Home

AutoQuant 8.0.2 Suite User`s Manual

Contents

1. January 2012 Page 52 Media Cybernetics Inc Manual amp Tutorials lt MIN Minimum Projection Icon Displays the active dataset in the Minimum Projection view a Za Sum Projection Icon Displays the active dataset in the Sum Projection view Slice View Icon Displays the active dataset in the Slice View v XY Icon Displays the active dataset in the XY view x Zz XZ Icon Displays the active dataset in the XZ view z YZ Icon Displays the active dataset in the YZ view Ja z XYZ Icon Displays the active dataset in the XYZ view El Copy Current View Icon Copies the active view to a new workspace a Cropping and Reshaping Icon Opens the Cropping and Reshaping Dialog ES Extend Slices Opens the Extend Slices Dialog Resize Image Icon Opens the Resize Image Dialog dad A ao gt eee Background Equalization Icon Opens the Background Equalization Dialog gt E Background Subtraction Icon Opens the Background Subtraction Dialog a Attenuation Correction Opens the Attenuation Correction Dialog a Optical Density Correction Icon Opens the Optical Density Correction Dialog January 2012 Page 53 Media Cybernetics Inc Manual amp Tutorials TI Slice to Slice Alignment Icon Opens the Slice to Slice Alignment Dialog Channel to Channel Alignment Icon Opens the Channel to Channel Alignment Dialog ph Image Algebra Icon Opens the Image Algebra Dialog 3 3D
2. Media Cybernetics Inc Manual amp Tutorials Guidelines for Collecting 3D Image Datasets Overview For transmitted light brightfield and widefield fluorescence microscopy the main considerations are the following e Collection of the optical sections Collection of bias frames Collection of the flat field frame Vibration control It is important to follow the guidelines outlined in the following sections The adage of garbage in garbage out is true for following these guidelines Even though AutoQuant is relatively robust against non adherence to any of these guidelines with each guideline that is not followed and depending upon how severely the guideline is broken the microscopist increases his her chances of having unsatisfactory experimental results It is best to follow each and every guideline to the letter wherever possible Do not violate a guideline for the sake of convenience Break only those guidelines for which you have no choice but to break because of restrictions caused by your experimental conditions Remember that with each guideline broken the risk of having unsatisfactory results increases Optics Alignment Before beginning any procedure be sure that the entire optical train is well aligned Follow the instructions provided by your microscope manufacturer on proper adjustment and alignment of the optical elements This should be performed daily prior to each experiment and should be re che
3. Performing DIC Restoration Use with Differential Interference Contrast datasets only The DIC Restoration feature converts a DIC image into an image that represents the optical thick ness of the specimen Restoration Method This section allows you to select between an Iterative and an Inverse Filter process to restore your dataset The Iterative process will yield better results whereas the Inverse Filter will yield satis factory results in less time January 2012 Page 99 Media Cybernetics Inc Manual Tutorials Explaining the DIC Image Settings Slice Selection In this section select the slice s to deconvolve Choose from All Slices Current Slice or Slice Range If Slice Range is chosen you must enter a range into the From and To text boxes Image Characteristics In this section input the data about your dataset to attain the best restoration results possible Shear Angle This is the angle at which the light in your acquisition equipment meets your object Select the proper angle for your equipment and the shadow box to the right of the drop down menu will mimic the lighting of your dataset Image Noise Select the image noise level that best matches your dataset Low Medium or High Channel to Deconvolve Select the channel to restore If you are restoring a grayscale image the only available option is Intensity if you are restoring a multi channel image select the channel to restore Advanced Settings
4. Previous Result Apply y Use Recommended Expert n a Settings Go to Expert Image Enhancement Y Output Settings Y Show Progress V Base Name 90 Blind raw tif C Save PSF Optics Information Settings OK Modality Widefield Fluorescence Spacings 1 1 x 0 304294 um Obj Lens N 1 3 Al 1 35 Ready x 246 y 4 cl 1147 c2 3557 c3 6164 January 2012 Page 50 Media Cybernetics Inc Manual amp Tutorials Starting at the top and moving clockwise the features indicated are 1 Docking Controls 2 Dia log Box 3 Operation Status 4 Image Enhancement Dialog 5 Data Browser 6 Data Properties 7 Data Manager and 8 Perspective Tabs Below each item will be explained 1 Docking Controls There are two icons in the Docking Controls section The tack 2 con trols whether the dialogs display all the time or whether they only display when they are clicked on If the tack icon is vertical as shown in the previous image then the dialog remains displayed all the time Once the icon is clicked the tack will appear horizontally and the dialog will remain open only as long as the mouse is held over it Once the mouse moves away from the dia log the dialog will slide closed It can then be reopened by placing the mouse over the tab for that dialog located along the side of the application window The default for the tack icon is vertical open at all times The othe
5. The least critical deviation is to make Ad larger than that indicated by the above equation This is because most of the noticeable smearing in images is along the axial dimension which is not affected by making Ad larger The next least critical deviation from this rule is to make Ad smaller than indicated by the above equation In fact there are distinct advantages in doing so Generally speaking the finer the sampling is along z the better is the deblurring along z The above rule is considered the optimal trade off between this axial deblurring and the potential pho tobleaching and other problems such as extra disk space required that may be caused by having finer axial sampling The next least critical deviation is to make Ad larger than indicated by the above equation Doing so lessons the amount of deblurring that is possible along the axial dimen sion but there should still be substantial deblurring that is noticed For example it is common to process datasets that have Ad equal to 2 or 4 times that indicated by the above equation and sub stantial deblurring and noise reduction is still usually obtained albeit generally to a lesser degree Making Ad smaller than as indicated by the above Equation is the most critical deviation from the rule Generally it is best to avoid bending this rule even though it has mixed advantages and disadvantages Making the in plane sampling finer than the Rayleigh resolution limit will improve t
6. This are raw binary file types that can con tain data represented with 8 12 16 or 32 bits per pixel TIFF tif Tagged Image File Format This is a common format for image files January 2012 Page 62 Media Cybernetics Inc Manual amp Tutorials Windows Bitmap File bmp This is a Windows bitmap file often used to store image data on the Windows platform Image Pro Sequence File seq This is an Image Pro file format It is based on the TIFF standard but it includes additional proprietary information used by Media Cyber netics Image Pro software Image Cytometry Standard File ids ics Image Cytrometry is a standard file for mat used in some microscopy applications Bio Rad PIC pic This file format is a proprietary file format used by Bio Rad Bitplane IMS File ims This is a proprietary file format used by Bitplane software Carl Zeiss LSM File lsm This is a proprietary file format used by Carl Zeiss soft ware MetaMorph STK File stk This is a proprietary file format used by Molecular Devices software Olympus OIF OIB File oif oib This is a proprietary file format used by Olympus software Olympus Fluoview File tif This is a file format based on the TIFF standard It includes proprietary information used by Olympus software Scanalytics IP Lab File ipl iplab This is a proprietary file format used by older Scanalytics IPLab software Note
7. This section will explain the user interface for the software explaining what is in each section and how to interact with it More detailed explanations can also be found in the Online Help that is included with the product When you first start the application the user interface will look some thing like this X3 AutoQuant X File View Processing Deconvolution Visualization Analysis Window Help 5698 MEQ aA rs 28 Operation Ready Once a dataset has been opened it will appear as shown in the following page January 2012 Page 49 Media Cybernetics Inc Manual amp Tutorials AutoQuant X File View Processing Deconvolution Visualization Analysis Window Help 5608 B8Qiats lar Gee Vas D Blind raw tif 3D Sum DER 4 Data Manager 1 Dataset eee D 13D Blind raw t OQ wx B pP Q 2 econvolution Methods Adaptive PSF Blind Fixed PSF Nor PSF Settings C Load PSF Dataset Theoretical PSF Spherical Aberrati None Summary g ES 3D Blind raw tif Deconvolution Settings Use Defa Algorithm Settin Volume 402 x 300 x 60 i z E 7 Total Iterations 10 Save Interval Time Points 1 Spacinas 1x1 x 0 30429 um Noise Level Low y gt Noise Value Michi Ch 1 650 nm 650 nm Performance Y Faster Processing Reduced F Ecz Ch 2 520 nm 520 ni E Use Gold s Method Hci Ch 3 4 Iteration Acceleration hindaliks Ulidafiald Chinrannande m Continue From
8. Gray values Minimum Maximum gray Maximum possible value in this frame possible gray gray value value Figure C Histogram of gray values in an image frame that has saturated pixels The maximum gray value of the frame and the maximum possible gray value coincide Typically though not always a spike will appear in the histogram as shown above If you observe saturated pixels lower the exposure time and repeat the grabbing and viewing pro cedure As a rule of thumb adjust the exposure time so that the brightest pixel is at approximately 75 of the well capacity of the CCD camera This condition is often detectable depending upon the sam ple again by using a histogram The 75 well capacity condition is detected by noting that the highest abscissa value having a non zero histogram value is at 75 of the range of the abscissa An example of this is shown in Figure D below January 2012 Page 23 Media Cybernetics Inc Manual amp Tutorials n O o S D g o o O a E 3 Z Gray values Minimum Maximum gray Maximum possible value in this frame possible gray gray value value Figure D Histogram of gray values in an image frame with no saturation Adjust the camera frame grabber gain and or illumination level such that the maximum gray value in the image frame is between 75 to 85 of the maximum possible gray value For instance for an 8 bit camera most RS 170 camera frame grabbers the
