensibleName e g gLanguage o See Appendix for prefixes O 69 Using 3 party controls set properties in code rather than in settings pages easier to copy and modify Concatenating strings always use amp not Error handling in all procs unless they should be handled by the calling proc or in simple Property Get etc Method always use Handler at end of proc elipse 16 2 3 User Interface Overall organisation and navigation o MDI main startup form with ActiveListBar at left and Menu Bar at top o MDI size equivalent to full screen at resolution 800x600 allowing for Windows task bar o Child forms not sizable DemoBatch o Blank or background form when program group changed Think ease of use o ListBar organisation logical order of groups and items items in a group should be in some way comparable Don t duplicate ListBar functionality in menu bar Helpfulness of labels Tool tips Tab order O O O O Data entry fields o Date fields always display as Medium Date o Lists use combo boxes o All fields must be validated for data type range etc preferably on leaving field On any change in input on leaving field o Wipe all results and graphs o Don t process the new values wait for user instruction click on Calculate etc Consistency checks o Numerical fields decimal precision o Language terminology abbreviatio
66. e on the websites and products of MI Labelling must be concise grammatically correct and consistent appropriate terms must be used for labels Usability The usability of the application measures the efficiency and satisfaction with which users can achieve certain goals using the program interface ISO 13407 1996 The product must be enjoyable and uncomplicated to use The usability requirements of this system are based on the Massachusetts Institute of Technology MIT information services and technology usability guideline The MIT accessibility guideline can be found in Appendix F Internationalization The application must be designed to work in diverse locales it must show dates in the user s locale format The date entry control must display months in the user s language Likewise numbers must be interpreted and validated with regard to the user s locale Moreover the system must be usable from most widely used browsers Accessibility The product must be accessible to users with common disabilities The system shall use colour schemes that enable partially sighted and colour blind users to use the system The system must be designed to comply with the W3C accessibility guidelines reproduced in Appendix F 20 Performance Requirements The system must provide a reasonable response time of less than nine seconds over an average internet connection Input validation must be performed on the client to ensure post backs ar
67. e separators in a date field PO 43 1 Error on leaving field Invalid date PO 43 1 Date is accepted PO 43 1 Error on leaving field Invalid Date PI 43 1 The validation code correctly completes the year when the user enters a year of length less than 4 digits PO 43 2 System correctly auto completes the year if the user has entered less than 4 digits for the year using the 30 year rule cut off point 20 2 5 Help Test Case Specification Help file PI 43 1 Help file exists PI 43 2 Help file does not exist 93 default English locale numerical and date separators and the separators of the user s locale If an invalid date or number is entered an error message is displayed and the field 1s refocused If the length of the year 1s less than 4 the validation code auto completes it Notes PO 43 1 A Help file download box opens asking the user to open the document or save it PO 43 2 User informed that help file not found 20 2 6 Caching Test Case Specification Notes PI 43 1 Static components are stored in the internet cache of browser PI 43 2 Non Static components are never cached PO 43 1 All images are cached and stored in browser cache they show up more quickly from the second access onwards PO 43 2 Non static components always change showing that they are not being incorrectly cached 20 2 7 Reporting system Test Case Specification Notes PI 43 1 The generated PDF patient report
68. ealised to be useful if implemented in particular the security of the system could be made more comprehensive In addition the batch calculation functionality would be a welcome addition for some customers if it were implemented in future versions Overall the initial reaction from the company and the MI project supervisor was very positive and the high level of user motivation indicates that they are likely to adopt the system However the real test of success will be based on the number of users who will use the system on a regular basis and adopt it as their main prenatal risk calculation system although it is early to predict anything the prospects of the system look positive 50 11 Bibliography 1 Agar Michael 1996 The Professional Stranger An Informal Introduction to Ethnography Academic Press 2 Boehm 1991 Software Risk Management Principle and Practices IEEE Software Vol 8 January 1991 pp 39 44 3 Chen 1993 The Entity Relationship Model Toward a Unified View of Data ACM Transactions on Database Systems Volume 1 Number 1 pp 9 36 1993 4 Crump 2005 Risk Management http www saferpak com project management art2 htm 10 March 2005 5 Cuckle H 1997 Estimating a woman s risk of having a pregnancy associated with Down s syndrome using her age and serum AFP Obstet Gynaecology Journal 1997 Volume 94 387 402 6 Cunniff C 2004 Prenatal screening and diagnosis for paediatricians P
69. easurement Site Id Diabetes Ethnic Dating Date Days Gestation Smoking weight Donor Egg Maternal Weight VE Egg Extraction date Unit Previous Disorders Donor Age at Extraction Value Is Normalised Down syndrome Age Type Edward syndrome Donor DOE Operations Calculate Help Patient Report Results Report MH UE Forename Expected Date Marker MoM prior Risk P Surname of Delivery EDD Gestation Final Fisk required to Id Age at EDD reproduce results Fesult Based on patient Interpretation details and dating parameters Operations Print Export Operations View Patient Report Print and Export Figure 7 User Interface Fields And results as in the results report 30 The security calculation and results components of the user interface view layer will be implemented as web pages The fields contained in those components will be represented by text fields check boxes and drop down boxes Operations will be initiated by clicking appropriate buttons The patient report will be created as a PDF or HTML file The user interface will be implemented according to the usability and accessibility guidelines described in Appendix G Design 6 4 1 Colouring and Graphics The system will use a colour scheme that makes it consistent with current Media Innovations software products Colour is a very important attribute of the user interface and can improve usability and accessibility of the application Text will
70. echnology 3 4 2 Legal and Political Medical software faces legal and political challenges that impose constraints on what such software can do Compliance with legal requirements is crucial for the system to be feasible since non compliance is not an option The system must abide by the Data Protection Act of 1994 which states that systems must protect the privacy of data and ensure that it is up to date accurate and not misused The system will encrypt highly sensitive data and authentication details Users will be required to login before they can access the system In addition the web server is protected by a firewall system that prevents unauthorised access to settings data used by the application Besides the normal data management laws medical software is bound by laws that regulate the functions that such a system can perform The system must be clinically validated before it can be used to that end MI will use external evaluators to test the system Under medical law the system must only be accessible to registered physicians The final solution will only allow registered medical users who have applied to MI for login details Beside the normal health regulations that govern medical institutions and practices so far there 1s no legal provision for telemedicine Nevertheless the system will be developed with due consideration to medical law and other applicable legal constraints 3 4 Ethical Prenatal screening has controversial benefit
71. efore reading them This will not only save you time but will enable you to get quality information whilst saving you time Modules Since the project is worth 40 credits 1t requires a lot work and time I therefore recommend doing at least 6 modules during the first semester such that in the second semester you will have 3 modules to do This will give you more time to work on your project Write up Begin the project write up as soon as you start working on your project it 1s better to refine what you already have than to not have anything at all The write up takes a 57 substantial amount of time to do and accounts for a major portion of the marks so I found it useful to start the write up during the project conception stage Coding I found out that writing your code in a clear well documented manner will not only enable you test and debug the code easily but will make it easier to write the implementation chapter External Companies My project was proposed by an external company when writing up the report I was assigned a project supervisor to be my point of contact I felt that it 1s very important to take the users into consideration whenever possible They will be using the system therefore their needs must be satisfied by application In addition try to establish the user requirements as early as possible to make sure that your plans are appropriate for the project Project Supervisor and feedback O
72. em must use development technologies that are accessible to MI since they will be the main user of this system and will host it on their web server The product was also required to be written according to the Software development guideline of MI the guideline can be found in Appendix E Analysis 3 3 Stakeholders This project was proposed by Media Innovations MI a company that specialises in developing screening diagnostic and behavioural health software They were actively involved throughout the lifecycle of this project The main project stakeholders were MI and the end users The developer had contact with end users such as consultants and some clinicians who work for the Leeds Screening Centre Their active involvement was crucial in ensuring the application met user needs They were an invaluable source of advice and information In particular MI assigned the developer Mrs C J Wilson as the project supervisor to represent the company s interests in this project She acted as the project contractee and was actively involved in a overseeing the development of the final system The majority of end users are people involved in screening clinicians consultants and screening centres It is possible that researchers may use it to test certain assumptions and compare the system to other existing systems Prospective clients of MI might also explore this system to find out more about the company and its products 3 4 Feasibility Analys
73. equirements specification in Appendix F assigns unique identifiers to every requirement to allow the design code and test cases to be traced back to the requirement IEEE 1998 The use of the VORD technique to divide the application into separate view points was done with a view of ensuring that the elicited requirements were consistent and compatible not conflicting with each other In the case where independent viewpoints had identical or conflicting requirements such issues were discovered and rectified using a simple comparison of each viewpoint s requirements to those of other viewpoints It is assumed that by following the guidelines explained here and evaluating the system requirements as specified the risk of implementing a system that does not satisfy requirements will be mitigated 22 5 Methodology and Development Technologies This chapter assesses web system development methods and technologies which were considered for developing the final solution and concludes with the final choices made regarding which method and technology was used Methodology in this context refers to generic software development patterns that describe fundamental software process activities their timings and the results of each activity Development technologies are the platforms that facilitate the development of software applications 5 1 Development Technologies There are a variety of mechanisms by which a server can generate web content dynamic
74. error message when a user enters wrong login information so that the system does not give away information unnecessarily User interface UI The system shall provide a data input page that accepts information required to calculate pregnancy dating markers and disorders results The system shall provide a control for initiating calculations display help and results report The system must provide an internationalised client based date picker that enables users to enter to select dates in an efficient manner The system shall provide a panel where users can enter additional information related to a patient such as weight diabetes smoking and all risk factors The system shall display the names of markers and their units of a markers panel to enable users to enter valid values The system shall enable the user to specify the disorders for which they want results to be calculated for The system provide a default security page that will be presented to the user requesting their authentication details The system display a 1 in before all risk values The system shall display gestation in weeks and days and maternal age in years and months The system shall provide a link to the adobe website to enable users to download acrobat reader for use when viewing patient reports After calculating results the system must automatically display the calculated results The system shall display all dates in the DD MMM Y Y Y Y format the short mont
75. es and performance of the application is acceptable 9 5 Usability and Accessibility The system was designed to meet Nielsen usability and MIT accessibility guidelines The purpose of such guidelines is to ensure that the system provides an easy to use interface that is accessible and meets acceptable usability standards After implementation the system was evaluated to check that it met the requirements of these guidelines The evaluation checked every guideline against the system to see if the system complied with all guidelines The results of this evaluation showed that the system met all the guiding principles except for five guidelines which were not crucial or irrelevant to this application Appendix E shows the guidelines that were used to evaluate the system s usability and accessibility Additionally the system s user interface uses simple well known medical terms in labels With minimal understanding of prenatal screening a novice who has access to the correct clinical input data can efficiently use the system in a straight forward manner and with minimal user involvement 9 6 Portability and Maintainability The final solution is a web based application Having been implemented using ASP Net the application is installable on any web server and the software required to run the application is easily accessible at a small cost However any machine can access the application via a web browser The system can also adapt to different user
76. etarded population UKGTN 2005 and is highly costly to both patients and society The benefits in the economical sense far outweigh the costs From Media Innovations point of view a comparison of the development costs and the business value the project brings to the company and their clients shows that the development costs of the final solution are minimal The requisite development software is already available on their system and their server has all the required capabilities for hosting the final solution After installation running costs will only amount to the cost of the disk space and bandwidth that 1s used by the application In contrast the company can only benefit from the revenue that might be generated from this system It is therefore estimated that the cost of producing the final solution requires a small budget and developing the system 1s economically feasible 3 4 5 Operational The operational feasibility of the application 1s concerned with the viability of the day to day running of the final application The application uses one main resource that accrues costs on an operational basis The system will require an internet connection and this costs the company an annual fee of 250 However regardless of the existence of this application the company would still require internet connectivity and the introduction of this application does not increase operational costs Therefore at best the operational costs are minimal The sy