9. If you have Image Pro installed on your machine you can export files to it for analysis To launch Image Pro select the Image Pro icon on your toolbar The Image Pro icon and menu item are only available if you have Image Pro installed on the same machine as AQX When you export a file to Image Pro it will be saved as a 16 bit seq sequence file and opened in Image Pro Multi channel files are opened as several single channel files The Image Pro Color Composite tool is opened so that the files can be re merged January 2012 Page 65 Media Cybernetics Inc Manual amp Tutorials Export to Imaris If you have Imaris installed on the same machine as AQX you will have an option on the File menu called Export to Imaris as well as an icon on the toolbar If you do not have Imaris installed on your machine you will not have these options If you have these options load a dataset into the AutoQuant X workspace Click the Imaris icon or select Export to Imaris from the File menu The dataset is saved as an ICS ICD pair and opened in Imaris If the maximum intensity of the dataset is equal to or less than 65536 16 bit limit the dataset will be transferred as 16 bit integer data If the maximum intensity of the dataset is larger than 65535 it will be transferred as 32 bit floating point data You can force data to be saved as 32 bit floating point by going to the User Options and changing Export to Imaris to FLOAT_32 Export to Meta
10. Pollen deb 0 7 1 515 Widefield 90 80 110 0 357 0 3636 0 4 Colensc tif 1 4 1 5 Widefield 256 256 1 2D dataset Brain avz MRI 256 256 64 Hip avz 3 CT scan 512 512 90 0 9 0 9 4 8 SmallHip avz CT scan 128 128 120 3 6 3 6 3 6 RotatedHip avz CT scan 128 128 120 3 6 3 6 3 6 Vicera avz CT scan 128 128 120 3 6 3 6 3 6 Skinless avz CT scan 128 128 120 3 6 3 6 3 6 HipMovie avm Movie CT scan 128 128 120 3 6 3 6 3 6 VoxelTLB avm Movie Brightfield 130 130 41 0 279 0 279 1 0 TLBview 1 3 1 515 Brightfield 198 208 64 0 653 0 0653 0 4 Star 0 0 25 1 0 Brightfield 494 348 46 0 6897 0 6897 8 0 Star 1t0 0 25 1 0 Brightfield 494 348 1 0 6897 0 6897 8 0 Star bs1 0 25 1 0 Brightfield 494 348 1 0 6897 0 6897 8 0 Star bs2 0 25 1 0 Brightfield 494 348 1 0 6897 0 6897 8 0 Starfish_crop2 deb 0 25 1 0 Brightfield 128 128 54 0 6897 0 6897 8 0 TobaMC tif 1 4 1 515 Confocal 128 128 32 0 23 0 23 0 4 Neuron deb 0 8 1 515 Confocal 384 512 64 0 94 0 94 1 0 Neuron _crop1 deb 0 8 1 515 Confocal 128 128 64 0 94 0 94 1 0 RedNuc 1 3 1 33 Confocal 128 512 64 0 07 0 07 0 15 January 2012 Page 152 Media Cybernetics Inc Manual amp Tutorials Table 2 File Name Numerical Refrac Modality X Y and Z X Y and Z Spacing Aperture tive Dimensions Pixels Microns Index Time Series Data 1 32 1 4502 Widefield 256 256 23 32 32
11. and the numerical aperture of the lens Of course according to good laboratory practice it is always best to avoid saturation but this property of the algorithm makes data collection and experiment design much easier since the avoidance of saturation is now a flexible requirement and no longer an absolute requirement Q Are noisy poor signal to noise images going to be problem for AutoQuant A AutoQuant is able to process extremely noisy images and deliver restored image sets with very little noise Noisy images are common with a confocal microscope an intensified camera or whenever low light is used to minimize bleaching Q How many iterations should I apply How do I know I have performed sufficient itera tions What happens if I perform too many iterations A Start with the rule of thumb With the Use Recommended Expert Settings box checked for widefield and brightfield use 10 iterations for confocal two photon and spinning disk use 10 iterations If by inspecting the images you are satisfied with the resolving power stop If you want still better resolving power increase or double the numbers Repeat this increasing until you are satisfied with the resolving power You will know you went too far if you notice obvious noise and arti facts which will be clear because structures will be shown which clearly cannot be present and which do not have remnants seen in the raw images If you notice such noise and artifacts cut the nu
12. use the Image Algebra feature The volumes must be the same dimensions The available Image Operations are Addition Subtrac tion Multiplication Division AND OR and XOR If Subtraction is selected the Second Image will be subtracted from the First Image If Division is selected the First Image will be divided by the Second Image The two original files will remain intact and a new data file will be created with the processed results To use the AND OR and XOR one of the datasets must be optically sliced in a Slice Viewer projection in the XY view and segmented Image Algebra ax Scale First Image Constant lee aso i Operation Scale Second Image Constant 1 ral 0 qu 1 Go to File Open From the Widefield directory open the PollenPlus5 deb dataset 2 From the Processing menu select Image Algebra When the Image Algebra dialog appears from the First Image drop down menu select PollenPlus5 deb and select Pollen deb from the Second Image drop down menu Leave the Scale for both the First and Second Image at 1 This is the default value The Constant Value field will be inactive January 2012 Page 90 Media Cybernetics Inc Manual amp Tutorials Note To add a Constant Value to an Image select None for the Second Image and enter a number in the Constant Value field The Constant Value is for intensity offset Every pixels intensity will be either decreased by entering a negative number or increa
13. 2 the Airy disk width with a wavelength that is the excitation wavelength The excita tion optics determine the width of the PSF This will be approximately 0 614 NA x 2 Where A is the emissive wavelength and NA is the numerical aperture Generally the result is 163 nanome ters or finer assuming a 1 4 NA lens If you can sample finer say 80 nanometers or 40 nanome ters try this also as an experiment It is expected that you may improve the x y resolution by at least 2 and possibly by 4 80 nanometers To improve the optical resolution by 4 to 80 nano meters you need to increase the spatial sampling by 8 40 nanometers Sampling Criterion for Z Sample at 1 2 of the PSF axial width which approximately follows the formula a AJMANA Where n is the refractive index of the lens and NA is the numerical aperture This is the formula for the axial width of a confocal PSF For 2 photon microscopy use the excitation wavelength in this formula Assuming a 1 4 NA lens this distance is approximately 800 nanome ters For more substantial deblurring along z experiment with even finer sampling For example 400 or 200 nanometers or finer Keep in mind that the above rules are only rules of thumb for ensuring that deconvolution will show substantial deblurring If on the other hand the experimental objectives do not call for an improved x y resolution but they do call for an improved z resolution then follow the Sam pling Criteri
14. 2012 Page 131 Media Cybernetics Inc Manual amp Tutorials Tutorial 1 Load Tutorial_ Data Multi Channel Colorpollenc tif 2 Launch AutoQuant 5D Viewer 3 Highlight Volume in the object tree 4 Change the volume s gamma to 2 and press enter This should brighten both channels 5 Highlight Ch 2 520 nm under Volume 6 Change the first channel s gamma back to 1 and press enter This should keep the red channel bright but the green channel dim January 2012 Page 132 Media Cybernetics Inc Manual amp Tutorials Navigation The navigational mode toolbar allows you to change between rotation selection and zoom modes Rotation is the default mode for AutoQuant 5D N The image can be rotated by clicking the left mouse button dragging the mouse and releasing the left mouse button While in rotation mode right click and drag to zoom in and out Zoom is set as the current A navigation mode by selecting the zoom button Left click and drag while in zoom mode will zoom the image The button enables selection mode Zoom and rotation are disabled In selection mode the region of interest ortho slices and oblique slices can be moved Keyboard Hotkeys e Home Brings the image back to default rotation and zoom Up arrow rotates the scene up counter clockwise around the x axis Down arrow rotates the scene down clockwise around the x axis e Right arrow ro
15. Clicking the Advanced button will open the Expert Settings dialog Only change these settings if you are experienced with the DIC Restoration feature Otherwise use the default settings 1 Navigate to the DIC folder and open test tif 2 From the Deconvolution menu select DIC Restoration 3 In the Restoration Method section ensure that Iterative is selected 4 In the DIC Image Settings select All Slices the current image is only 1 slice 5 Still in the DIC Image Settings section select 135 from the Shear Angle drop down menu Once the angle is selected a sample will appear to the right of the drop down menu the light pat tern in this image should match that in the dataset 6 Select Low for Image Noise 7 Click Advanced to view the advanced features Verify that Total Iterations is set to 15 Verify that the Regularization Iterations and Regularization Contribution are both set to 10 Make sure all of the Processing Options are checked 8 Click the OK icon Y The results will display in a new dataset window January 2012 Page 100 Media Cybernetics Inc Manual amp Tutorials Expert Settings Note This section applies to users of AutoQuant only 1 2 3 4 5 Navigate to the Widefield folder and open FitcDapi_crop tif Go to the 3D Deconvolution dialog Uncheck Use Default to expand the deconvolution settings Uncheck Use Recommended Expert Settings Click the Go t
16. Convergence of EM Image Reconstruction Algorithms with Gibbs Smooth ing IEEE Transactions on Medical Imaging 9 4 439 446 Llacer J and E Veklerov 1989 Feasible Images and Practical Stopping Rules for Iterative Algorithms in Emission Tomography IEEE Transactions on Medical Imaging 8 2 186 193 errata 9 1 112 1990 Lucy L B 1974 An Iterative Technique for the Rectification of Observed Distributions The Astronomical Journal 79 6 745 765 Macias Garza F K R Diller A C Bovik S J Aggarwal and J K Aggarwal 1989 Improvement in the Resolution of Three Dimensional Data Sets Collected Using Optical Serial Sectioning Journal of Microscopy 153 2 205 221 Martin L C and B K Johnson 1931 Practical Microscopy Blackie and Son London January 2012 Page 157 Media Cybernetics Inc Manual amp Tutorials Miller M I and B Roysam Bayesian Image Reconstruction for Emission Tomography Incor porating Good s Roughness Prior on Massively Parallel Processors Proceedings of the National Academy of Sciences Vol 88 No 8 pp 3223 3227 April 1991 Miller M I and D L Snyder 1987 The Role of Likelihood and Entropy in Incomplete Data Problems Applications to Estimating Point Process Intensities and Toeplitz Constrained Covari ances Proceedings of the IEEE 75 892 907 Oppenheim A V and R W Schafer 1975 Digital Signal Processing Prentice Hall Engle wood Clif
17. Media Cybernetics Inc Manual amp Tutorials When a minimum projection is selected the background is automatically set to white It can be set back to black or another color by selecting Scene in the object tree and changing its Background color parameter Sum projection Each slice is added together to give a composite rendering of the entire volume This makes it easier to see trends in the data but many times background is summed and makes it hard to see features e Alpha blend projection Each slice in the image is blended together from back to front Details far away from the point of view impact the final rendering the least Details close to the point of view influence the final rendering the most This gives a true to life rendering of the image January 2012 Page 136 Media Cybernetics Inc Manual amp Tutorials When an alpha projection is selected the background is automatically set to grey It can be set back to black or another color by selecting Scene in the object tree and changing its Background color parameter The projections can be changed from the projection tool menu as well wax Max ws Min de al Alpha Sum The projection tool menu and the volume object s projection parameter are linked Storage Hint Render the volume as a 3D or 2D texture 3D is the default and recommended mode Volume size Change the subsampled volume size If the original dataset s width or hei
18. No Neighbor deconvolution methods The execu tion speed of the Inverse Filter is between that of Nearest Neighbor and Blind Deconvolution The Inverse Filter should be used in cases where speed is important A typical processing time is under 2 minutes with a 256x256x32 dataset Pentium III 450 Mhz 1 Navigate to the Widefield folder and open Pollen deb 2 From the Deconvolution menu select 3D Inverse Filter The Inverse Filter Dialog will appear January 2012 Page 96 Media Cybernetics Inc Manual amp Tutorials 3D Inverse Filter ax 3D Blind raw t1f 1 Input PSF Load PSF Dataset y Theoretical PSF amp Spherical Aberration None Algorithm Settings Noise level Medium v Phase content expected Output Settings Y Image 3D Blind raw tif g PSF Q Ly q 3 Select Medium from the Noise Level drop down menu 6 The Phase Content Expected should be used if the specimen is a Transmitted Light Bright field dataset and exhibits significant phase characteristics e g areas of the specimen appear brighter than the background For this dataset leave the Phase Content Expected box unchecked 7 Click the OK icon Y AutoQuant will now execute the Inverse Filter deconvolution The status bar will indicate the progression of the Inverse Filter 8 When the Inverse Filter method has finished a new XY Max Projection view will appear This is the result of the Inverse Filter
19. Save PSE AAA Optics Information Settings OK Modality Widefield Fluorescence Spacings 1x1 x 0 304294 um Obj Lens NA 1 3 Al 1 35 Emw 650 520 450 nm Subyolume Information Number of Subyolumes 60 240 180 1 Available Memory MB 1279 Available DiskSpace MB 13126 Confocal Defaults Nyquist Change Defaults Z Q v 8 January 2012 Page 95 Media Cybernetics Inc Manual amp Tutorials 3 Make sure that Adaptive PSF Blind is selected in frame 1 of the dialog 4 Select Theoretical PSF in frame 2 of the dialog 5 In frame 3 set the Total Iterations to 10 and check the Faster Processing Reduced Resolu tion box Ensure that the Use Recommended Expert Settings box is checked 6 In frame 4 leave the Base Name as 2D Data t1f 7 Click the OK button Y The 2D Deconvolution will start The results will be displayed in the main viewing window Using the Inverse Filter Use with Widefield Fluorescence and Transmitted Light Brightfield datasets only Inverse Filter does not work with Confocal datasets The Inverse Filter is a one step non iterative deconvolution method based on inverse filtering the ory It utilizes optimal linear filtering Inverse Filter is one of the simple deconvolution methods offered in AutoQuant It is very useful for obtaining quick results but is not as accurate as Blind Deconvolution Inverse Filtering is typ ically more robust than the Nearest Neighbor or
20. This Guardband prevents artifacts at the seams of these subvolumes The possible values are integers from 0 to N 2 where N is the depth of the XZ or YZ field Guidelines The Z Guardband should never be larger than the subvolume overlap region A value of 6 is adequate for most image stacks Intensity Correction Selecting this option will correct for differences in intensities between slices The default setting for this option is unchecked for Confocal images and checked for Widefield and TLB images It works much like the Optical Density Correction feature in the Processing menu Minimum Intensity Removal The Minimum Intensity Removal feature will subtract the minimum intensities from the image Thus with Minimum Intensity Removal selected an image that has intensities ranging from 11 245 will be adjusted to have intensities of 0 234 The default for Minimum Intensity Removal is Yes Accelerated SA Detection Selecting Accelerated this is selected by default will accelerate the process of detecting the spherical aberration of the dataset Accuracy is slightly better with acceleration not selected but the trade off of speed with the acceleration is greater than the accuracy compromise Pre Condition Imported PSF This option will pre condition an imported PSF Object First Guess The Object First Guess has four options Flat Sheet Original Data Object Input and Filtered Ori gnal The First Guess selects the initial estimate of the o
21. XY field in pixels whichever is smaller Guidelines An overlap of 10 or 25 pixels usually works best If the result of the deconvolution contains artifacts having rigid lines edges or an obvious grid structure start with a value of 10 If doing so reduces the problem but does not eliminate it then increase this number again Overlap ping regions are deconvolved twice so making this number too large e g 100 will increase the deconvolution time January 2012 Page 104 Media Cybernetics Inc Manual amp Tutorials XY Guardband Pixels The Guardband size defines the width of a border surrounding each subvolume This border is the region where the subvolumes are processed to seam them together This guardband prevents arti facts at the seams The possible values are integers from 0 to N 2 where N is the width or height of the XY field in pixels whichever is smaller Guidelines Generally the larger the guardband the fewer the artifacts and the better the image quality However deconvolution time increases with guardband size so a default value is set which minimizes deconvolution time and eliminates artifacts in most cases The default value is 10 This number may be increased if seaming artifacts appear If so first increase the number to 15 then 20 and then 25 until the seaming artifact is gone Z Guardband Pixels The Z Guardband specifies the number of slices that will be added at the top and bottom of the subvolume
22. adverse laboratory condi tions temperature changes room lights being turned off and on and movement in the room caus ing shadows over the microscope Although the source of these flicker effects is difficult to pinpoint exactly it is certain that they occur and that they occur regularly 7 Close all views before continuing on to the next section Aligning an Image Note This feature including Manual Automatic and Channel to Channel Alignment is only available as an added plug in Contact your dealer for purchase information January 2012 Page 85 Media Cybernetics Inc Manual amp Tutorials For several reasons a 3D dataset can be misaligned and in several different ways There can be shift in between slices a live cell image can move during acquisition or rotate or even become warped and a multi channel dataset can have misalignment from channel to channel The Image Alignment feature can correct for these problems producing an aligned image for easier viewing and more accurate analysis Slice to Slice Automatic Alignment This feature corrects for X amp Y translation and warp nonlinear distortion between frames Mis alignment and warp can be caused by stage motion movement of the sample during live imaging and other causes 1 Open the Unaligned avz file from the Ocular folder in the Tutorial Data directory 2 From the Processing menu select Image Alignment the select Slice to Slice The Image A
23. and click on it 3 The dataset window will refresh in Sum projection mode For information on how to select Projections using the View menu see the View Menu chapter Selecting a Single View As with the Projections Single Views can be selected both from the View menu and from the toolbar in the dataset window but additionally for Single Views the dataset window contains tabs with which single views can be opened and closed 1 In the dataset window click the Single View icon y and the available single views appear in the drop down menu 2 Select the XZ single view and click on it 3 The dataset window will refresh in the XZ view Alternatively 1 With the dataset still in the XZ view click on the XY Tab on the top of the dataset 2 The XY view will open above the XZ view For information on how to select Single Views using the View menu see the View Menu chapter 3 Perspective Tabs These tabs located along the right side and bottom of the image display window open new perspectives of that dataset within the same image display window The tab along the right side opens and closes the YZ perspective and the tabs along the bottom open and close the XY and XZ perspectives Clicking on one of these tabs will open that tab s perspective Once opened clicking on that tab again will close that perspective Creating a Region of Interest Several functions in AutoQuant and 5D Viewer require a region of interest to be c