77. f immediate revisions are necessary increment Minor Increment Major for upgrades only 16 2 2 Coding Always check whether we have existing code examples or components before writing your own look in MI SRV MIDew Utilities Development tools packages o Databases Microsoft SQL Server 2000 Access 2000 ActiveX components Visual Basic 6 Win32 clients Visual Basic 6 Web clients Visual InterDev 6 Visual Basic Net Front ends for database configuration Access 2000 adp Visual Basic 6 Visual Basic Net Program structure o Think client server ensure that it will be easy to encapsulate the functionality in a dll etc No server functionality in forms Minimise repeated code use modules with shared subs and classes Hardcoded values use constants Think DCOM client server and database may be on different PCs Pass values ByVal etc O O OQ O O O O O Providing objects classes o An object should generally contain data and functionality that together represent some entity real world object related group of tasks etc o Public properties should be general to the class Input data for a particular sub should be passed as parameters conversely data required by more than one sub should be public properties o f your class contains methods and functions only no properties then it should probably be a standard module unless obviously it is part of an ActiveX interface 68 o f your class is just
78. f the date to inform them of the selected date 48 The security system of the application could be further improved so that all sensitive information transmitted between the client and the server is encrypted or protected in some way Currently the application only encrypts authentication details during transmission from the client leaving clinical information vulnerable to electronic eavesdropping In future implementing such a feature would be a useful improvement The current help system is not adequate a search utility that would enable users to enter a keyword and display the relevant help topics would make help more accessible It would enable the user to quickly find the information they are looking for The application validates numeric input and dates using the user s locale formats However non numerical fields are not validated it would be useful to implement code that validates non numerical input such as names in future versions The application performs basic validation numerical and date validation on both the server and the client this 1s done to cater for clients whose browsers are not JavaScript enabled Ideally the application should be able to do more complex validation on the client side as well checking that values are within acceptable ranges as 1s done on the server would be more efficient if done on the client where possible This would prevent unnecessary delays because a user would be informed of an invalid
79. g affected by a given disorder This project satisfactorily achieved that objective however there is still room for improvement The developer aimed to deliver functionality first before embarking on other secondary aims of improving the usability of the application The usability of this project was based on the Nielsen s usability guidelines The application was a success in terms of usability according to those guidelines because throughout the implementation stage newly developed components were analysed to identify possible means of improving usability Users and Media Innovations Ltd were fully involved in the development process and this meant they had a sense of enthusiasm about the project due to their involvement After implementation the application went through testing and validation processes These processes evaluated the correctness of the results calculated by the application The testing process was made up of several cycles of error discovery and resolution On completion of testing the user carried out acceptance testing to check 1f the program met their requirements Because the user had been actively involved throughout the development process acceptance testing was a secondary procedure for the user to perform a final check on the system This project achieved the set objectives and requirements although there is still room for expansion Some functions which had not been thought of during requirements elicitation were r
80. h name must be in the current user s language format The system s date picker must display full month names in the users language The user interface must be consistent with other software products of Media Innovations 79 UI 15 UI 16 UI 17 ID RE 1 RE 2 RE 3 RE 4 RE 5 RE 6 RE 7 ID HP 1 HP 2 HP 3 HP 6 HP 7 HP 8 HP 9 HP 10 The user interface must be labelled using correct and easy to understand English Controls on the user interface must be laid out in a clean and well formatted manner Numbers must be formatted to two decimal places before display Reporting RE The patient report must display user details to enable identification of the physician who calculated the results The patient report must show all the input data entered on the calculation form and must also show the results calculation date The patient report must have a disclaimer with a message that tells the user that the responsibility for entering valid data lies with them The patient report must show the version number of the system The system must display results on all reporting forms and documents in a consistent manner The system must produce the report in PDF and HTML formats The PDF patient report must be printable on most printers it must have the default A4 page dimensions Help Requirement HP The system shall provide a help system that explains how to use the system The system shall provide a context sensiti
81. h prospective end users and other stakeholders really widened my people handling skills and enlightened me to the industrial world This project enabled me to learn many skills technologies and improved me in many ways the paragraphs below explain ten different lessons that may be useful to other students undertaking a final year project in the future Plan before doing things taking a moment to organise things imagine and analyse possible solutions to a problem is very important Doing so will help you to do things using the easiest and time efficient approach Plan before doing most tasks write up detailed plans for major tasks and think up smaller plans for mundane tasks before starting to do them In addition planning enables you to do things in a structured manner that makes in easier to manage your project and ultimately enable you to complete it successfully and on schedule Project schedule Create a reasonable project right at the start and stick to it Spare adequate time for completing a task before starting another one Take note that when writing up and implementing the solution problems do arise and they might take time to solve so leave some unallocated time between some tasks to ensure that you don t fall out of schedule as soon as a delay happens Background reading Although it is important to read a very wide variety of books for background research purposes take some time to validate the relevance of articles b
82. he STRIDE acronym Microsoft 2005 will be used to categorize threat types Spoofing Spoofing refers to users gaining access to the system using a false identities gained by stealing user credentials Tampering Tampering 1s the unauthorized alteration of stored data or data in transmission over a network between a client and the server Repudiation Repudiation is the ability of users legitimate or otherwise to deny that they performed specific activities or transactions 3l Information disclosure Information disclosure is the unauthorised revelation of private data For example a user might view the contents of a table or file he or she is not authorized to open or monitors data passed in plaintext over a network Denial of service Denial of service is the process of making a system unavailable This threat might be accomplished by bombarding a server with requests to consume available system resources or by passing it malformed data that could crash an application process Elevation of privilege Elevation of privilege occurs when a user with limited privileges assumes the identity of a privileged user to gain privileged access to an application 6 6 1 Countermeasures For most threat categories described above the security system will implement countermeasure techniques that will mitigate or reduce risk where possible Thea 000000000 Countermeasure to be implemented Spoofing user identity The system wi
83. he complexity of results interpretation from the user and uses a more improved database that is normalised to third normal form Benetech PRA 2006 is a good system with an advanced data storage system and good security However it 1s rather limited in terms of the number of markers that can be used in calculations This means the final solution is better since it enables users to use the most suitable marker collection for their purposes In addition the current database system of final solution is adequate for the current market needs In conclusion the final solution gives added benefits to users it meets the needs of the market that were identified in the project summary The system provides an improved quality of service than most of the existing systems and delivers functions that utilise some of the latest developments in prenatal screening 9 8 User evaluation The MI project supervisor carried out a system evaluation on completion of the beta test version The evaluation was split into three sections the first analysed how well the system satisfied user requirements The second described the tasks that were required to create a satisfactory system and the last section was an evaluation of the system in terms of performance usability and efficiency Mrs C Wilson the MI project supervisor compiled a user evaluation document which was reproduced in Appendix F Evaluation The user evaluation showed that MI was satisfied with the final soluti
84. his project The supervisor was actively involved in the all stages of the project lifecycle final user acceptance testing and user evaluation 1 3 Project Aim The main aim of this project is to develop a web application that will accept maternal biochemical and ultrasound test results from clinicians and calculate the likelihood of the given pregnancy being affected by congenital disorders especially those caused by chromosomal anomalies 1 4 Project Objectives The objectives of this project are as follows Research the theory of screening and the causes of congenital disorders such as genetic malformations and evaluate the benefits of prenatal screening Evaluate existing prenatal screening systems noting any useful features that could be implemented into the final solution and weak points to capitalise on Elicit requirements from stakeholders and develop a requirements specification Evaluate software development methodologies and technologies that could be used to develop the application and choose the most suitable ones with justification Design and develop a software application that meets at least all the minimum requirements and some of the requirements extensions 1f possible Test and evaluate the developed application against a test plan and the evaluation criteria This project aims to produce a beta test version of the final system further independent evaluation and testing by MI and external evalua
85. ided fields PO 3 2 Not allowed to proceed to main screen Error on login 20 2 3 Date Picker Test Case Specification Notes PI 43 1 Date picker shows the correct days for the selected month Test dates and year test for leap years normal years and all months 31 02 2000 01 12 2006 PO 43 2 The Date picker displayed the correct number of days for all 27 02 1996 months in leap years or ordinary years PI 43 1 Date picker pastes the selected date into the date field correctly Try 5 different dates PO 43 2 The Date picker reads the selected value and correctly saves it in the appropriate date field PI 43 1 Date picker user interface works correctly and responds correctly to user interactions PO 43 2 Based on the test performed above the date picker responded well to all clicks and other user interactions 20 2 4 Input Validation Test Case Specification Notes Numerical validation The program PI 43 1 User enters an invalid number in numerical field validates dates and numbers using the 92 PI 43 1 User enters an valid number in numerical field PI 43 1 User enters an number with invalid number separators in numerical field PO 43 1 Error on leaving field Invalid number PO 43 1 Number is accepted PO 43 1 Error on leaving field Invalid number Date validation PI 43 1 User enters an invalid date in date field PI 43 1 User enters an valid date 1n date field PI 43 1 User enters an number with invalid dat
86. information or functionality and is updated as often as the inaccessible original page 84 Priority 2 Checkpoints In General Priority 2 YesNoN A 2 2 Ensure that foreground and background color combinations provide sufficient contrast when viewed by someone having color deficits 3 1 When an appropriate markup language exists use markup rather than images to convey information 3 2 Create documents that validate to published formal grammars 3 3 Use style sheets to control layout and presentation 3 4 Use relative rather than absolute units in markup language attribute values and style sheet property values 3 5 Use header elements to convey document structure and use them according to specification 3 6 Mark up lists and list items properly 3 7 Mark up quotations Do not use quotation markup for formatting effects such as indentation 6 5 Ensure that dynamic content is accessible or provide an alternative presentation or page 7 2 Until user agents allow users to control blinking avoid causing content to blink 1 e change presentation at a regular rate such as turning on and off 7 4 Until user agents provide the ability to stop the refresh do not create periodically auto refreshing pages 7 5 Until user agents provide the ability to stop auto redirect don t use mark up to redirect pages Instead configure the server to perform redirects 10 1 Until user agents allow users to turn off spawned windows do n
87. ion The marker value is not normalised and the gestation value amp type are not supplied Error on calculate Missing or invalid gestation information The marker s value is normalised and the gestation value amp type are supplied Correct calculation since gestation values are not required in such a case The marker s value is not normalised and a gestation value amp type are not supplied Error on calculate Missing or invalid gestation information The markers value is normalised and the gestation value amp type are not supplied Correct calculation since gestation values are not required in such a case Twin Gestation O Twin value not given but normalised and gestation value amp type are not supplied O Error on calculate Missing or invalid gestation information O Normalised value and gestation value amp type are supplied O Correct calculation Weight O Weight not supplied O Correct calculation O Invalid weight supplied e g weight lt 0 O Error on calculate Invalid maternal weight SiteID O SiteID not supplied 99 O Correct calculation if default records exist O Invalid SiteID supplied 0 O Error on calculate Invalid SiteID 7772 19 Covariables Invalid CovarID supplied lt 0 O Error on calculate g O The same covariable is submitted more than once O Error on calculate Unable to add element to collection existing element with same key O A value 1s not supplie
88. ion Error on calculate Marker s not supplied Only the required markers in the marker set are supplied Correct calculation Extra markers that are not in marker set are supplied in markers collection EN o BM O BN o BC Same results as when only the markers in the marker set are supplied extra markers are not used PrenatalLink must use only those markers that are in the specified marker set O Non required markers that are in the marker set are supplied in markers collection 101 O BM o H Correct calculation all markers that are in the marker set are used A marker set for NT has AFP only and AFP is supplied in markers collection Correct calculation NT markers sets must have AFP only A marker set for NT has AFP and other markers An NTD marker set can have only AFP Error on calculate Invalid marker information Calculating correct disorders results 2 28 29 O O Prior Risk Posterior Risk Results Sample calculations Calculating correct result compare PrenatalLink V3 results to PrenatalLink V1 results for the same settings and input The results must be correct for the Posterior risk result Calculating the correct result in relation to risk cut offs truncation limits gestation limits and twin results cut offs The calculated results must be correct for the input data 20 4 Requirement vs Test Case Traceability Matrix Requirement UR RESI UR RES2 UR RES3 UR RES4 UR SOUI
89. is The feasibility study was integral to this project it was important that the system was shown to be theoretically practical before commencing on design The feasibility study analysed how the system would operate under different assumptions or constraints and to infer whether the system was viable In order to evaluate the project s potential for success the feasibility study focused on five main sections Technological Legal and Political Ethical Economical Operational and Scheduling 14 3 4 1 Technological This section analyses whether it is possible for the system to be successfully implemented given the current technology The system will be hosted on a web server running either SQL Server or Microsoft Access Both database management systems are satisfactory for the data storage requirements of the application The final solution will be developed on a platform that has language facilities for developing web applications Of the current technologies that are available most are either free or already installed on MI machines The MI web server 1s connected to the internet via a gigabyte internet connection Therefore it will be able to provide the bandwidth required to satisfy clients using the application The web server has about 150 gigabytes of disk space enough disk space to store the settings data required by the application Therefore the developer concluded that it was possible to develop the final solution using the current t
90. ith a camera into the uterus to observe the fetus 2 3 2 Non Invasive Techniques Non invasive methods involve taking fluids blood or urine from the mother for analysis or using external diagnosis mechanisms Ultrasound The most common method of prenatal screening is ultrasound testing It uses ultrasound waves high frequency sound waves to produce an image or sonogram of a fetus In addition some other measurements can be carried out for further analysis ultrasound screening is explained in more detail in section 2 5 3 Maternal Serum Maternal serum screening involves measuring biochemical markers present in the mother s blood during pregnancy Haddow et al 1999 Abnormal levels of these chemicals may indicate increased risk for certain birth defects and genetic diseases Physical examination of a pregnant woman s stomach may reveal the presence of an abnormality to an experienced physician 2 4 Reasons for Prenatal Screening Knowing that a child will be affected by a disorder enables parents to plan for birth and receive counselling Clinicians may also prepare for the birth and have necessary treatment ready Medline Plus 2006 Diagnostic tests such as amniocentesis are expensive and risky screening gives a means of deciding which pregnancies warrant further investigation Prenatal screening eradicates illness pre emptively and in doing so it has altered fundamental society attitudes towards parenting and
91. ject Stepanek 2005 To that end the developer compiled a user requirements specification that was reviewed by the user before design was started During the early stages of the project the developer negotiated with the MI project supervisor to establish the main requirements of the system such that the project s scope 13 could be established as early as possible and for drawing up the project schedule The project schedule was a useful tool for project management and progress monitoring 3 2 Constraints Constraints are items that by their nature restrict the choice of solutions and constrain the scope of a project Various limits and constraint exist in a project development lifecycle due to financial environmental legal and other factors Schwartz 1997 Consequently it is crucial for such constraints to be evaluated before the development of an application commences to ensure informed decisions are made This project was developed for an external company who along with other stakeholders were likely to impose various requirements and constraints on the project These requirements had to be fulfilled in order for this application to satisfy their needs The product would have to interface with existing software a dynamic link library that encapsulates the clinical rules for risk likelihood calculation The final system is scheduled to be finished by the end of March 2006 a month before the deadline for project submission The syst
92. l aEHEEEEEEEE wemmmemim HHHH A PA Technical documentation mE E Pl NN Userdocumentation E A REN 11 1 11 3 Testexeontion LLL T H Op PI EN 1 A Wee A AA kd T mEn H Hd UE 8 Schedule Gantt Chart Key Work in progress Exam and holiday period 62 16Appendix C Analysis Technical specifications are typically written to describe how a project will be implemented They are created from the functional specifications 16 1 Risk Register Risk Register Risk Likelihood description amp high Requirements instability Incomplete Requirements aims are clear Unclear requirements Infeasible requirements Hardware constraints capacity limitations incompatible development Meeting human needs Unsuitable development environment low medium medium Low the main Risk Management risk reduction plans Try to establish main requirements as soon as possible Considerable time and effort will be set aside for eliciting documenting and reviewing requirements The existence of a project supervisor contractor means clarification will be available when required The development platform limits what can be done A feasibility study of the available technology will ensure only feasible tasks are carried out The application will be regularly tested
93. l model such that the chapters of this report will mirror the stages of the Waterfall model 12 3 Analysis This section will analyse the general issues concerning this project s development and will evaluate its feasibility and associated risks It aims to model the system in real world terms taking into account the project s constraints It analyses the project constraints and development issues and describes the results of the feasibility study in detail Furthermore alternative systems will be evaluated with the aim of identifying ways of improving this application and to ensure that the assessments of those systems are taken into consideration before embarking on design This analysis is performed to ensure the project is feasible and to pinpoint the requirements constraints and scheduling needs of the project and maximise its likelihood of success Requirements analysis cannot start until a statement of the objectives scope and feasibility of the project have been established 3 1 Considerations Successful projects are those that increase business value and satisfy or exceed user needs They are delivered on time and on budget Before commencing analysis the developer took some time to reflect on the issues and problems that have affected other projects both negatively and positively Considering such issues is important because as Stepanek 2005 explains doing so can help the developer learn from others mistakes and good practices
94. layers as shown in figure 4 After this initial analysis appropriate design models were chosen for designing the system components in the MVC layers The Model layer components were to be designed using an object oriented model because the results calculation objects represent real world entities and are particularly well suited for such a model 21 6 2 Architectural Design Once design patterns had been established the first stage of the design process was to decompose the system into a set of interacting subsystems Figure 5 is a structural model of the architecture of the system based on Figure 4 that shows the system s components in their respective MVC layers Client Side System Event Handling Reparting System D System View Model security Calculation Data Access Controller system System System Figure 5 System Structure 6 2 1 Model Layer Design The design of the system components that belong to the MVC model layer was carried out using an object oriented approach The objects in this layer carry out calculations and data access Input Validation Calculate PregnancyDating strForename arrionths PreznancyDating datingProperties strsurname arrDays Markers imtld chrseparators Disorders cetsettinesiPregD Ly dtescanDate 2 Calculate intGest amp tscan validateDates Lg zetGestations setResults dteMaternalDOB WalidateN ums zetMoMs intNumF ets cetDisorders elipse M3 intWweight fillcambaboxe
95. lementation unit and integration testing the performance behaviour of the system was checked to ensure that response times were acceptable and that the system could cope with normal levels of usage and data In addition the system underwent tests under extreme conditions such as high numbers of users and unexpected input data All tests checked that the system produced correct results and that the system s exception handling system produced the 41 correct error messages The security system was also tested to make sure unauthorised users could not gain access to the system The system testing section of Appendix I Testing describes the system test cases and their results Acceptance Testing User acceptance testing encapsulates the test procedures that result in the formal acceptance of the system End users of the system were actively involved in this test The test plan for acceptance testing was primarily based on the user requirements to which the system must conform User Acceptance testing was done to scrutinise the performance of the system in real world conditions and to ensure that the user is satisfied that the system meets their requirements and performs at an acceptable level The MI project supervisor was responsible for accepting this software solution therefore she was the appropriate person to carry out user acceptance testing A user acceptance testing report written by the MI project supervisor can be found in Appendix F
96. ll use strong authentication and will not transmit login details in plaintext Tampering with data Data hashing and signing are possible solutions for this problem but due to performance and project constraints this solution was deemed too complex to implement in this version Repudiation The use of strong authentication and encryption means that users can be identified easily However the repudiation threat does not have harmful consequences for this system Information disclosure Login details will be protected by encryption however takes considerable time processing power and bandwidth to encrypt all the other information transmitted to the client Denial of service The current server system have resource and bandwidth throttling techniques for overcoming this threat and the system will validate and filter input to protect the application Elevation of privilege The principle of least privilege will be used when accessing resources and the server s security system will prevent external access to the internal file system Table 5 Threats and Countermeasures 6 7 Database Design The entity relationship modelling technique was used to create the structure of the database An entity relationship diagram is an illustration for describing high level abstract data models using a graphical notation Chen 1993 It contains four key elements entities attributes identifiers and relationships Entities represent real world
97. ls across a wide range of industry and professional sectors It is also a significant developer of Training Systems Risk Management and Diagnostic Support Tools to the medical profession Isaacs 2004 Some of the software products that are developed by MI are based on research projects of its own the University and other consultants Its market is made up of the general public healthcare professionals prenatal screening centres and other companies such as Perkin Elmer Ltd one of the leading developer and distributor of healthcare software in the world The MI range of products can be divided into two categories multimedia and scientific software products Most of the multimedia software is designed for use as medical self help training materials while the scientific software products are principally medical risk management and diagnostic support tools 1 2 1 Organisational Structure MI is made up of three departments Scientific Software Multimedia and Textiles departments The Multimedia department specialises in developing self help training materials The Scientific Software department develops medical risk management software and diagnostic support tools The textiles 2 department develops self learning products and e learning systems on various textiles and apparel subjects This project was proposed by the scientific software department of MI The department assigned a project supervisor who was responsible for overseeing t
98. lues population marker median of the testing laboratory Correct analysis of the different biochemical marker results depends on knowing the gestational age precisely and proper adjustment of the marker values for Covariables Covariables are pregnancy related factors of the mother such as weight smoking and diabetes For example weight gain indicates good fetal health therefore marker values have to be adjusted for these factors to enable correct regression analysis Posterior risk 1s also affected by other risk factors Examples of such risk factors include previous T21 or T18 positive pregnancies These factors may increase the likelihood of a pregnancy being affected by a disorder For example a previous positive Downs or Edwards syndrome pregnancy may increase the risk of a new pregnancy being positive due to the increased risk of recurrence if a previous fetus was affected 2 5 3 Final Risk The final risk value is calculated from the posterior and age risks a constant risk cut off value is then used to determine whether a pregnancy is positive or negative and if positive further definitive tests are then carried out Definitive prenatal diagnosis 1s then carried out using invasive techniques such as those discussed in section 3 3 2 2 6 Project Methodology In order to solve the problem of this project the developer considered various methodologies for software development These methodologies are generic models which give useful abs
99. mented in the Calculate class and is supported by three other classes Pregnancy Dating Disorders and Markers and the Elipse object An SQL server database was created using the schema described in section 8 6 and database classes Connect and DBView were implemented to handle data access The frmCalculate web page was created using ASP Net web forms designer to enable data input A security class and login page were implemented to enable user authentication and for protecting login details during transmission A help was developed and can be accessed from the calculation page A JavaScript html based control was implemented to enable efficient entry of date values The application uses random session keys to track users and validates dates and numerical fields to ensure appropriate values are entered and to remove any code or injected SQL commands All images and static content are cached on the client side in order to reduce response times and bandwidth usage The application is usable in Internet Explorer Netscape Fire fox and Mozilla The application displays dates in the user s language and validates both dates and numbers according to the user s locale formats A help system was developed and shows short messages explaining how to perform the current task Figure 11 Requirements vs Features Delivered 38 7 5 2 Features Not Delivered Figure 12 describes the requirements extensions that were not implemented in the final s
100. mosome Men with Klinefelter s syndrome are usually sterile and maybe dyslexic Medline Plus 2006 Spina bifida malformation of the spine resulting in a spinal opening Effects include paralysis nerve damage and bowel bladder complications GM Stocks 2005 Ancephaly a condition that results in incomplete brain formation its effects include missing brain hemispheres and life expectancy for affected infants 1s less than a few days 2 3 Screening and Diagnostic Methods Screening and Diagnostic methods are the techniques used to detect fetal abnormalities These methods can be subdivided into invasive and non invasive procedures Invasive methods give more conclusive results and involve elaborate risky and penetrative procedures for diagnosis Non Invasive methods estimate the likelihood of the abnormality and are more prone to error 2 3 1 Invasive Techniques Invasive procedures involve inserting a needle guided by ultrasound into the uterus abdomen or other internal parts of the body and withdrawing a small amount of fluid for cell analysis Amniocentesis A procedure in which a needle guided by ultrasound is inserted into the uterus and withdraws some amniotic fluid for cytogenetic cell analysis and diagnosis Chorionic Villus Sampling CVS A small amount of villi hair like fringes of the placenta is removed using a needle inserted into the abdomen Cunniff C 2004 Embroscopy Involves putting a probe w
101. n it reduced the number of errors that were likely to be encountered during later testing stages Integration testing was carried out thereafter the main focus of this test was to prove the compatibility of a given unit to the units it would interoperate with These tests enabled the developer to discover integration errors when the knowledge of the likely cause of the error could be easily discovered and to reduce the number of post implementation errors and interoperability issues System testing was of great importance to the evaluation of the system It emphasised on checking the correctness of the results calculated by the system and verifying that the system was working correctly interpreting user input and settings data correctly The system calculates correct results based on the comparison of the results it produces to the results generated by the MI calculator The pregnancy dating results that it generates were verified by comparing them to manually calculated results Several test cases were executed to test the performance of the application in extreme conditions such as large numbers of settings records low bandwidth and large numbers of users The results of these tests showed that the system was reliable and could cope in extreme conditions as demonstrated in Appendix E In addition the system was subjected to invalid input tests to check how well it recovers from errors or exceptions Exception handling tests results Appendix E demons