24. and used for publication or presentation purposes 6 Click the Track button The program will track the counted objects through all of the time points The statistics from these tracks will be displayed on the Tracking Grid and Tracking Sum mary Statistics in the statistics frame LJ 7 Click the Save a icon which will save all of the statistics into an xls file format so it can be opened and manipulated in a spreadsheet Track Selected Objects This option allows you to track a small number of selected objects To select an object click on it in the Counting and Tracking preview window or highlight it in the Object Stats tab of the grid control in the Counting and Tracking PlugIn Dialog You have the option to specify the maximum displacent of the object between time frames If the object moves farther then the maximum the track will end You may also specify a maximum deformation number between 0 and 100 that is the percentage energy difference of the feature vector of the object between time frames If the object deforms too much the track will end January 2012 Page 111 Media Cybernetics Inc Manual amp Tutorials Preprocessing The Preprocessing menu contains tools for refining a dataset before performing the Counting and Tracking functions on it Clicking the down arrow next to Preprocessing will display the Prepro cessing tools e Background Equalization Move the slider to adjust the amount of background equ
25. are treated as decoration objects They can be toggled on or off from the tool menus or from the object tree Toolbar AutoQuant 5D Viewer contains a toolbar for some of the more common actions performed on the dataset Rotate 2 When the Rotate button is depressed use the left mouse but ton to click and drag on the object to rotate it along any axis e Select To select an oblique slice or orthogonal slice click this icon then click on the desired slice plane Projections To select a projection mode click on the icon to see the drop dow menu Maximum Projection Sum Projection and Alpha Blend Projection e Contrast Click one of the icons to adjust the brightness and contrast in the Viewer e Zoom When the Zoom button is depressed hold down the left mouse button and move the mouse from left to right and back Dragging the mouse to the right will zoom out dragging the mouse to the left will zoom in Use the right mouse button to click and drag on the object to rotate it along any axis January 2012 Page 129 Media Cybernetics Inc Manual amp Tutorials If the dataset is more than one channel view objects will have child objects one for each channel M Scene Camera Region of Interest Scale Bar M Volume Ch 1 520 nm Ch 2 620 nm Each channel can be turned on or off by using its associated check box Tutorial 1 Load Tutorial _Data Widefield BlueGreen stk 2 Launch AutoQuan
26. as valuable property Please save the following information for future use 1 GRANT OF LICENSE Subject to the terms of this Software License Agreement Media Cybernetics Inc Inc Media hereby grants to a single user Licensee non transferable except as expressly provided in Section 4 below and non exclusive license to use and execute the Auto Quant software SOFTWARE on a single computer or facility COMPUTER at any time during the period of validity of this license This software may only run on one computer If you have a multiple user license the SOFTWARE may be run simultaneously by no more persons than the number of users licensed 2 COPYING RESTRICTIONS In order to affect your license rights granted in Section 1 above you may install the SOFTWARE by copying it onto the hard disk drive of one computer You may make a full or partial backup or archival copy ofthe SOFTWARE for your own use or use at the facility to which the SOFTWARE is licensed You AGREE that 1 your use and pos session of such copies shall be solely under the terms and conditions of this Agreement and 11 you shall place the same proprietary and copyright notices and legends on all such copies as January 2012 Page 8 Media Cybernetics Inc Manual amp Tutorials included by Media on the media containing the authorized copy of the SOFTWARE originally provided by Media You may not copy or provide copies of the SOFTWARE in whole or in
27. deconvolution can be performed at once Generally however many CPU s the computer has is the maximum number of operations that should be allowed Transfer to Imaris Determines what bit depth should be used to transfer files to Imaris Your options are e UNSIGNED _INT16 WHEN POSSIBLE If the maximum intensity of the dataset is equal to or less than 65536 16 bit limit the dataset will be transferred as 16 bit integer data If the maximum intensity of the dataset is larger than 65535 it will be transferred as 32 bit floating point data FLOAT _ 32 Forces data to be transferred to Imaris as 32 bit floating point data in all cases If Imaris is not installed changing these options will have no effect on the behavior of AQX To change any of these settings 1 From the file menu select User Options 2 Click on Temporary File Directory This icon al will appear click on that icon A Windows Browser will appear 3 Scroll to the desired location highlight that folder and click OK 4 Repeat this procedure for any settings that need to be changed 5 When all changes have been made click on the OK icon Y Or to discard the changes click the Cancel icon t This will cancel all changes and close the User Options dialog Exiting AutoQuant 5D Viewer This function closes the software and all files that are open You do not need to save the images generated in this tutorial section From the File menu select Exit If there are an
28. download the FRET Application Note from the Support section Select the desired algorithm by clicking on its radio button FRET Images and Calibration for Cross Talks This section makes associations between the images and excitation emission specimen set To optimize the use of the FRET feature the proper datasets must first be obtained Below is a list of headings and the images they correspond to that will allow for optimal use of the FRET feature when naming your datasets make sure you know which one corresponds to which excitation emission specimen set Headings and Corresponding Datasets DDf Donor excitation and Donor emission applied to the FRET specimen DAf Donor excitation and Acceptor emission applied to the FRET specimen AAf Acceptor excitation and Acceptor emission applied to the FRET specimen DDd Donor excitation and Donor emission applied to the Donor specimen DAd Donor excitation and Acceptor emission applied to the Donor specimen January 2012 Page 114 Media Cybernetics Inc Manual amp Tutorials AAd Acceptor excitation and Acceptor emission applied to the Donor specimen DDa Donor excitation and Donor emission applied to the Acceptor specimen DAa Donor excitation and Acceptor emission applied to the Acceptor specimen AAa Acceptor excitation and Acceptor emission applied to the Acceptor specimen Not necessary but can be used with the Gordon and Herman and Media Cybernetics Inc Inc algorithm
29. error message will appear stating The dimensions of this image are different from the other image Do you want to change to this new image for colocaliza January 2012 Page 123 Media Cybernetics Inc Manual amp Tutorials tion analysis Selecting Yes will load the new image and clear the other image at which point a second channel will have to be selected Selecting No will not load the new channel and the chan nel loaded for the other image will remain loaded Next to each image dropdown menu is a dropdown menu with a color selector The channel selected in this dropdown menu will determine how the channel is displayed in the Coloc Viewer The options are Red Green and Blue Once a color is selected for one channel that color will not be available on the dropdown menu for the other channel Colocalization Preview The Colocalization dialog controls the Colocalization Preview which appears in the AutoQuant X2 workspace Tools This section contains tools to manipulate the Colocalization Preview X Q The zoom controls will zoom the image in and out The button on the left decreases the image and the button on the right increases its size Holding either of these buttons down will continue to either increase or decrease the image s size depending on which button is being pressed The dropdown menu displays the image size relative to its original size as a percentage Alternatively it is possible to either select
30. for High Speed CCD Data Correction because Dark Current does not effect the image in this situation 6 Enter the value for the frame s per optical slice Leave the default setting of 1 for this example For Bad Pixel Treatment select Median Filter Click the OK icon Y and AutoQuant will correct the High Speed CCD dataset Note To slow down the viewing time per each frame increase the frame s per optical slice value Correcting Attenuation If your dataset suffers attenuation due to any of the following issues e Excitation light absorption Fluorescent emission light absorption e Photobleaching Changes in the optical point spread function as a function of depth which in turn is fundamentally due to refractive index mismatches between the sample embedding medium coverslip and lens immersion medium The Attenuation Correction feature can correct the attenuation as a function of depth into the sam ple To determine if a dataset requires Attenuation Correction look for darkening within an image as you go from the top to the bottom of the view 1 From the File menu select Open Navigate to the Confocal folder and select RedNuc 0 2 From the View menu select Projections then select Sum projection From the View menu select Single View and click on XZ to generate an XZ View of the Sum Projection Notice how the dataset appears darker as you go from the top to the bottom of the view 3 From the Processing menu select A
31. how many slices you want to add to the top and bottom of your sam ple In the Extend Depth box enter 4 for the Top and Bottom field values The new Depth value will be displayed in the New Dimension frame Leave the Zero Pad box unchecked 2 Click the OK icon Y The extended dataset will be displayed Click on the XZ view icon and notice the difference between the real data slices and the false slices just generated 3 Close all views before going on to the next tutorial Resizing an Image In order to reduce data storage requirements reduce pixelation blockiness in small datasets and improve rendering a dataset can be resized using the Resize Image feature This function allows you to change the overall size of an image There are two methods for doing so the Linear and the Ideal method The Ideal method is recommended however the Linear method is somewhat faster than the Ideal method Resizing a dataset resamples the data to produce a new stack of optical sections with dimensions different than those of the original stack unlike changing the size of the image on the screen Note Resizing an image and Sizing zooming an image are different functions January 2012 Page 77 Media Cybernetics Inc Manual amp Tutorials 1 Navigate to the Widefield folder and open FitcDapi_crop tif 2 From the Processing menu select Resize Image The Resize Image dialog box appears This dialog box contains three ways in whic