102. name FFF la po EDD 2 Oct 2006 EDD 23 82 Previews Prior Risk Print Risks Final Risk Result Final Risk Gestation 2nd MoM 2nd Gestation Twine Result Figure 14 Results Page 40 8 Testing Testing was conducted throughout the implementation process this chapter explains the validation or verification of the system Validation tests check whether we are building a system that meets the user requirements Verification test cases check the code to ensure 1t works correctly The software was subjected to unit integration and system testing unit testing checked every unit of the system to ensure that 1t worked as expected while integration testing checks whether a collection of units interoperate correctly At the end of the testing process system testing was done to check whether the system was acceptable to users and assess how 1t coped in various conditions During the analysis and design stages verification test procedures were used to review the deliverables of each completed stage The reviews were conducted in conjunction with the MI project supervisor The specification review would check that the specification document captured all the user requirements and that the requirements were feasible 8 1 Code Testing Two main techniques were used to test software units white box and black box testing In White box testing test cases aimed at checking code fragments such as loops algorithms and c
103. nes In particular documents must have version addition control systems and must be written to a very high standard of grammatical correctness and consistency Software documentation must be written during the development process to ensure details are not overlooked or forgotten A list of tasks to be done must be maintained for every document in order to ensure necessary additions are always done Product documentation can be found in Appendix G Product Content Legal Requirements The system shall fulfil the validation requirements specified for medical software It must comply with the Data Protection Act of 1984 medical and civil software regulations Reliability and Availability The system shall be available for use 24 hours a day and operate independently without need for regular human intervention 4 5 External Interface Requirements The final solution will provide a web form based user interface These web forms will have controls that enable users to enter clinical data calculate and view results This section describes the main components that the user interface will have The system is required to provide a graphical user interface that will allow users to Login securely 21 Input and calculate single patient results Import and calculate batch multiple patient results View and save results preview and print reports and results Access and view help In addition user controls must be implemented for a
104. ng the application and making it customise to the user s locale and display dates in the user s language 1 6 Deliverables Deliverable name Description Delivery date Mid term report A progress report and background chapter 09 December 2006 User manuals A user guide describing how to use the system 18 March 2006 Technical documents Documents explaining the system s components and 07 March 2006 other technical implementation information Application Elipse The final product and the database used to store 01 March 2006 and Database clinical data for use in calculations Testing documents Document describing the validation and testing 25 March 2006 processes of the system Final project report A report on this project that documents the project 25 April 2006 lifecycle Table 2 Project Deliverables 1 7 Schedule A detailed project schedule is included in Appendix C Introduction this section gives an overview of the project schedule Tasks Implementation Period Relevant research background reading and planning 27 Sep 2005 01 Jan 2005 Report Planning and Write Up Incremental 12 Nov 2006 15 Apr 2006 Inception and Requirements Analysis 13 Nov 2005 26 Nov 2005 Methodology and Design Specification 27 Nov 2005 15 Dec 2005 Implementation 15 Jan 2005 24 Mar 2006 Verification and Validation 05 Mar 2006 18 Mar 2006 Review and Evaluation 19 Mar 2006 24 Mar 2006 Release Packaging amp Report Finalisation
105. ns o Fonts o Colours o Control size and alignment especially textboxes 16 2 4 Locale Independence All code must work correctly in any Windows locale General o Program components must receive parameters properties in the correct type dates double integer etc and NOT variant o Don t pass a date strings directly to a recordset date field instead convert it using CDate first and pass it using the MIUtils sqlDate function o Use formatting to convert a date number into a string when displaying dates and numbers o Dates and numbers must be stored in variables of the correct type o Use validation see the user interface section to prevent invalid values being passed to a conversion function the source value must be valid for the destination data type or an error occurs o Most VB Net formatting and parsing methods have the ability to dynamically change based on the current culture Always make sure a method has such capabilities before using it and if it doesn t you will need to send it Culture and formatting information to ensure the correct formats are used Dates o Always use named formats when formatting dates for display Use the MIUtils DisplayDate function whenever possible 70 Numbers Avoid storing dates as strings in your code Enter dates in code in the format month day year to ensure that the date will be interpreted correctly in any system locale The System Globalization namespace contains classes th
106. nt are handled automatically Perl is excessively complex and compact once code is written it is very difficult to understand it seems Perl has more than its fair share of terse and complex language constructs Perl does not enforce data security access to private data is not prevented by convention 5 1 4 Python Python is a dynamic interpreted object oriented programming language developed as an open source project and managed by the Python Software Foundation Python is easy to use due to its clean and simple language constructs It is free and highly portable like Java It 1s an object oriented language that supports important modern language features such as garbage collection and exception handling Python 2006 24 As a new language Python is not widespreadly used scripts may not work on some machines and as a rapidly developing language it is highly unstable 5 1 5 PHP PHP is a scripting language that depends on pre processing of HTML tags When a server s PHP pre processor encounters PHP language tags it invokes the PHP engine to execute the code PHP 19 free since its open source software there are no upgrades or licensing fees PHP works with the APACHE web server which has a proven track record of reliability and good security Hull 2006 It is a cross platform technology that can operate on Linux Windows Solaris and other UNIX platforms PHP produces shorter code as a result PHP pages use memory efficientl
107. ny reusable graphical component to enable efficient date entry and page navigation The user interface must be adaptive to user locales and display dates and numbers in correct formats for the user s locale Software Interfaces The application will interface with an existing software component that encapsulates the rules for calculating marker results The system is required to communicate with the external software components by passing data in the correct format and data type 4 6 Requirements Evaluation The International Standards Organisation ISO 9126 1991 specifies 6 quality evaluation criteria for software requirements In theory good requirements are feasible clear verifiable complete consistent and testable Essentially requirements must be testable this is important for validation purposes However some requirements are not testable by nature and for those requirements other evaluation criteria were used to evaluate those requirement is the Testing chapter A test plan was drawn up based on the functional requirements the tests cases in the test plan will be used to check the functionality of the final solution Additionally the test plan will serve to prove the testability and verifiability of the system requirements The test plan can be found in Appendix E Testing If a requirement cannot be tested or is infeasible it will either be eliminated or scaled down to a more attainable specific and testable requirement The r
108. o B function encipher f var s f stream value var k salt f key value r k b encoding buffer var p 0 n Math floor s length b cl while n gt ptt I self status Encrypting block p n r encrypt s substr p 1 b b f key value Ln C f stream value E self status Done un 86 19 3 Validation function validateNumbers identity separatorz this method formats the entered string by removing characters fand extra commas i e separator separ var theform document Forml the patients page var Control document cqetkElementById identity edited textbox var validchars lz34567582 4 separators faccepted values var thousandiep separators char t D client s thousands separ var decimalSep separators char t l client s decimal separ var EnteredString Control value the entered numerical string var validWNumber true fboolean var for validating user input var num fbecSep number of decimal separators in string if EnteredString return end if Enteredatring Enteredstring replace Ittra T remove invalid characters from the entered input for var i 0 i lt Enteredstring length i var letter EnteredS5tring charAt i toLowserCase the char at position i if letter thousandSep Enteredatring EnteredString replace letter l end if if validchars index f letter 1 validMumber false
109. o what quality it satisfies user requirements aims and objectives 9 2 1 User Requirements Table 7 evaluates the quality of the functionality delivered by the final system and the extent to which 1t satisfies the main requirements of the solution Requirement How it was achieved Web application During implementation and testing tests were carried out to check the for calculating performance of the system s components The system responds in real time to risk results most functions that are executed on the client side This means that the quality of the system s performance is very high in comparison to existing systems The system can be easily used over a network enabling a screening centre to monitor all operations from a single server machine The system delivers good usability and minimal understanding of prenatal screening is required to use the system The system delivers more by means of a multi stage results reporting system that interprets the results for the user giving them more valuable information Database system During implementation a database was created for storing settings data The for data storage database was normalised to third normal form which means the database organises data efficiently and reduces the potential for anomalies during data editing The database organises data in a consistent manner as a result and allows quick data access and updating with minimal risk of causing inconsistencies Data inp
110. ogin first before they can access the system If they successfully login they will be transferred to the calculation page On the calculation page users are required to provide valid clinical 39 information in the fields of the calculation page This information is the used to calculate Pregnancy Dating Markers and Disorders results User input is validated immediately on leaving a field and if the input data is invalid an error message is displayed to notify them of the error On the server more comprehensible validation is carried out to ensure input values are clinically valid and within acceptable ranges Only when the provided information is valid does the system proceed to calculate results and generated patient results reports 7 7 Reporting System The application produces Results and Patient reports in PDF and HTML formats When a user calculates results a results page is automatically displayed showing the calculated results and the user can request a PDF version of the report which they can print save or open on their machine The figure to the left shows a flow chart illustrating the routine for displaying reports The Patient report is generated dynamically in code and downloaded to the client Appendix H explains how a PDF report is generated in memory Figure 14 below shows a screenshot of the Results page see Appendix H for a sample patient report Patient Markers Gestation Twinz Surname ss Fore
111. olution and gives reason for their non implementation Feature Requirement Reason for non implementation Batch calculation system that Storing patient data would bring considerable complexity and allows users to calculate results security demands on the system It would also put a huge strain on offline server resources so it was decided not to implement this feature Import and export functions for This feature was no longer required since the batch processing file upload and download feature was not being implemented Figure 12 Extensions Not Delivered In summary all the minimum and additional requirements were met except for the batch calculation features that were not implemented 7 6 How the System Works Exception handling is done whenever a method is executed and an error message is displayed when an exception happens The inbuilt try catch construct of ASP Net was used to handle exceptions Security C Start gt Ser accesses the syste VIa a browser Server responds by sending the Login page User enters Login details cs Login details are encrypted and sent to server Connect Class used Server Lozin details are decrypted amp checked To retrieve user details user s login details server tranzfers User to the calculation page Calculation Display Error Message Request cor rect login details Figure 13 System Flow Diagram Users have to l
112. omalies Smith Bindman Study 2001 2 5 2 Posterior Risk Posterior risk 1s calculated from the results of clinical tests carried out on the mother s blood to measure biochemical markers ultrasound markers results and other pregnancy related factors The most common biochemical markers used in maternal serum screening are 10 Alpha fetoprotein AFP a protein made in the liver of the fetus In neural tube defects the fetus skin is not closed so large amounts of AFP leak into the mother s blood Low levels of AFP in the second trimester 14 26 weeks gestation may indicate a T21 positive pregnancy Human chorionic gonadotrophin HCG A hormone produced by the placenta it indicates pregnancy and a part of if beta sub unit 1s increased in affected pregnancies High levels of HCG in the second trimester may indicate T21 Unconjugated oestriol UE3 A female sex hormone produced in the placenta it is decreased in Down syndrome affected pregnancies Pregnancy associated plasma protein A PAPP A low levels of PAPP A in the first trimester may indicate T21 and other disorders Dimeric Inhibin A an active form of the glycoprotein Inhibin A that is produced in the placenta T21 affected subjects may have high levels of Dimeric Inhibin A in the second trimester Once values of some markers have been obtained a Posterior Risk value is calculated based on the difference of the marker values to the normal marker va
113. on is warranted The likelihood of a disorder is normally calculated using two main factors the risk of the disorder due to age maternal age risk and the risk due to test results posterior risk from maternal serum markers ultrasound tests and other risk factors Cuckle H 1997 The final likelihood of having a given disorder is then calculated from the age and posterior risk 2 5 1 Maternal Age Risk The incidence rate of fetal anomalies is directly related to maternal age The risk of having an affected child increases exponentially after the age of thirty as shown in Figure 1 Historically maternal age has been used as the first screening test for fetal genetic disorders Maternal age risk 1s calculated using a population age risk equation and maternal age Thompson SG 1997 The age risk is based on the average disorder incident rate in the population for the mothers maternal age Figurel shows the relationship between maternal age and the risk of Trisomy 21 Risk of Down syndrome in live births 96 IS 26 27 25 29 J 317 34 33 34 35 36 37 38 33 40 41 4 43 44 45 Maternal age years Figure 1 Age Risk in Relation to Maternal Age Hook EB 2000 Furthermore before serum results can be evaluated the gestational age GA of the fetus has to be estimated as accurately as possible because correct analysis of the different biochemical marker results depends on knowing the gestational age precisely The three basic methods used f
114. on risk management however it is more suited for large projects Schaum 2001 Some development methodologies are better suited for some parts of the system than others for example the Waterfall model is not suitable for user interface development while the evolutionary approach would be well suited for such a task The Spiral Model was considered unsuitable for this project because project scheduling is difficult at the start since some tasks are not analysed until they are going through design Eventually it was decided that a mixed process was the most suitable model in order to incorporate the best features of the Waterfall and evolutionary methodologies It would involve using an evolutionary approach to resolve requirements for a given component and then finish implementation using a more structured waterfall based approach System components whose specifications were unambiguous would be developed using the Waterfall model while other parts were developed evolutionarily The mixed approach 1s justified in that it enables the developer to reap the benefits of both methods It enables development to start before requirements are fully established and as users develop a better understanding of the problem this can reflected in the software This methodology makes it easier to maintain software and when errors or new required functionality 1s identified the system can easily evolve to remain useful This project will chiefly follow the Waterfal
115. on s performance and level of functionality 9 9 Evaluation Results In conclusion the system achieved the required level of functionality The final solution met all the user requirements except for a small number of supplementary requirements It performed reliably and efficiently during testing and generates responses promptly The system is highly portable and can be accessed from most platforms with web browsers In addition the system can be installed on most web servers which have an installation of the requisite software The system met over 95 of the Nielsen usability guidelines and 9096 of the MIT accessibility guidelines The system passed the user evaluation and performed satisfactorily from the developer s and user s point of view therefore the final solution satisfies all the evaluation criteria 9 10 Suggested improvements During post implementation reviews testing and user evaluation it was realised that some functions of the system could be improved and new ones could be added These improvements would enable the system to offer better efficiency and security During evaluation users raised a need for the date picker to have a dynamic dropdown control for the year so that they could select or type in the year This would give users the option of typing in the date when selection from a drop down list would take considerable time In addition users felt that the date picker must provide a textbox showing the current value o
116. on the deployment server Continuous prototyping and demonstration will be used to ensure mutual agreement Background research will be carried out to familiarise with the technology 63 Contingency plans for when the risk happens Discuss with client if a requested take too long to implement Unallocated time will be left to code the missed requirements components will be created using an exploratory model to obtain clarification from user reviews The requirement will be scaled down or changed to fit into the requirements Consideration alternative solutions Consider hosting the application an alternative compatible servers User review may be carried out to get more specific feedback Affected units can be created on other platforms amp imported into the system Medium Political risks Failure of the management process Lack of time to perform all testing Lack of cooperation or conflict Insufficient budget medium Project unacceptable to the University Project fails to meet legal constraints gt Effects Existence of a single person to report to makes 1t unlikely for a communication breakdown to happen As much testing will be done during testing so that final testing does not take too long A good established relationship was during the developer s placement at MI The development costs are minimal Constant dialogue will be maintained with the
117. onditional statements to ensure that they executed correctly Black box test cases tested the functionality of the code without regard to how the code worked Black box test cases were based on checking the visible functionality of the code by validating the results calculated by the system Unit Testing Soon after implementation a unit was individually subjected to white and black box test cases All the units that it interacted with were replaced by dummy units For the purposes of this report a unit was assumed to be a single class Small programs were written to automate most unit tests and to act as dummy units During unit testing when an error was discovered the developer would resolve the error in code before repeating the test again to ensure the unit worked properly Because unit testing was done soon after implementing a given unit not all the errors rectified during the implementation could be documented in the test plan The test plan can be found in Appendix E Integration Testing When units had successfully gone through the unit testing procedure they were combined with other units to form an interoperating coherent component The emphasis of integration testing was on checking that units work properly together The main components of the application were security results calculation data access and the reporting system See Appendix E Testing for the integration test cases System Testing After finishing imp
118. or calculating the gestation of a fetus are menstrual history clinical examination and ultrasonography with the latter being the preferred method due to its accuracy When GA is calculated from menstrual history the gestation is based on the first day of the last menstrual period LMP such that the GA 1s equal to the day of testing minus the LMP date GA estimation by clinical examination of the pelvis or abdominal palpation is useful up to 28 to 30 weeks gestation after which it becomes too inaccurate for GA calculation Wilcox 1993 Ultrasonography is by far the more accurate technique for estimating GA Ultrasound methods for calculating GA include crown rump length CRL biparietal diameter BPD and head circumference HC CRL is the longest verifiable length of the fetus excluding limbs while BPD is the diameter of the fetal head An appropriate formula is then used to calculate the gestation from the obtained CRL BPD or HC measurement Westerway 2000 Ultrasound Screening is mainly used to determine the gestational age of a fetus however other serious medical problems such as blocked intestines or organ defects may be detected during ultrasonography The most established ultrasound marker is nuchal translucency NT normally sampled between weeks 11 13 of gestation 30 of fetuses with increased nuchal translucency found in the nuchal area at the back of the neck were found to have chromosomal disorders or other fetal an
119. or interactive systems International Standard SO 13407 18 IEEE 982 2 1987 Guide for the Use of Standard Dictionary of Measures to Produce Reliable Software 19 Isaacs 2006 The Media Innovations Ltd http www media innovations ltd uk mainframeset htm 19 ISO 13407 1999 Human centred design processes for interactive systems International Standard ISO 13407 20 ISO 9126 1991 Information technology Software product evaluation Quality characteristics and guidelines for their use 21 Jones 1999 Discussions with Dr Rick Jones Leeds Teaching Hospitals Trust and a Media Innovations Ltd Medical Consultant 22 Julie MeMullan 2006 Achieving Software Product Quality ESSI SCOPE Publisher Den Bosch pp 76 87 23 Klein 2005 Software Project Secrets Why Software Projects Fail Expert s Voice Apress 2005 24 Kotonya 1998 Requirements Engineering Processes and Techniques John Wiley amp Sons pp17 25 Kristol 1993 Picture Perfect The Politics of Prenatal Testing Elizabeth Kristol 1993 pages 17 24 Publisher First things the journal of Religion Culture and Public life 25 Lerdorf 2002 Programming PHP O Reilly Media Inc 1 edition pp 2 26 Liang Y Daniel 2003 Introduction to Java Programming Fourth Edition ISBN 0022R 5 0 13 100225 2 27 Mahoney 1990 The Ethics of Sex Selection In Medicine Medical Ethics and the Value of Life John Wiley amp Sons
120. osomal disorders such as Trisomy 21 caused by incorrect chromosome distribution metabolic disorders structural malformation of a fetus and anatomical malformations In addition during prenatal screening several other conditions e g syphilis or infections that may adversely affect the pregnancy may be discovered or tested for 2 2 Congenital Disorders Congenital Disorders are medical conditions that are present at birth that have been caused by genetic or environmental factors Chromosomal disorders are caused by an irregular expression of genes in a person resulting in a clinical phenotype Oncology 1999 Possible causes are gene mutation and abnormal chromosome distribution as in Down s syndrome extra chromosome 21 Some of the common congenital disorders are Down syndrome a variable combination of congenital malformations caused by Trisomy 21 It is the most commonly recognized genetic cause of mental retardation Because of the morbidity associated with Down syndrome screening and diagnostic testing for this condition are always offered as part of prenatal care Hernandez et al 1996 Edward s syndrome a trisomy of chromosome 18 whose effects include mental and serious health problems S Patau Syndrome Trisomy 13 a trisomy of chromosome 18 its symptoms and effects are similar to those of Edward s syndrome Klinefelter s syndrome an abnormal gene distribution affecting men that results in an extra sex chro
121. ot cause pop ups or other windows to appear and do not change the current window without informing the user 11 1 Use W3C technologies when they are available and appropriate for a task c nd use the latest versions when supported 11 N Avoid deprecated features of W3C technologies 12 3 Divide large blocks of information into more manageable groups where natural and appropriate 13 1 Clearly identify the target of each link 13 2 Provide metadata to add semantic information to pages and sites 13 4 Use navigation mechanisms in a consistent manner 85 19 Appendix F Implementation 19 1 Database 19 2 Security E J Javascript for encrypting user login details s is the string Qu Pn PHP and E is the hash key uti var cipher block size 543 encoding buf fer l z4 function b f c return c lt 16 2 D c to5tring l6 c tostring 16 Ha Ln em function bftf c I return parselntic 16 LO A function salts var n U for var 1 U iss length i n 143 charCodeat i return b f ntz55 Hl function encrypt s k if s length gt cipher block size var nm parselnt s length cipher block size t Hath round m Zz cipher block size return encrypt s substr D t E encrypt s substrit kE var r parzselInr Math random z55 o b0f r i for i i s length i o BOE 3 charCoded t i r k charCodeAt isk length h return
122. people s notions of medicine Screening is very useful in identifying and verifying conditions deduced from symptoms or physical examination and can even detect diseases in the pregnant mother When a chronic condition can be avoided early diagnosis not only saves the health system money but gives the individual an improved quality of life Post screening treatment may reduce the risk of developing a given condition or its complications Prenatal diagnosis of fetal anomalies gives parents the choice of continuing or terminating an affected pregnancy Moreover it is crucial for them to be prepared for a newborn disabled child 2 4 1 Issues in Prenatal Screening Although prenatal screening is considered routine in our century a number of issues have arisen from the differences in opinion of various stakeholders and the controversy created by the somewhat negative effects of screening on society Invasive methods do pose known dangers there is a miscarriage risk of 1 2 percent following CVS Jones 2005 Some experimental forms of genetic screening are socially controversial examples include vitro fertilisation a procedure in which embryo s are genetically analysed and those which have signs of genetic defect are discarded People could abuse the system leading to wrongful abortions 2 5 Prenatal Risk Calculation The risk values calculated during prenatal screening are used to determine whether further investigati
123. re 9 shows a screenshot of the final calculation screen the main page of the application elipse PrenatalLink Screening Calculator Mi media innovations Patient Markers Surname Default Date Ultrasound Site Default P Forename Marker Twinz Value Unit Id Ultrasound w MT mim Ethnic Default NB MK S NK ON Marker Value 1 Date 2 value 2 E Serum Number of Fetuses AFP Pregnancy Moncey sous HCG Dating Type None vw HOGh Measurement Inhibin 4 Dating Date PAPP A Donor Egg MES Egg Extract Date MAA 2 Extraction Additional information Date 1 Value 1 Date 2 Value 2 T21 Ma Truncation weight Kg T48 Donor DOB T43 Results Ed Diabetes Smoker VE E PrevT21 PrevT18 C T18 13 Figure 9 Calculation page screenshot 36 7 3 Security The system needs to protect sensitive information from unauthorised access As shown in section 6 5 some protection features are provided by the security system of the server As a result the application only needs to provide some of the absent security measures The system protects login details during transmission to prevent session hijacking Encryption is used to protect login details of a user by obfuscating them before transmission to the server Figure 10 illustrates how login details are transmitted from the client to the server Customer Id customer d Username username Password T 4ff5e eabe1z alab t cba416al Hf ascome
124. renatalLink Screening Calculator Medis innovations Patient Markers Surname Default Date Ultrasound Site Default v Forename Marker Value value 2 Turin Unit Id Ultrasound NT mm E Marker Value 1 Date 2 Value 2 Unit serum Ethnic Default NB Pregnancy Number of Fetuses 1 l AFP HMonozygnus HCG Measurement Inhibin A iu ml iur L nz ml pz ml mL L mu L Dating Date v PAPP A E Donor Egg UES Egg Extract Date SS Additional Information Tests Date 1 Value 1 Date 2 Value 2 121 E MA Truncation Mot Truncati Weight Kg v Donor DOB MA Extraction T18 T13 Uu ES Help O Diabetes Smoking ivFL Prevr21L Prewri8L 11813 El 25 1 Date Input a To enter dates click on the drop down icon of the date field to display a date picker Date field Default Date X April MENTA SB M hon Tue Wed Thu Fri cm 1 2 B 7 13 14 15 O 21 22 2 28 Za b Select the year and month from the year and month dropdown boxes respectively 112 c Choose a day by clicking the required day number e Click to paste the selected date into the date field and close the date picker f Click Xl to cancel and close the date picker 25 2 Validation Input validation is performed for numerical and date fields When inputting dates and numbers ensure valid values are entered before starting calculations Measurement d Microsoft Internet Explorer
125. resentation of results A permanent report that can be saved on the user s PC and printed The web pages must run satisfactorily on user PCs of various specifications on a Windows platform e The application must work correctly in different locales 21 1 4 Project tasks In order to carry out the project Courage was required to e Learn about the scientific background to Prenatal Screening e Understand the Elipse calculation engine its various components and its interfaces e Within the constraints of Media Innovations approved development platforms decide on the division of functionality through the application e g between the server and the client e Research and select the appropriate implementation methods and technologies e Carry out the problem solving development testing and fault finding tasks with minimal assistance 21 1 5 Evaluation The Beta testing is at an early stage so we do not have much user feedback Based on my own assessment and the responses we have received The development met the commercial deadlines The product meets the specification So far the Beta testing has revealed only minor errors which Courage has corrected The overall look is very professional and appropriate to a clinical application The layout of the user elements data entry form the results screen and the printable report is well thought out and logical e The product does not require much further work to be developed into