32. in any form by any means electronic mechanical by photocopying recording or otherwise in whole or in part without the written permission of Media Cybernetics Inc SOFTWARE United States copyright laws and international treaty provisions protect the Auto Quant or AutoQuant 5D Viewer software package It is illegal to copy the software except as specifically allowed in the License Agreement stated below Any person found storing or using an unlicensed copy of this software is liable to be sued by Media Cybernetics Inc to the maximum extent permissible under applicable US and International laws and International Treaty Provi sions Software License Agreement This is a legal agreement between you either an individual or an entity the end user and Licensee and Media Cybernetics Inc Media or we the Licensor BY OPENING THE SEALED MEDIA PACKAGE or by installing this software on a computer YOU ARE AGREE ING TO BE BOUND BY THE TERMS OF THIS AGREEMENT By doing so the user is acknowledging that he she has carefully read and considered all the provisions of this Software License Agreement and agrees they are fair and reasonably required to protect Media s interests The end user thereby agrees to all of the terms spelled out in this Agreement IF YOU DO NOT AGREE TO THE TERMS OF THIS AGREEMENT RETURN THE UNOPENED MEDIA PACKAGE AND THE ACCOMPANYING DOCUMENTS TO Media The following informa tion is important and should be treated
33. is iterative and nonlinear it is capable of properly constraining the solution 1 e constraining the probe concentration and PSF solu tions to be nonnegative This is especially important for both quantitation and for resolution improvement especially z resolution Our algorithm is capable of optimally restoring the miss ing cone of frequencies that are inherent in the PSF for widefield microscopes For a detailed description of the problem of missing cone of frequencies the reader is referred to the paper by Streibl see Bibliography Linear methods such as the well known Nearest Neighbor method and the Inverse Filtering methods see references are not capable of implementing such constraints and are thereby not nearly as suitable for quantitation or resolution improvement However imple mentations of both of these simple and fast deconvolution methods are included in AutoQuant January 2012 Page 46 Media Cybernetics Inc Manual amp Tutorials Dark Current Bias and Flatfield Corrections The data correction utility is a component in AutoQuant This provides features for automatic data correction including automatic bias and flatfield correction This component corrects data from video rate and cooled CCD cameras Processing Many Datasets AutoQuant allows the sequential processing of many datasets by using its Batch Processing feature New in Version X3 This version of AutoQuant contains many improvements including
34. maximum gray value is 255 For a 12 bit cooled CCD camera the maximum possible gray value is 4095 If your cooled CCD camera provides gain adjustments or frame averaging follow the manufac turer s directions to achieve the above described intensity histogram profile RS 170 Cameras Many RS 170 cameras provide some control over the video gain Some cameras provide a gain adjustment device on the camera itself while others provide this adjustment as part of a camera electronic control box Often frame grabber cards provide a software selectable gain factor Finally the lamp brightness control and the neutral density excitation filter control provide yet two more mechanisms for selecting the gain Please look through the manuals provided by your microscope camera and frame grabber manufacturers to determine which if these options are available to you and methods to access them Similarly cameras and or frame grabber cards provide an offset black level adjustment mecha nism Again please refer the manufacturers manuals on methods to access the offset adjustment January 2012 Page 24 Media Cybernetics Inc Manual amp Tutorials Once you have determined the above information the following procedure is recommended for setting these adjustments 1 Set the illumination light intensity for transmitted light brightfield or the neutral density filter for fluorescence first Do this by direct eye viewing and selecting th
35. of the oblique slice Normal X Y and Z Controls the orientation of the axis on which the oblique slice sits January 2012 Page 139 Media Cybernetics Inc Manual amp Tutorials The oblique slice can be animated by selecting Set Start Frame from the movie tool menu moving the slice and then selecting Set End Frame from the movie tool menu When the movie is played the slice is animated between the two locations January 2012 Page 140 Media Cybernetics Inc Manual amp Tutorials Region of interest By default the entire extents of the object are displayed The object can be subvolumed by entering selection mode El toolbutton and then clicking and dragging any of the green control boxes at the edges of the volume The exact locations of the sides can be set within the parameters section of the Region of Interest object The region of interest can not be animated Scalebar Scalebar for the image appears at the lower right hand side of the image window The scale bar can be turned on or off from the object tree or decoration menu toolbar The scale bar responds to spacing changes in the data manager If no spacing information is available in the data manger then pixel values are displayed Axis Displays the X axis as red the Y axis as green and the Z axis as blue Grid The grid is displayed on the back sides of the image s bounding box The grid is linked to the scale bar s markings and w
36. open BILTI Num 01 tif and BILT2_Den 01 tif For each file it will ask which pattern denotes the slice number select the pattern in which the pound sign falls at the end of the file name 6 Assign BILT Num 01 tif to the Numerator and assign B LT2_Den 01 tif to the Denomi nator in the Images frame of the Ratiometrics dialog 7 Verify that the Remove Spot Noise feature has been checked Click Calculate A new image will appear This is the result of the ratio 8 Click the Analysis button in the Ratiometrics dialog to display the statistical information on the new image 9 Close all images before moving on to the next section Colocalization Note This feature is only available as an added plug in Contact your dealer for purchase infor mation Often times in multi channel datasets two separate dyes will occupy the same area In order to display the areas in which two dyes become colocalized use the Colocalization feature This function displays and analyzes the overlay of two separate dyes in a multi channel dataset To use Colocalization a multi channel dataset must first be opened Image 1 Image 2 In the mage I and Image 2 sections the channels to be analyzed are selected and assigned a color in which to be displayed All available datasets will appear on the dropdown menus so make sure that the channels selected are the same dimensions If 2 channels with different dimensions are loaded into these dropdown menus an
37. other rights of third parties that may result from its use No license is granted by implication or otherwise under any patent rights of Media 9 CHOICE OF LAW This notice will be governed by and interpreted in accordance with the laws of the State of Maryland in The United States of America 10 US GOVERNMENT RESTRICTED RIGHTS Use duplication or disclosure by the US Government is subject to restrictions as set forth in subparagraph c 1 i1 of the Rights in Technical Data and Computer Software clause at DFARS 252 227 7013 or subparagraphs c 1 and 2 of the Commercial Computer Software Restricted Rights at 48 CFR 52 227 19 as appli cable Manufacturer of this SOFTWARE is Media Cybernetics Inc 4340 East West Highway Ste 400 Bethesda MD 20814 11 COPYRIGHT NOTICE Copyright 2003 2011 by Media Cybernetics Inc Inc The SOFTWARE package is protected by United States copyright laws and international treaty provisions It is illegal to copy the software except as specifically allowed in this Software License Agreement Any person found storing or using an unlicensed copy of this software is lia ble to Media to the maximum extent permissible under applicable US and International laws and International Treaty Provisions Should you have any questions to this Agreement please contact Media Cybernetics Inc 4340 East West Highway Ste 400 Bethesda MD 20814 301 495 3305 12 TERM This Software License Agreement shall remain
38. part to any other party 3 SOFTWARE USE RESTRICTIONS a Reverse engineering and tampering You may not reverse engineer de compile modify translate or disassemble the SOFTWARE Any tamper ing with the license protection system such as the license protection key that attaches to your printer parallel port or of the SOFTWARE is not permitted Violators will be prosecuted to the maximum extent permitted by applicable laws b Not a Medical Device This software is not intended for usage as a Medical Device The definition of a Medical Device appears in section 201 h of the FD amp C Act A Medical Device is an instrument apparatus implement machine contrivance or other related article including a component part or accessory which is 1 recog nized in the official National Formulary or the United States Pharmacopoeia or any supplement to them 11 intended for use in the diagnosis of disease or other conditions or in the cure mitiga tion treatment or prevention of disease in man or other animals or 111 intended to affect the structure or any function of the body of man or other animals and which does not achieve any of its primary intended purposes through chemical action within or on the body of man or other ani mals and which is not dependent upon being metabolized for the achievement of any of its pri mary intended purposes c Usage with Confocal Microscopes This software may be used for visualizing or anal