126. risk es a Ma A a a al 9 aa B IT txtDizorder txtPriorRisk2 Mi tdPostRisk2_ E nu exul d txETwvinFinalFi l BIG Resulta tdDisorder lbdPriorRisk3 Bl bdPostRisk3_ l bip d txtTvvinF inal l G Resuta seri L BE ust MI SEGUE wm zewvers S10 nal k CElipse yers zji anl rerl er Mi txtPrenatalLink erzioan BjlblDizclaimerg J his report was produced by Elipse TERT 1 1 which is an unvalidated Beta Jest version THESE RESULTS MUST Nifty OR CLINICAL PURPOSES 88 Figure 16 Template Patient Report PDF Form The generated report is then downloaded to the client s browser where they can open or save the PDF file Figure 2 shows an example patient report Elipse PrenatalLink Screening Calculator Patient Report 24 Apr 2006 Patient Information Surname Screening Gestation at 24 Apr 2006 11 weeks 3 days Forename Prenatal Fetuses 2 Id DD Weight 77 Kg 8 Kg DOB 01 Jan 1983 Other Details Diabetes Smoking Previous T21 Previous T18 MAEDD 23 yoars 10 months Markers Twin 2nd Measurement MoM Delta Gestation MoM Delta Gestation 0 89 11 weeks 3 days 1 66 11 wooks 5 days 1 87 11 weeks 3 days 1 45 11 wooks 6 days 0 34 11 weeks 3 days 0 36 12 weeks 0 days 0 81 11 weeks 6 days 1 17 11 wooks 1 day 22 70 12 weeks 0 days 27 29 11 wooks 5 days Results Twin 2 Age Risk Final Risk Result Final Risk Result 1 in 125 1 in 2483 Screen Hogativo 1 in 172 Screen Positiva Username leeds Customer l
127. ructions PI 43 1 The system was installed on the Media Innovation Ltd all the test cases where performed on the application running on the server under the same conditions as those encountered in normal use 20 3 Calculation Test Cases This section outlines the proposed tests to be performed on PrenatalLink to check if it calculates the correct results for the supplied data 95 In every test all the other input parameters will be valid except for the input parameter under investigation Doing so ensures that the test results are mainly influenced by the parameter under investigation since all the other parameters are valid For example in the Pregnancy Dating section on the SiteID test when the Invalid SiteID test is performed all the other Pregnancy Dating inputs have to be valid while the supplied SiteID is of the type suggested in the test 20 3 1 Pregnancy Dating Must be an integer gt 0 Patient DOB Valid date not more than 60 years less than the sample date Dating type Within the EnGestationMeasure enumeration A GestationMeasureT record must exist for the DatingID Dating date Valid date less than the Sample date Must be supplied where necessary for the dating type Dating measure Must be within the range specified in the GestationMeasureT record for the Dating type Site and SampleDate Must be supplied when required e g Gestation CRL BPD HC Monozygous Can only be true for twin pregnancies note all Assi
128. rver sides to ensure users enter valid dates and to cater for non JavaScript browsers The validation code for dates and numbers is locale sensitive and uses the correct thousands decimal and date separators for the user s region Input validation 1s done on leaving a field for JavaScript enabled browsers and on the server side when a form is submitted 37 7 5 Development Summary This section describes the features that were implemented in relation to the system s requirements The software for the application was developed according to Media Innovations software development standards which are reproduced with permission in Appendix H The final solution s code is located on the accompanying compact disk at D Software PrenatalLink 7 5 1 Features Delivered Figure 11 describes the features implemented to satisfy minimum requirements and some of the requirements extensions Requirement Web application to allow calculation of results Database system for data storage Data input and results reporting systems Security system for the application Help system Date input system Automatic client side input validation and session tracking systems Page caching functionality for static web page components Wide browser compatibility Internationalising the application so that it customises to locales An integrated help system that uses compiled help pages Implemented Feature The calculation feature was imple
129. s k strRiskF actors showHelp L Connection createPatient getProperties DbDataAdapter zetWeight resultsProparties setProperties Dbcammand getDate DataSet cetSettings mks di calculate zetsattings setResults DbConnection l m strCustomerld DbDataAdapter i strUsername abjConnect DbCommand strDBName Dataset intConfigl d lagin Markers ESAS H Eins Cei constructori canstructors resetDetails initialise zetsettings Dis E getmonths chkDetails sl zet serinfa calculate cetFormatinfo SetFocus getCustinfol setResults Figure 6 System Classes Figure 6 shows class diagrams of the objects and their relationships each class will be implemented in a separate file The following paragraphs explain each class in more detail 28 6 2 2 Connect This class connects to an SQL Server database using a native database connection object It then uses a precompiled SQL command to retrieve user details and store them in a dataset A precompiled command is used to prevent attackers from injecting malicious SQL commands into login fields It is mainly used by the security class to retrieve user details from the database 6 2 5 CurrentUser The CurrentUser class stores the name user identity and configuration details of the current user this information is used to identify the correct settings records of the current user when calculating results and for reporting purposes I
130. s a last updated date Consistency The same word or phrase is used consistently to describe an item E Link reflects the title of the page to which it refers T Browser page title is meaningful and reflects main page heading gt Error prevention and correction Users can rely on recognition not memory for successful use of the site y Site tolerates a reasonable variety of user actions Site provides concise instructions for user actions including entry format e Error messages are visible not hidden E Error messages are in plain language Error messages describe actions to remedy a problem y Error messages provide a clear exit point T Error messages provide contact details for assistance Architectural and visual clarity Site is organized from the user s perspective v Site is easy to scan for organization and meaning T Site design and layout 19 straightforward and concise Site design and layout are redundant only when required for user productivity T White space is sufficient pages are not too dense T Unnecessary animation is avoided Colors used for visited and unvisited links are easily seen and understood X Bold and italic text 1s used sparingly a 18 3 Accessibility Guidelines W3C Checklist of Checkpoints for Web Content Accessibility Guidelines 1 0 In General Priority 1 YesNoN A 83 1 1 Provide a text equivalent for every non text element e g via alt longdesc or in element content 2
131. s and the social issues arising from it are even touchier Eradicating illness through screening requires a change of fundamental attitudes towards parenting and social responsibility this has given rise to a multitude of ethical issues related to screening Prenatal diagnosis with a view to termination of the pregnancy is widely acceptable Mahoney 1990 Most people fear that prenatal screening may lead to people choosing what children they want to have Is a mentally or physically different child less valuable to society Although parents are fully informed prior to every stage of the screening process making decisions regarding life is very difficult and will always be controversial but as Stephen Thacker of the centre for disease control says Screening examination in early pregnancy is here to stay Mahoney 1990 Prenatal screening 15 has undoubtedly widespread advantages as well as negative social consequences however the medical community generally agrees that screening has considerable benefits Jones 2005 3 4 4 Economical Policy makers and health services are under increasing pressure to improve health provision and cut down soaring health costs Scott 2005 Widespread screening and termination of positive pregnancies would reduce both problems although it is arguable whether cost benefit analysis 1s acceptable when it comes to health For example Down s syndrome accounts for 15 percent of the institutionalised mentally r
132. s that were as complete as possible in the hope that we would uncover the highest number of errors and correct them before user testing began in earnest 42 9 Evaluation The main aim of this section was to appraise the system s functionality and to retrospectively assess the quality of the system It also evaluates whether the intended benefits have been achieved and identifies unintended benefits of the system to end users The evaluation processes assess the final solution by describing its good features and where applicable show how things could be improved It concludes with a description of additional features which would be useful additions to the system 9 1 Evaluation Criteria The evaluation of the system was based on the success of integral functions and the evaluation criteria set out in section 1 9 during project conception The criteria are based on established software evaluation standards The International Standards Organisation ISO specifies six characteristics for software evaluation ISO 9126 2003 as shown in figure 15 The widespread use of these standards for evaluation means they are a useful means of appraising software McMullan 2006 Accu racy Compliance Interoperability Functianality security Suitability Fault tolerance Recoverability E Resource amp Learnability T Operabilit Adaptabili P E Portability OM Understandability
133. st tustomerid casus Retrieves ee username e Hash using Authentication SHA 1 Hash details a Algorithm 5baa6lesdcotear3m6s22 obec 91h72 i i i i i i 249ba36000029bbe97499c03db5a500f i i i i i i Figure 10 Secure Hash Authentication The security class uses challenge hash authentication to avoid transmitting passwords in clear text The password entered by a user is hashed and sent to the server where the hashed password is compared to the hash code of the stored password In addition to hash authentication users are tracked using automatically generated session keys This ensures that users cannot access other people s data also session keys are destroyed when users log out or when their session expires 7 4 Data Entry and Validation A date input control was created using dynamic HTML and JavaScript The control enables users to enter dates in a convenient and efficient manner and prevents them from entering invalid dates The date picker is local sensitive and displays month names in the Mon Tue Wed Thu Fri Sat Sun 1 2 3 4 85 7 8 910 11 12 they use this control since it only displays the valid days for a given language of the user s region Users cannot enter invalid dates when 14 315 16 17 18 19 21 22 23 24 25 26 28 280 30 31 month and year Alternatively users can enter dates directly into a date field Numerical fields and date validation 1s done on both the client and se
134. sted reproduction fields will be disabled unless the dating type 1s assisted reproduction Donor Egg Can only be true if the pregnancy was started through fertilisation Ma at Extraction Required if AR Donor DOB is not supplied or if patient DOB is not supplied in a assisted reproduction pregnancy Must be Fertilisation or Implantation AR Donor DOB A valid AR Donor DOB must not be more than 60 years less than the sample date Invalid inputs l SiteID l l Invalid SiteID e g negative SiteID 1 2 Error on calculate Invalid site id 2 Patient DOB Invalid Patient DOB Resulting age outside of the range 0 lt age lt 60 O Error on calculate Invalid maternal age or pregnancy dating information O Patient DOB gt Sample date O Error on calculate Invalid maternal age or pregnancy dating information 96 10 11 12 Sample Date O Invalid Sample Date e g a sample date that 1s less than the dating date O Error on calculate Invalid pregnancy dating information Dating type O Invalid Dating type GestationMeasureT record not present for Site and Sample date Error on calculate Missing configuration data for BPD 01 01 1999 Same Dating type used twice in the same calculation Error on calculate Unable to add element to collection existing element with same key Dating date O Dating date gt Sample date O Error on calculate Invalid pregnancy dating information O Dating
135. stem does not require any attendance from a human being except for occasional maintenance tasks 3 4 06 Scheduling This section analyses the viability of the project being completed on time Development of the final solution is predicted to end on the 12 of March 2006 as shown in the project schedule Appendix C The implementation of the solution and documentation are scheduled to be completed by the 27 of March 2006 Enough time has been allocated to cater for delays and testing Although this project is complex and requires a considerable amount of time to complete it should be possible to complete it on time if things go according to plan 3 5 Risk analysis Theoretically every decision in a project should be subjected to a risk assessment in order to manage risks effectively Crump 2005 however such an approach is impractical for all but major project risks Effective risk management involves both informal awareness and a structured approach in 16 which potential risks are identified and mitigation strategies are systematically implemented Complex software development is risky and understanding risks is a crucial step in avoiding becoming another failed software project Many post mortems of project disasters indicate that problems would have been avoided or reduced 1f there had been explicit concern in identifying and resolving high risk elements HRS Consulting 2000 This section will identify analyse and prioritise risks It
136. sts are then carried out before prospective courses of actions are considered The broad scope of this project is to respond to the present need for software systems that are configurable available on the internet and incorporating some of the latest research findings on ultrasound and serum markers A system that produces results at each major stage of risk calculation instead of one final risk value and hides the complexity of result interpretation from the user Wilson 2006 There has been widespread need for provision of screening services from the internet with the aim of enabling clinicians to utilise the advantages of web based systems Jones 2005 The delivery of prenatal screening systems that allow clinicians to configure calculation parameters and customise them to local population averages would be a useful development in the prenatal screening industry The aim of this project is to deliver such a system a system that will be web based and incorporating some of the latest technological advances in prenatal screening The system will serve both clinical and research interests of clinicians due to its highly configurable functionality This project will be developed in conjunction with Media Innovations Ltd and they will be involved in every stage of this project s lifecycle 1 2 Media Innovations Ltd Media Innovation Ltd MI is a subsidiary company of the University of Leeds It is a leading developer of multimedia training materia
137. t 0 Invalid Inputs 21 Disorder type 22 SiteID O SiteID not supplied O Correct calculation 1f defaults exist O Invalid SiteID supplied string or lt 0 O Error on calculate 23 MAEDD O Invalid Maedd supplied lt 0 O Error on calculate Invalid Maedd O Maedd out of Disorder s age range O Error on calculate Maternal age out of range 24 Number of Fetuses Number of Fetuses not supplied Correct calculation default number 1 Invalid number of fetuses supplied lt 1 Error on calculate Overflow Valid number of fetuses supplied but out of calculatable range 0 lt Num Fets lt 3 EN O mM O BO Error on calculate Cannot calculate risks and results in grand multiple pregnancy 25 DisorderID Disorder not supplied invalid DisorderT record does not exists Warning on calculate No disorder results requested DisorderID is Combined and 2 extra disorders are not supplied Error on calculate Must include gt 2 disorders for combined risk When DisorderID is Combined and 2 extra disorders are supplied Correct calculation The same Disorder is specified more than once in the same calculation O oOo o BN o DO Error on calculate Unable to add element to collection existing element with same key 26 Markers Collection and Marker Set Marker Set not supplied Correct calculation all supplied markers are used Required markers in marker set are not supplied in markers collect
138. t Effects Impact medium gt Action gt Rationale high Operational Intensive testing and evaluation must be done before system installation Software audits must be taken regularly Such risks are expected to occur and if monitored correctly quick response will ensure that few customers are affected Rising competition Low sales funds and low return on investment Likelihood Type medium Operational Counter measures to combat competition must be implemented Competition will always exist in any successful industry and companies must keep ahead to stay alive Lack of maintenance Degradation of service quality low sales and system failure Likelihood Type medium Operational Maintenance registers must be filled at regular cycles and administration must ensure that regular system servicing 1s done Data loss Widespread loss of service Likelihood High Type operational Regular backups are compulsory And backup copies are stored offsite in very secure surroundings If backups are done minimal data loss must be incurred and only a few customers will be affected while restoration is done System security failure Disruption of service and disclosure of confidential information Likelihood Type medium operational The security system of the server will protect stored information we aim to develop a security system to protect data in transit The information an attacker will be able to gain is not of much use so
139. t has a get and set method for every property and the set methods validate new values before updating the current value of a property 6 2 4 Security The Security class provides security functions for the whole application it encrypts user s login details in preparation for transmission over the internet using a secure hash algorithm When users initially access the application their login details are checked against stored values If the login details are correct a CurrentUser object 1s created and stored in a session variable 6 2 5 Patient The Patient class stores details of the current patient patient information is displayed on patient reports for identification purposes 6 2 6 PregnancyDating This class calculates pregnancy dating results using an Elipse object the gestational age of a fetus and mother s maternal age at expected date of delivery 6 2 7 Markers The Markers class calculates marker median values from the input biochemical marker test results using an Elipse object 6 2 8 Disorders This class calculates the final risk values for the selected disorders using the results from the pregnancy dating markers and other additional risk information 6 2 9 DBView The DBView class connects to the database and retrieves settings data for use by the PregnancyDating Markers and Disorders classes in calculating results 6 2 10 Calculate The Calculate class is the main class of the system it has PregnancyDating
140. t issues and project feasibility The design chapter explains the modelling of the system and is followed by an implementation chapter that explains the creation of the final system based on design decisions The testing and validation of the system 1s described in the chapter after implementation The penultimate chapter evaluates the final solution and the report concludes with a conclusion section that explains how the system can or could have been improved This report also includes a glossary in Appendix B to explain the medical and technical terms used in this report 2 Background This chapter provides a general background to prenatal screening it explains in general terms the theory behind prenatal screening how the screening process works and identifies the current issues related to screening It concludes with an explanation of the risk calculation process 2 1 Screening Theory Screening is a systematic process of identifying the possibility of an individual being affected by a disorder or a disease when no problem is suspected It is a means of finding out those individuals who are most at risk of having a certain condition and give them further more expensive inconvenient and even dangerous diagnostic tests Wald et al There are a multitude of diseases for which prenatal screening programmes have been developed Several conditions can be tested for in a fetus or embryo before it is born Jones 2005 The aim is to detect chrom
141. t user feedback and enable unit testing of the system s software components as they are implemented 6 8 1 The Implementation Process The software design will be transformed into executable code using the following guidelines Analyse the design specification of a software unit and plan its code Implement the unit s functionality and write its code in accordance with Media Innovations Software development guidelines Test each software unit immediately after implementation before integration into the system Evaluate code and test results using the following criteria satisfaction of requirements appropriateness of coding methods conformance to guidelines and performance Improve the software unit and repeat the testing process before updating the test plan and the implementation documents before moving on to another unit After all components and improvements have been implemented integration testing will be preformed and the application will be validated as explained in the testing chapter Table 6 outlines the functionality to be added at each phase of implementation Iteration Functionality Added Project setup in the ASP Net development environment creation of directories l documentation files setting permissions and test to make sure everything works 2 Database creation and query testing using SQL server query analyser 2 Skeleton login page with user identification fields without any styling or colours 3 Data
142. te Required Date at which Measurement taken Donor Egg Egg Extract Date Measurement Required Measurement of the type selected in MA 97 Extraction the Dating Type field For twin pregnancy with ultrasound PAS Truncation measurements enter a measurement for each fetus Donor DOB The remaining fields are for Assisted Reproduction AR 114 pregnancies patient s stored egg or embryo or donor egg and are only enabled for that AR Donor Egg Checkmark if egg donation pregnancy Storage Date Optional For stored egg embryo the date at which the egg embryo patient or donor was extracted and put into storage for adjusting MAEDD MA at Storage Optional For stored egg embryo if Storage Date and Donor DOB for egg donation pregnancy are unknown enter an approximate maternal age at date of storage in years plus decimal part year for adjusting MAEDD MA Truncation If MA at Storage is in completed months half years or years select the appropriate truncation Donor DOB Optional For an egg donation pregnancy either this field and Storage Date or MA at Storage must be completed 25 4 3 Markers Markers Ultrasound Site Default wv Value 2 Twin Default Date 26 Apr 2006 Marker value Ultrasound Date NT MB Marker serum AFP HC HCGb Inhibin A 26 Apr2006 v 26 Apr 2006 Date 1 Walue 1 Date 2 26 Apr 2006 Y 26 Apr 2006 Y 26 Apr 2006 v 26 Apr2006 v Value 2 PABRP A
143. the commercially released version some minor modifications and more extensive validation 104 This was quite a complex development requiring a good understanding of the scientific background a variety of IT platforms and components and issues of commercial development and deployment It gave Courage considerable scope for using his own initiative in design and implementation decisions and problem solving and he worked throughout without needing close supervision am very satisfied with Courage s contribution to the project and with the finished product 28 April 2006 Carol Wilson Scientific Software Ma 105 22 Appendix G Product Content 22 1 Installation and Setup 22 1 1 System Requirements Microsoft IIS server Microsoft SQL Server 22 1 2 Installing the database Open Enterprise manager Go to database and right click the databases tab Select restore database Use the back up file at D Software ElipsePrenatalLink DLLs elModel BatchTest In the dialog box that opens give the file the database the name elModelV3 Follow all the other instructions and choose security options as required Open the restored elModelV3 database and click logins Give the elUser and elAdmin permission to access the newly restored database 22 1 3 To install the system Create a real folder to store the system files and copy all the files under D Software ElipsePrenatalLink into that directory Create an IIS Virtual folder for the s
144. the fetus and in addition other defects such as malformed organs might be noticed There are several indicators that can be found during ultrasound screening which have a significant association with chromosomal abnormalities These markers in combination with other biochemical markers can be used to discriminate between unaffected fetuses and positive ones Smith Bindman 2001 Examples of ultrasound markers are nuchal translucency NT the thickness of the nuchal area at the back of the neck and neuroblastoma NB a cancer that affects nerve cells of immature or developing cells such as those found in a fetus In addition to ultrasound screening physicians carry out serum screening a procedure in which they check a mother s blood to obtain certain measurements such as alpha fetoprotein AFP unconjugated estriol UE3 and human chorionic gonadotropin HCG The blood test results of these independent markers together with the fetus gestational age and other risk factors related to the pregnancy can be used to calculate the risk of a fetus having congenital disorders or chromosomal anomalies Complex risk calculation procedures are then used to determine the likelihood of a pregnancy being affected based on the clinical test results gestation population parameters and other factors related to the mother Wald et al 2004 If the results are positive justifying further investigation more dangerous definitive and invasive te
145. tion information o In Class Terminate set this variable to Nothing For database read sub returns a recordset o For database write data supplied as recordset O Code layout o Use Option Explicit and always declare variable type Declare private objects together at beginning of proc Declare global variables together in a single modDeclarations module o Commenting use comment blocks at start of object section etc and in line comments Res asp elipseWeb v2 include At start of each module and proc what it does Each variable Code section for a particular task at start what it does Why something is done in a particular way if not obvious All database fields unless completely obvious and component type libraries o Indent to show grouping nesting prefer 2 or 3 spaces In general don t have code lines much longer than screen width use o Use llf If Elself With only where this makes code clearer O Naming conventions o Never use the same name for different entities For entities representing the same quantity e g property module variable representing it parameter procedure variable representing it variables with overlapping scope use variations on the root name o Ensure the name conveys sufficient information and don t abbreviate unnecessarily e g dtmTransactionDate not dtmTrnsDte o StrWordStringsLikeThis not strWord Strings Like This or strwordstringslikethis o For module variables us
146. tions Ltd She was crucial to the success of this project and provided invaluable help advice and encouragement Many thanks to Dr S R Roberts my University of Leeds project supervisor for his constructive advice and helpful suggestions And finally thanks to my friends Marion Braun Mathew Bosquet and Carmen Gonzalez for their support throughout this project A considerable number of people were involved in requirements analysis testing and evaluation of this project I would like to express my deepest gratitude to all those who have not been mentioned here It 1s with great appreciation that I acknowledge the information counsel and support you provided 111 Contents SUM MV AAA cadi o ND ed rod tesi a 11 A O eei tdm ono t ERU tube bd une e md ce dr DPA eR tU IL mU LAU 111 E Introduce 1 1 1 The Problem Domaica iaa l 132 Media nt Oy AOS Eid aiii 2 1 3 A A o a Med 3 14 Projgeet OD e E as 3 D MM CIES a ines pad 3 I Deliver leurs acia 4 1 7 cnc dl aio 4 ls Relevance O Dret alcoi ada 5 ES Evaluation Criterio dd dd dE Dd ibi 5 LOS REPO SUU U Caena a as 6 Zi WHACK SOU UN 7 24 SAN AAA esu ear A Pee A e dU Enc S dc Pe Pe En M 7 22 CONSELL DISOFUG S e a tan sesto dte nca aus qudd ute te ae ees etu 1 2 3 Sercenino and Diagnostic Tea L 8 24 Reasons tor Prenatal e en nET 8 D gt Prenatal Risk Calcular EA A A a 9 2 6 Project METIO Te Te A A A ute quim ae 11 A O O ILL Le 13 3 1 Considerations ss 13
147. to hereditary and gene related issues The period from the conception date to the current date A hormone produced by the placenta it indicates pregnancy and a part of if beta sub unit is increased in affected pregnancies A feature of the Net framework that enables the auto completion of code statements Abnormal gene distribution affecting men that results in an extra sex chromosome Men with Klinefelter s syndrome are usually sterile and maybe dyslexic Biochemical markers present in the mother s blood during pregnancy Haddow et al Abnormal levels of these chemicals may indicate increased risk for certain birth defects and genetic diseases The history of the menstrual cycles of the mother Refers to digestion disorders it refers to the uptake digestion and utilisation of food Massachusetts Institute of Technology is an engineering research institute and University in Cambridge United States of America Birth defect resulting in improper development of the brain or spinal cord Patients and all those who receive the services of this system Procedures that are used definitively test for the presence of a medical condition in an embryo or a fetus Factors that increase the risk of an individual developing a disorder The risk of a pregnancy being affected by a given disorder e g 1 in 100 means the probability of the pregnancy being positive is 0 01 A systematic process of identifying the possibility of an individual being affected by a
148. tors will be conducted before the system can be used clinically However during the external evaluation the beta system will be available for non clinical use to registered physician for evaluation and testing purposes It will also serve to advertise the system before the release of the clinically validated alpha version 1 5 Minimum Requirements The minimum requirements of the project and their possible extensions are Minimum Requirement Possible Extension A commercial web application for Batch calculation system that allows users to calculate calculating risk results results offline Database system for data storage Import and export functions that enable users to easily upload input from files Data input and results reporting systems An advanced data input system that allows users to enter dates using date pickers etc Security system for the application An automatic client side input validation system and session tracking system Help system delivered in the form of a web An integrated help system that uses compiled help page pages or case sensitive functionality Table 1 Minimum Requirements and Extensions Possible additions to the minimum requirements and extensions are Page caching functionality that stores static elements of the applications web pages on the client side to improve performance Enabling the application to be usable from Internet Explorer Netscape and Fire fox browsers nternationalisi
149. tractions that explain different approaches to software development The methodologies considered for providing a guiding framework for this project are described below 11 Waterfall model The waterfall model was derived from engineering processes Royce 1987 It separates fundamental software processes into separate stages requirements specification design implementation testing and maintenance This model enables the development of well structured systems based on initial requirements and easier project planning It suffers from inflexibility making it difficult to respond to changes in requirements under this model Somerville 2004 Prototyping An evolutionary development model that interleaves the specification development and verification of systems An initial system is rapidly developed and several refinement cycles are carried out based on user feedback to produce the final system It enables rapid software development but has drawbacks in that software documentation is difficult and prototyped systems are often poorly structured due to continual change Herndon 1997 Spiral Model The spiral model splits the development process into a sequence of activities Setting objectives risk assessment development and validation and planning At each stage or loop the above activities are carried out in order leading to incremental development It enables risk minimisation because of its emphasis
150. trated that the system always displays an error message and cancels the current operation when a run time error occurs All software routines of the solution use the error handling routine explained in section 7 6 to ensure faults are detected handled and reported to the user The validation system aims to prevent errors from happening by validating input to prevent buffer overruns data type conversion errors and formatting errors Indirectly the validation of input also increases reliability because users can only enter valid input which the system 1s capable of handling Since the 10 of April 2005 the beta version of the system has been operating on the MI web server without any reported faults Wilson 2006 9 4 Efficiency The efficiency of the system measures the system s level of performance in relation to available resources The main efficiency measurements criteria for the solution are response time bandwidth and storage usage The total size of the whole system is 11 9 megabytes MB and during operation this value will only be increased by additional settings records added to the database making its usage 46 of storage space very efficient The response times of the application are less than 5 seconds on average over a 1 MB connection since the mean size of the application s pages 1s less than 1 MB The patient report is 256 kilobytes in size and takes a second to download over a megabyte connection showing that the response tim