39. pixel to twice the above mentioned optimal sample spacing 0 5 micrometers for a 1 2 NA lens or higher Of course other experimental conditions may prevent you from doing this and this may be alright but as with other rules of thumb this is the so called recommended optimal setting Signal To Noise Considerations AutoQuant will operate with a wide range of signal to noise levels The Expert Settings have a Noise Smoothing feature to handle different signal to noise levels Widefield datasets generally contain low levels of noise therefore the default setting is Low Confocal datasets generally con tain medium levels of noise therefore the default setting is Medium For Signal to Noise Ratio SNR levels of 10 or below a Noise Smoothing setting of High should be selected This will cause AutoQuant to be extremely robust against the noise For SNR levels between 10 and 100 the Noise Smoothing setting of Medium should be selected For SNR levels of 100 or higher the Noise Smoothing setting of Low should work well See Table 3 to identify noise levels approximately by visual comparison Table 3 SNR 160 1 SNR 40 1 SNR 10 1 Table 3 Left to Right Simulated SNR of 160 1 40 1 and 10 1 respectively January 2012 Page 38 Media Cybernetics Inc Manual amp Tutorials 2 Photon Dataset Collection Guide for Obtaining the Correct Collection Parameters Sampling Criterion for X Y Sample at 1
40. star bs0 from the drop down list Click Open 6 To Load the Bias Field 2 file check the Bias Field 2 box and select star bs1 from the drop down list 8 For the tutorial data enter the following parameters Average 0 03 Flat Field 0 03 Bias Field 1 60 Bias Field 2 120 Note When using your own datasets these numbers should have been recorded and archived when the image stacks were collected The Average exposure time is the exposure time of the CCD camera that was set while the 3D dataset was collected To learn how to obtain these numbers refer to the operator s manual of your camera check the documentation of the software used to collect the image stacks or contact the camera and data collection software manufacturer 9 The Bad Pixels Treatment setting should have the default value of None Note Bad pixels occur when the dark current of a pixel is so severe that it causes a bright spot in the picture Most image stacks do not require any treatment of bad pixels Therefore the recom mended default selection is None If the selection of None causes the corrected pictures to become black this is an indication that the dataset requires the treatment of the bad pixel s In this case the recommended Bad Pixel Treatment setting is Median Filter This is the most accurate means of treating bad pixels It replaces the bad pixel s intensity with the median value of its nearest neigh boring pixels The Wiener Regularization settin
41. than one will emphasize dim features A gamma smaller than one will suppress dim features e Min Max Threshold Used to include or exclude certain intensity ranges By increasing the minimum threshold background is removed By decreasing the maximum thresh old bright features are removed Number of slices Changes how accurate the volume rendering is Most vol umes can be over one hundred slices thick in any direction Instead of blending all slices together a subset of slices are blended together The more slices the better the rendering but the slower the performance If the number of slices control parameter is set to Manual or All changing this value will have no affect on the image Number of slices Control Allows you to manually set the number of slices to render the volume with use all the slices or have the software do it automatically recommended e Projection Affects how volume slices are combined or blended e Maximum projection The brightest portions of the volume are used to render the volume This is the default rendering style for volumes It is the recommended projection for fluorescent images bright features on a dark background Minimum projection The parts of the image with the lowest intensities are dis played This is the opposite of the maximum projection This is recommended for viewing brightfield data dark features on a bright background January 2012 Page 135
42. that is the corrected FRET image DAf If you have just run the Correct FRET Cross Talk algorithm the proper images will already be assigned All three images must be assigned in order to run statistical anal ysis on them Dye Pair s Foster Distance Ry This section is where you can enter the dye pair used in the images This will allow for a calcula tion of the actual Forster s distance when calculating the FRET efficiency Select the proper dye pair from the dropdown list by clicking on the dropdown arrow or if the distance is known yet not displayed in the dropdown menu it can be entered into the textbox manually by checking the Use New RO then entering the RO into the text box underneath thr checkbox Fixed Threshold This section has two sliders which allows you to set the upper and lower percentages of the inten sities to be analyzed Set the intensities by either clicking on the pointer and dragging it to the desired location or by clicking on the desired location to incrementally move the pointer to that spot or by entering the desired intensity in the text box Current Rectangle ROI The Image ROI section displays the coordinates of the selected region of interest Select the region of interest by clicking and dragging on the desired image to create a rectangle around the desired region Top Left refers to the top left corner of the rectangle whereas Bottom Right refers to the bottom right corner of the rectangle For each the
43. the camera position on the optical tube and grabbing an image until it is clear that the graduation lines are perfectly vertical or horizontal see Table 2 for the x and y axis calibration respectively The stage micrometer will have a specifica tion of micrometers per graduation Denote this specification as Ax micrometers graduation Table 2 N 25graduations N 512pixels 10x25x12 7_ Ax 115 512 0 54um pixels Table 2 Stage micrometer image used for calibrating the pixel size The distance between the first and the last of the six lines on the upper half of this image was measured to be 11 5cm W on a video screen The stage micrometer dimension was known from the man ufacturer to be 10um graduation The width of the entire screen W was measured to be 12 7cm on the video screen The resulting calculations are illustrated above January 2012 Page 33 Media Cybernetics Inc Manual amp Tutorials AutoQuant or 5D Viewer can be used to measure the x y locations of pixels on a screen using a ROI box the width of the box will be displayed on the Status bar of the main window If the num ber of pixels between graduations along the x and y directions is n and n respectively then the number of micrometers per pixel along the x and y directions denoted Ax and Ayp respectively is given by Ax micrometers graduation Ax micrometers pixel TT n pixels graduation Ay
44. the graduations on the fine focus controller Each graduation specifies a certain number of microns A careful recording of these graduation positions will allow you to identify slippage in the motor controller and in the linkage between the motor and the microscope but it will not allow you to determine if there is slippage in the micro scope gears themselves Check with your microscope s manufacturer to determine what type of gear mechanism is used and if this gearing may be prone to slippage Regardless of the design of this gearing any micro scope may have problems since any gearing mechanism may become stripped and easily so if it experiences harsh usage such as in educational environments Some microscope models have ball bearing friction linkages between the focus control and the stage and thereby are prone to slippage under normal conditions There are a number of manufacturers that sell turnkey 3D dataset collection systems which includes both the needed hardware and software for widefield microscopy These products carry January 2012 Page 28 Media Cybernetics Inc Manual amp Tutorials out the above instructions automatically Some of the manufacturers to check are listed here Uni versal Imaging Corporation Downington PA Compix Cranberry Township PA MediaCyber netics Bethesda MD Nikon Melville NY Leica Bannockburn IL Avoiding Backlash and Hysteresis While stepping the focus always scan agains
45. thus creating a more accurate image Calculate Clicking this button will run the Ratiometrics process and create a new image Statistics Clicking this button will open a statistics window displaying a histogram and statistics of the intensities For more information on the Statistics window see page crossreference Close This button kJ will close the Ratiometrics dialog Help This button Y will open the Online Help opened to the Ratiometrics topic Ratiometrics Tutorial 1 Navigate to the Ratiometrics folder then open HighCa1 001 HighCa2 001 LowCal 001 and LowCa2 001 For each file it will ask which pattern denotes the slice number select the pat tern in which the pound sign falls at the end of the filename 2 From the Analysis menu select Ratiometrics January 2012 Page 122 Media Cybernetics Inc Manual amp Tutorials 3 Click on the Use Grynkiewicz Equation for Ion Concentration box 4 Click on the Calibrate button 5 For the Numerator Wavelength section assign the LowCa_Num 1 and HighCa_Num 1 images to the Low Jon and High Jon images respectively For the Denominator Wavelength sec tion assign the LowCa_Den 1 and HighCa_Den 1 images to the Low Jon and High Ion images respectively Click the OK button This will populate the Calibration Parameters frame in the Ratiometrics dialog 5 Go to File then Open and navigate to the Ratiometrics folder then the Calibration folder and
46. turn the cur sor into a Select Object cursor Left click on the ROI and notice that the histogram and statistics update automatically Three Dimensional Statistics Note this section applies only to users of 5D Viewer Three dimensional statistics may be calculated when an image is in a Slice Viewer projection To calculate three dimensional statistics check on the AlI Slices box 5D Viewer will then perform the calculations on the selected volume Selecting Single Slice will cause 5D Viewer to analyze the displayed slice two dimensional only Please close all views before going on to the next section Line Profile This function allows you to obtain line profile information from the dataset by drawing lines on any view 1 Open a view you would like to draw lines on you can use the TobaMC tif from the multi channel folder for an example 2 From the Analysis menu select Line Profile A Line Profile dialog box will appear 3 Click on the Line Mode and select Square January 2012 Page 113 Media Cybernetics Inc Manual amp Tutorials 4 Left click the mouse within the view and move the cursor left clicking again will com plete the current line segment Observe the Line Profile within the Line Profile dialog will show the data properties for the line being drawn The distance of the line segment is displayed beneath the Line Profile Note For multi channel datasets the Line Profile will display colored li
47. unknown random quantity In an intuitive sense the image reconstruction algorithm is searching for the 3D fluores cent probe concentration which is the most likely one to have caused the raw optically sec tioned image dataset taking advantage of any available constraints In a mathematical sense the algorithm is based on the formulation of a mathematical function called the likelihood function This likelihood function depends on both the unknown 3D fluorescent probe concentration and the 3D PSF The algorithm iteratively solves for the 3D image that maximizes this likelihood function In other words the likelihood function is solved for the correct probe concentration and PSF that provide this maximum value solution This solution then in effect becomes our recon structed 3D probe concentration and PSF This optimization model is based on an assumption that the noisy camera data follow a Poisson random signal model This assumption is physically justi fied assuming a quantum photon limited photo detector which is certainly the case in many situ ations especially for very noisy low light level situations Thus AutoQuant has a sound mathematical basis which makes it statistically optimal Partly due to this the algorithm inher ently reduces noise in the dataset Additionally it incorporates numerous improvements and valu able features January 2012 Page 48 Media Cybernetics Inc Manual amp Tutorials User Interface
48. used for publication or presentation purposes Statistics This function provides you with specific statistical information about a dataset The Statistical Information displayed will be the following Min Intensity Max Intensity Total Number of Vox els Sum of Voxel Intensities Average Intensity Standard Deviation Variance ROI Region Of January 2012 Page 112 Media Cybernetics Inc Manual amp Tutorials Interest Area ROI Perimeter and a Histogram of ROI of Current View You can obtain statistical information about the dataset in any view except for Montage view and Movie view 1 Open the TobaMC tif dataset from the Multi Channel folder From the View menu select Slice Viewer 2 Under the Analysis menu select Statistics or click on the Statistics icon A Statistics dia log box will appear 3 Check the AlI Slices box which will perform statistics on all slices 4 Left click on the Plots Over Slices of Current Time Point Notice the dashed red line in the graph 5 In the Slice Viewer scroll through the slices Notice that the red dashed line moves to dis play the values for that slice and that the values in the Statistics of XY View also update automati cally to correspond to the new slice Click on the Histogram button 6 In the dataset window left click the ROI icon ROI then select Square Left click and drag out a rectangle then click to complete the rectangle Right click in the image to
49. 0 0b8 Cybernetics Inc Tel 1 301 495 3905 Version 30 ast west Hwy Sute 400 Web www medacy com Released 9 12 January 2012 Page 146 Media Cybernetics Inc Manual amp Tutorials Frequently Asked Questions Q How does AutoQuant perform an accurate deconvolution without a measured point spread function PSF A AutoQuant s adaptive blind deconvolution automatically extracts point spread functions from the raw image dataset It does this by analyzing the spatial frequencies that make up the raw image volume and creating one or more PSF s that are based on this information Because the performance of a microscope varies across the field of view and especially along the optical axis AutoQuant will often automatically derive several PSFs to process a single image volume This assures that results are accurate even though the optical system scope cover slip immersion medium mounting medium specimen is not linear Blindly restored PSFs are also more accurate than empirically measured PSFs because they are free from noise contamina tion and are based on the distortion actually present within the dataset rather than when the PSF was measured Q Can I process 2D images and time lapse movies A Yes AutoQuant contains an iterative blind deconvolution for use on 2D images It is for researchers who do not have a Z motor or who wish to improve resolving power of their standard 2D images or who want to improve resol
50. 0 nm rey IO 3D Blind raw if 402 x 300 x 60 nts 1 0 12987 x 0 12987 x 0 30429 F 650 nm 520 nm Scene Options Be 4 S Background Color Black Plane Outline True 450 nm Widefield Fluorescence relens 13NA verture 13 a Apply u Enhancement y Background Color hu 236y 0 c1 2802 6903 172 Operation Ready The view can be changed using the mouse keyboard and tool buttons or by changing parameters in AutoQuant 5D Viewer dialog January 2012 Page 128 Media Cybernetics Inc Manual amp Tutorials The top portion of AutoQuant 5D Viewer dialog is an object tree containing a list of all objects displayed in the scene or view Only a few default objects are present when AutoQuant 5D Viewer is started New objects can be added by using the object tool menus max Max uu Scale Bar Q volume min Min k Axes lsosurface 2 Sum H Grid ortho Slices O Alpha O Region of Interest Oblique Slice Objects can be turned on or off by checking or unchecking their associated box within the object tree Volumes isosurface ortho slice and oblique slices are treated as different views of the same object If one is added to the scene for the first time all other views are turned off in the object tree They can be manually turned on to display several views at the same time by checking multiple objects in the object tree The scale bar axes grid and region of interest
51. 122 Voxel Size 78 Warranties 7 Warranty Implied 7 Wavelength 21 Well capacity 23 What is Blind Deconvolution 45 When do artifacts appear 149 When should I use optical density or attenua tion correction 149 Why won t my images load into Photoshop 150 Wiener Inverse Filtering 46 Wiener Regularization 80 Window 144 Window List 145 Writeable CD ROMs 14 US GOVERNMENT RESTRICTED X XY Montage 104 XY Guardband 105 XY Guardband Size 105 Z Z Montage 104 Zero Pad 77 Z Montaging 104 Zoom 73 January 2012 Page 6
52. 2D Histogram changing the displayed slice or changing the ROI Colocalization Statistics This frame displays the statistics for the Region of Interest in the Colocalization Preview Along the top of the ROI frame are the statistics that are gathered on the ROIs The first ROI listed is always the Active ROI If no ROI has been drawn then the Active ROI will be the entire Coloc Viewer Delete The Delete button will delete the selected ROI s Multiple ROIs can be selected by either click ing on the first and last ROIs to be deleted while holding the shift button or clicking and dragging from the first to the last ROI to be deleted Save Statistics Clicking the Save Statistics button will save the statistics as a csv file A Save As browser will appear in which the user can navigate to the desired directory and name the file Clicking Save will complete the save January 2012 Page 125 Media Cybernetics Inc Manual amp Tutorials Generate Coloc Image Clicking this button will generate the image showing the colocalized areas based on the intensity ranges set up in the 2D Histogram This new image can be saved as an independent file Close Clicking this button kad will close the Colocalization dialog box Help Clicking this button will bring up the Online Help menu opened to the section on Colocal ization Colocalization Tutorial 1 From the File menu option select open and navigate to the Multi Channel fo
53. 4783 Test tif 256 256 1 Test_parameter set A file containing datasets parameters Triple label segment seg A file containing datasets labels Volume amp Surface Area text A text file of the measurements for Surface Area amp Volume of a dataset Toolbar Icons File Toolbar Icons GIBE Visualization Toolbar Icons min SEPP EIS TETTI View Toolbar Icons qa Q January 2012 Page 153 Media Cybernetics Inc Manual amp Tutorials Processing Toolbar Icons ES EEE E AA Analysis Toolbar Icons SUR HHO DeconvolutionToolbar Icons A OR Zoom Toolbar Icon Qe 10 Br l gt G a January 2012 Page 154 Media Cybernetics Inc Manual amp Tutorials Bibliography Agard D A 1984 Optical Sectioning Microscopy Cellular Architecture in Three Dimen sions Annual Review of Biophysics and Bioengineering 13 191 219 Aikens R S D A Agard and J W Sedat 1989 Solid State Imagers for Microscopy Meth ods in Cell Biology 29 291 313 Ayers G R and J C Dainty 1988 Iterative Blind Deconvolution Method and Its Applica tions Optics Letters 13 7 547 549 Bertero M P Boccacci G J Brakenhoff F Malfanti and H T M Van der Voort 1990 Three Dimensional Image Restoration and Super Resolution in Fluorescence Confocal Micros copy Journal of Microscopy 157 1 3 20 Carrington W A 1990 Image Restoration in 3D Microscopy with Limited Data Bioimaging
54. A M Hammoud and R L White 1993 Image Recovery from Data Acquired with a Charge Coupled Device Camera Journal of the Optical Society of America A 10 5 1014 1023 Snyder D L M I Miller L J Thomas and D G Politte 1987 Noise and Edge Artifacts in Maximum Likelihood Reconstructions for Emission Tomography IEEE Transactions on Medi cal Imaging 6 3 228 238 January 2012 Page 158 Media Cybernetics Inc Manual amp Tutorials Streibl N 1984 Depth Transfer by an Imaging System Optica Acta 31 1233 1241 Turner J N K L Szarowski S M A Marko A Leith and J W Swann 1991 Confocal Microscopy and Three Dimensional Reconstruction of Electrophysiologically Identified Neurons in Thick Brain Slices Journal of Electron Microscopy Technique 18 11 23 Van Trees H L 1968 Detection Estimation and Modulation Theory Wiley New York Veklerov E and J Llacer 1987 Stopping Rule for the MLE Algorithm Based on Statistical Hypothesis Testing IEEE Transactions on Biomedical Imaging 6 4 313 319 Visser T D J L Oud and G J Brakenhoff 1992 Refractive Index and Axial Distance Mea surements in 3 D Microscopy Optik 90 1 17 19 R H Webb and C K Dorey The Pixelated Image Chapter 4 in The Handbook of Biological Confocal Microscopy 2nd Edition James Pawley Editor Plenum Press New York 1995 Willis B J N Turner D N Collins B Roysam and T J Ho