151. ty Sun 2005 It has a highly structured architecture that enables the development of object oriented applications that are maintainable in a straightforward manner JSP pages are highly efficient since the Java virtual machine stays up and a new lightweight thread 1s created to handle requests It 1s generally free or very cheap in contrast to most commercial quality web systems that are relatively expensive Debugging JSP pages is still complex due to an absence of a powerful integrated development environment Database connectivity 1s not as efficient as it should be since connection pooling has to be done manually using custom code Horton 2003 To date there is no what you see is what you get editor for JSP pages consequently web page design and development has to be done using other HTML mark up technologies 23 5 1 2 Active Server Pages Net Microsoft ASP Net is a programming framework for building web applications It is part of the Net framework and succeeds Microsoft s Active Server Pages ASP The Net framework is a development and execution environment that allows different programming languages and libraries to interoperate seamlessly in a language neutral common runtime language MSDN 2006 The use of a common language runtime enables ASP Net to provide a multi language platform such that web pages can be written in VB Net C J It provides true object oriented programming with capabilities to handle polymorphism
152. ule The solution that was implemented is currently undergoing clinical validation and trial before being launched online 1f all goes well It gives the author great happiness in knowing that this project produced a useful application that could improve and save lives Working on this project has been a very enjoyable experience and a good learning process I believe that I could have improved things by starting my write up earlier 58 14 Appendix B Glossary Alpha fetoprotein AFP Amniocentesis Amniotic fluid Anatomical malformations Ancephaly Anencephaly Application Biochemical markers Chorionic Villus Sampling Chromosomal Abnormalities Chromosome Clinical phenotype Clinical phenotype Congenital Disorders Cytogenesis DBMS Embroscopy Fetal Anomaly Fetus Gene mutation Gene mutation A protein made in the liver of the fetus Low levels of AFP in the second trimester 14 26 weeks gestation may indicate a T21 positive pregnancy A procedure in which a needle guided by ultrasound is inserted into the uterus and with draws some amniotic fluid for cell analysis and diagnosis Protective fluid that fills the sac surrounding the fetus A structural malformation of a fetus A defect in the closure of the neural tube during foetal development A failure of the neural tube to close properly in a fetus it may also have symptoms such as incomplete brain and skull bones The program written for this proje
153. ut and The data input system enables users to input values in a convenient manner it reporting validates input immediately after editing to ensure that users get real time alerts to systems invalid inputs The validation checks input according to the user s local formats such that a French user can enter a number with French decimal and thousands separators and the number will be validated appropriately and interpreted correctly The reporting system produces reports in various formats so that users can save the reports e mail them or print them The PDF format for patient reports enables users to secure reports on a file level use them across platforms 44 and the appearance of printed PDF files is of high quality Security system The system authenticates users and encrypts user login details to prevent for the unauthorised access The technique used for encrypting data is secure and no application encryption keys are transmitted from the client to the server The security system is efficient and fast because it does not decrypt user details instead it encrypts the comparison details stored on the server and compares them to the encrypted login details sent from the client side This protects the server from any malicious input However the security system is not comprehensive there are some threats that still need to be mitigated for the system to fully protect data during transmissions Help system The help system created for the applic
154. value soon after inputting it greatly improving the system s performance and usability When a user moves out of the application using the back and forward browser navigation buttons they can come back into the application by navigating in a similar way This is a security vulnerability since an attacker might be able to access a page of the application that a user has just visited beforehand Future versions of the system must monitor the navigation on the client browser so that when users navigate out of the systems boundaries the system automatically logs them out and destroys their session variables to prevent anyone from accessing the session data or pages the user might have visited earlier This solution would greatly improve security and reduce wastage of server resources used to maintain session data The batch calculation function was cancelled due to project constraints however future versions of the system must provide an offline batch calculation service to serve the needs of customers who handle large numbers of patients Finally the performance of the system could be optimised by the use of embedded images This would ensure that images are transmitted to clients before the system needs to display them greatly improving response times and the appearance of the system 49 10 Conclusion The main focus of this project was to develop a web system that would enable clinicians to calculate the likelihood of a pregnancy bein
155. ve help on the major tasks results calculating reporting validation pregnancy dating markers disorders and security if possible The system shall provide a form for viewing help General Requirements The system implementation of the system must be documented in an implementation document explaining the decisions made during development and their rationale The system must be implemented according to media Innovations guideline it must be written in a clear consistent and well documented manner The developer shall write maintenance documents for the system This document must explain the systems functions their locations and how the system works The system must be usable on Explorer browsers since they are the default browsers in the health system The system must use Media Innovations Ltd logos and copyrighted graphics on the user interface 80 18 Appendix E Design 18 1 Database Normalisation Normalisation is an iterative process that eliminates redundancy organises data in an organised manner and reduces the potential for anomalies when records are updated Edgar Codd established originally established 3 normal forms The first normal form states that attributes can only store a single value A table is in second normal form if all non key attributes are fully dependant on the primary key A table is in third normal form if it is in second normal form and there exist no non trivial dependencies between non key attributes
156. ven pregnancy being affected by congenital disorders especially those caused by chromosomal anomalies The final solution calculates the risk value for a pregnancy It uses some of the latest ultrasound markers such as Nuchal Translucency and Nasal Bone and calculates results for more than eight disorders The system can calculate results for combinations of between one and eight markers from different sample dates enabling results for multiple markers to be calculated in a single transaction The calculation of risk values is amongst the most mathematically complex tasks in medical support Wilson 2006 and the level of interpretation provided by the system in results presentation makes the system very useful indeed 45 9 3 Reliability Software reliability 1s the ability of a system to perform its required functions under stated conditions for a specified period of time IEEE 982 2 1987 The correctness of the results generated by the system 1s crucial to its quality and reliability incorrect results are unacceptable The system was tested using unit integration system and acceptance tests The unit tests were carried out soon after implementation to resolve any coding errors and to check program logic After verifying that the code of a unit was correct the unit was integrated into a larger system component such as the security component The rationale for unit tests was to resolve coding errors immediately after implementation In additio
157. ver the years your project supervisor will have seen many final year projects arrange to have weekly meeting with your supervisor to get feedback from him The meeting will enable you to get constant help and advice as well as inform you supervisor of your progress I found the feedback very useful by meeting my supervisor every week it meant that he clearly understood my project and was always up to date on what I was doing and could give me advice proactively Team work some of this project s development had to be done at Media Innovations Ltd During this time I was required to work with other Media Innovations employees and this project s MI supervisor Developing a project for an external company gave me the opportunity to experience how things are done at a professional level I would advise future students that developing a solution that solves a real problem simplifies a lot of tasks and may give you the opportunity to improve skills such as team working When dealing with third parties always plan for the meeting beforehand and behave in a highly professional manner in addition value their time because most people are likely to be very busy My project enabled me to learn the ASP Net framework and VB Net It also honed some of the skills that had been learnt during my degree Overall I feel that my project was very successful the aims minimum requirements and objectives were achieved to what I felt was a high standard and on sched
158. xisting element with same key Two Markers at different gestations added to the Markers collection Correct calculation the two markers will have different markers collection keys 98 14 AssayDate 15 16 17 18 O O O AssayDate not supplied Error on calculate Missing marker information Value and Normalisation Value is lt 0 when Marker normalisation is not equal to delta and ValueIsNormalised S Error on calculate Invalid Marker information Value is lt 0 when Marker normalisation is equal to delta and ValueIsNormalised m Correct calculation m Valid MoM when Marker normalisation is delta lt 0 Correct calculation m Value is lt 0 and marker normalisation Level or MoM S Error on calculate Invalid Marker information Value is outside the MoM limit range specified in the Marker distribution record S Error on calculate Invalid Marker information m Supplied MoM is outside the MarkerDistributionT MoM range Error on calculate Invalid Marker information m Value not supplied or equal to 0 Error on calculate Missing marker information Gestation type and gestation value O BM OO E O O O GO O O A gestation type value is supplied but the gestation value 19 not Error on calculate Missing or invalid gestation information A gestation type is not supplied but a gestation value is supplied Error on calculate Missing or invalid gestation informat
159. y Courage Chibiya in conjunction with Media Innovations Itd Contact Media Innovations Ltd 3 Gemini Business Park Sheepscar Way Leeds LS7 3JB UK Tel 44 0 113 262 1600 Fax 44 0 113 262 1605 110 a Start up your Internet Explorer browser and enter the URL shown the image below into the address bar or if you have installed the Prenatal Link system on your machine point the browser to the path of the virtual IIS folder that contain the system See Appendix XX Product Content for installation instructions pack T gt f Jj Search 5 7 Favorites 2 ddress httpijf213 237 225 18 PrenatalLink Def aul aspx v Eco b Enter the correct Customer Id Username and Password into the appropriate fields and click the Log in button Login Customer Id User Password Remember customer id and user name an this computer your browser settings may not allow this L Note All login details are compulsory If you enter incorrect login details and wish to empty the contents of all login fields click Reset For testing purposes you can login using the following login details e Customerld ltest e Username Leeds e Password pass 111 25 Calculation The figure below shows a screenshot of the calculation page To calculate results you will be required to supply patient dating markers and disorders information The following paragraphs will explain how to calculate results P
160. y and execute quickly Lerdorf 2005 There are problems in PHP s object model It was not designed as an object oriented language and as a result it lacks features such as exception handling and garbage collection As with most scripting languages PHP lacks a proper debugging tool and as result debugging can be very complicated 5 1 6 Conclusion Based on the performance security ease of use and advantages of the systems described in this section 1t was concluded that ASP Net and JSP would be ideal platforms for implementing the final solution The developer has studied all the above technologies except for Perl and PHP PHP was considered unsuitable for this project due to its lack of object oriented architecture Perl was discounted due 1ts considerable complexity and lack of security drawbacks Python does not provide a graphical interface for designing the applications user interface consequently the developer felt that it was unsuitable due to its lack of an integrated development environment IDE With the tight project schedule in mind it seemed prudent to use a platform that the developer was very proficient in and a platform that provides a powerful IDE Having to use an independent HTML editor to create web pages in JSP made it less attractive as a medium for developing the final solution Based on these requirements ASP Net was chosen for developing the final solution because its minor drawbacks are unlikely to negatively affe
161. ystem according to the IIS guidelines and set the permissions according to your security requirements and point it to the real folder Start up you browser and point it to the virtual folder and the login page of the application will be displayed 106 22 2 Implementation Document Sample page from the implementation document describing the security class 22 2 1 UserConnect Class This class encapsulates the functionality required to read user records from the ElipseUserT table Public Sub New Used by the Security class this method initialises the variables required to connect to elModelV3 SQL Server 2000 database Private Sub LoadUsersDateset Called from default px this method connects to elModelV3 uses the select commands to fill the Users and Customers datasets with records from elModelV3 Public Function GetUserInfo strFieldName strUsername Takes a name of an ElipseUserT field and a username and returns the value of the specified field from the row that h the given username Public Function GetCustInfo strFieldName strCustid Takes a name of an ElipseUserT field and a username and returns the value of the specified field from the row that h the given username 22 2 2 Security Class Private Sub Page Load sender event Handles MyBe Load This method is executed every time the page 1s loaded it mainly initialises the login vars Private Sub butLogin Click sender event Handles butLogin Click This method is exe
162. ystem would be broken up into viewpoints as shown in Figure 2 and interviews would be used to gather requirements from the stakeholders for all real viewpoints All Viewpoints Conceptual Real Viewpoints Viewpoints peus Settings Reporting Input MI Ltd Data system system IM Results c Calculation Developer Figure 2 System Viewpoint Hierarchy The requirements of conceptual viewpoints would be based on the services that the conceptual viewpoints were required to provide in order to satisfy the real viewpoints Figure 2 For example the user viewpoint requires a list of markers to be displayed on the user interface the data source viewpoint can meet that requirement by storing data in a database and making it available to the system So in turn the data viewpoint would have a requirement for a table that stores marker information During the project inception stage interviews were carried out to establish the needs of users and the scope of the project Interviews were carried out with MI representatives and prenatal screening consultants The interviews were carried out to find out what the users wanted from the final solution Prior to every interview an interview plan would be created the plan contained a list of interview questions and a description of the goals of the interview The user also had a copy sent to them prior to the interview Appendix F User Requirements shows a sample interview

Courage Chibiya BSc Computer Science (Industry)

Contents

Download Pdf Manuals

Related Search

Related Contents