55. Ch 3 450 nm Modality Widefield Fluorescence 0 bjective Lens 1 3 NA Imm Medium 1 518 Selecting this combo drop down displays the color picker window Here you can choose from a standard palette of colors or a wavelength Different colors can be assigned to each channel of a multi channel dataset For single channel images a third color tab containing look up tables is available This is available for single channel datasets only Changing the channel color will affect the coloring of volume renderings isosurfaces orthogonal slices and oblique slices Volume Rendering By default a volume rendering is displayed when AutoQuant 5D Viewer is first opened and a volume object is added to the object tree Some preprocessing is done to the data prior to displaying it the first time as a volume This includes rescaling of intensity ranges to fill the full dynamic range of the monitor resizing it in X and Y to speed up rendering and converting it to 8bit so the graphics card can display the volume Though some contrast and XY resolution are lost this allows the image to be rotated in real time Bright Change the overall brightness of the volume rendering e Gamma Changes how intensities are mapped to screen colors from a linear mapping to a curve A gamma value of one will create a one to one intensity to color display of January 2012 Page 134 Media Cybernetics Inc Manual amp Tutorials the image A gamma larger
56. Data Properties DIO da ii A dia 55 Using the Image Enhancement Tool ii cas 56 Stat s Display anem n E a A a a a E 57 Using th Dataset Window escisiones ea e K R E T R 57 Zooming a Dataset AR 57 Selecting a Profecias 58 Pee LEC CMO a Sinple View a es 58 Creating a Region of Interest ncawss dian A daaawus 58 Saving and Copying the Current View an 59 A are a e a E n a aa means 59 Tutorials for File Menu Items occoonncccconccccnnnnononnncanonnnnanos 61 Openme a Datase tisna ieena n cain alc aip chs cactaia tne E EE 6l Savine Lo a T eas 62 Guidelines for naming your file s seneesesseeseesessseessessesessseesreseesseessesersseessesress 62 SAVE AS A A A A 62 Saving the Application La ea 63 Editing the User Options soneron A a ao 64 Exiting AutoQuant 5D Viewer 0 AA A AN 64 Recent Eiles iS diana 65 AutoQuant COmMent e da 65 Export to MASE PLO siria ii iaa AR AEE AESA 65 EXPOrtto Mars ss 66 Exportto Meta Morph Loto ld alta 66 Tutorials for the View Menu IteMS ccccoonnnnnccccconnnancnnnnanan 67 January 2012 Page 2 Media Cybernetics Inc Displaying the Data Browsers ii a A A dai ev Add dao 67 Displaying the Status Dialog vita di rs 67 Displaying the Task Bathe ade 67 Disparo E elutes 69 Using the Max Min and Sum Projections iia di ita adi diia 69 Usine the Shee VIEWER 69 Synchronizing Slice Viewers si lcd 71 Scrolling through ss EES siz ES A A a i 71 Adding and removing Datasets from Slice Sync
57. Dataset Window Many manipulations can be done on datasets via the tools within the dataset window There are toolbars that appear across the top of a dataset window Zooming a Dataset Datasets can be zoomed in three ways Click either the icon to zoom in or the icon to zoom out e e To the right of the Zoom icons click the image size percentage icon and a drop down list will appear with pre determined zoom percentages from which to select To the right of the Zoom Percentage icon is the Aspect Ratio icon H Click this icon and select Maintain Aspect from the drop down menu Then left click and drag on one of the corners of the dataset window and the dataset will zoom in and out as the window size increases and decreases Note Selecting Stretch Aspect can alter the dataset by changing the height and width dispropor tionately depending on how the window is resized if the width is changed more or less than the January 2012 Page 57 Media Cybernetics Inc Manual amp Tutorials height and vice versa Selecting Maintain Size will not change the size or aspect ratio of the dataset at all only the window itself will be resized Selecting a Projection Projections can be selected from two places the View menu and the View icon in the dataset win dow gt 1 In the dataset window click the projections icon MAX and the available projections appear in the drop down menu a 2 Select the Sum projection icon La
58. Deconvolution Icon Opens the 3D Deconvolution Dialog 2 2D Deconvolution Icon Opens the 2D Deconvolution Dialog A 30 3D Inverse Filter Icon Opens the 3D Inverse Filter Dialog N p JN No Nearest Neighbor Icon Opens the No Nearest Neighbor Dialog v 5D Viewer Icon Opens AutoQuant 5D Viewer and the Properties Form Dialog that controls the AutoQuant 5D Viewer display e Image Statistics Icon Opens the Image Statistics Dialog am 4 Line Profile Icon Opens the Line Profile Dialog 1 Count Track Icon Opens the Count Track Dialog FRET gt FRET Cross Talk Correction Icon Opens the FRET Cross Talk Correction Dialog FRET 57 FRET Efficiency Calculation Icon Opens the FRET Efficiency Calculation Dialog Q Colocalization Icon Opens the Colocalization Dialog A Y B Radiometrics Icon Opens the Radiometrics Analysis Dialog January 2012 Page 54 Media Cybernetics Inc Manual amp Tutorials Using the Interface This section will go over the basic controls used to navigate and manipulate the User Interface Using the Data Manager The Data Manager is a simple way to manipulate the display of a dataset When a dataset is opened the dataset name will appear in the Data Manager as a file tree Under the dataset name will be listed the channels z slices and or time points associated with that dataset Deselecting a channel will change the image display of that dataset to not display that channe
59. G 116 Specified Cross Talk 116 Subtract Background 116 FRET Conversion Factor G FRET 116 G Gain 22 24 25 Adjustment 24 Gaussian Smoothing Ratiometrics 122 Gaussian Width 107 Generating Different Views and Projections 67 Grynkiewicz Equation for lon Concentration Ratiometrics 121 H Haze 40 Help 146 Histogram 23 25 Histogram of ROI of Current View 113 histogram stretching intensity filter 56 Hourglass Central Radius 103 How are saturated pixels handled 149 How do I know the restored features are real 149 How many iterations 148 Hysteresis 29 I Image Algebra 90 Image Aspect Retain Aspect Ratio Stretch Aspect Ratio 73 Image Characteristics 100 Image Macro 90 Image Size 78 Immersion Medium 21 Impalement Sites 44 Important Features of the AutoDeblur Sys tem 46 Importing Multi Channel Data 64 improved morphometry 42 Intensified CCD 46 Intensity Correction 105 Introduction 40 3D Deconvolution 40 Inverse Filter 96 Deconvolution Menu 96 Inverse Filtering 96 Nearest Neighbors or No Neighbors 96 Invert Data 90 Isosurfaces 137 Iterative Method 46 January 2012 Page 3 Media Cybernetics Inc Manual and Tutorials Jansson Van Cittert 46 K Kd Ratiometrics 122 L Least Squares 46 License US Government 11 Likelihood Function 48 Limitation of Liability Notice 7 Line Mode 113 Magnetic Tape 14 Magneto Optical disks 14 MANUAL 8 Manual Alignment 88 Maximum Likelihood 46 Median setting 80 Microscope Conde
60. Guess a model of the theoretically expected spreat of light based on the optics settings that have been applied to the dataset The default Axial Stretch Factor is 1 for widefield data and 3 for confocal data It is recommended that the Disable PSF Con straints box be unchecked Doing this maintains limitations placed on the Point Spread Factor PSF to prevent the deconvolution from considering highly unlikely or impossible solutions The PSF Waist default setting is 1 for both widefield and confocal data Thisis the size of the narrow est part of the PSF usually measured in Airy Disc diameters Autocorrelation This is used by the 2D Deconvolution to generate a a PSF by examining the image itself rather than the optics used to obtain the data It is not presently applicable to 3D Deconvolution Flat Sheet This option uses as the first guess a volume filled with a con stant value equal to the average of the entire image stack For very noisy data this is an appropri atestarting point for the deconvolution e PSF Input Selecting PSF Input allows the user to provide a dataset con taining the first guess PSF This is used anytime a measured PSF is supplied to the non blind deconvolution It is also the appropriate selection when using a PSF saved from a previous decon volution as a starting point for a new deconvolution Axial Stretch Factor The axial stretch factor is used to stretch a theoretical first guess PSF in the axial directio
61. Media Cybernetics Inc Manual and Tutorials MEDIA CYBERNETICS INC User Manual Version X3 From Images to Answers January 2012 Media Cybernetics Inc Manual and Tutorials MEDIA CYBERNETICS INC User Manual Version X3 From Images to Answers January 2012 Media Cybernetics Inc Limitation of Liability ii cien 7 Limitation of Liability Notice et Asi 7 Copa N AO E e a ese A A EE EA e AOA 8 Softwar License Agreement o 8 Installation Transferring a License Uninstalling and Hard ware Requirements sssssssnunnnnnnnnnnnnnnnnnnnnnnnnnnennnnnnnnnnn ennnen 13 Sales Tnformatioi ratios 13 Technical SPP 13 Software and Hardware Requirements Windows Systems ccccceeseeesseesteeeteeeeees 13 Software Instala A e a aii 14 Dongle Installation and Requesting a Permanent License cococnconocnnccnoconoccnnccononanonncinnonos 15 Updating ithe Dongle e tits ad 15 Starting the AutoQuant Software Windows Systems ooooocccoocccoccconccconoconncnnncconccconocinnos 15 AMO Pd a ca 16 Uninstalling the Software AA te ciee shinee st ey tcieuntredect wmaeene Sradelectucthageamaniceentaas 17 How to Use This Manual cc cecceeeeeeseeeeeeeeeeeeeeeeeeeeeeeeees 18 Guidelines for Collecting 3D Image Datasets 19 ONTE BA eee eee oP eee rE aT PRG UA Sea eee rere 19 A eee EAE 19 Collecting Optical S ctions con ns cick paren Gade ca auth ace Ia ans alia ace uae oat 20
62. Morph If you have Metamorph installed on the same machine as AQX you will have an option on the File menu called Export to Metamorph as well as an icon on the toolbar If you do not have MetaMorph installed on your machine you will not have these options Active files are sent to Metamorph through a COM based inter application exchange January 2012 Page 66 Media Cybernetics Inc Manual amp Tutorials Tutorials for the View Menu Items For this tutorial you may use your own data type Fluorescence Widefield Laser Scanning Confocal Brightfield Transmitted Light Spinning Disk Scanning Confocal Two Photon Fluorescence or you may follow along with the recommended dataset to use All 3D datasets can be viewed in a Single view All 3D datasets can be Zoomed Enhanced Inverted and its Aspect Ratio changed Choosing different views will display the image from different perspectives Displaying the Data Browser The Data Browser option also known as the Data Manager displays the Data Browser along the left side of the application window For more information on the Data Browser see the User Inter face chapter Displaying the Status Dialog To display the status of all operations click on View gt Display Status This will display the Status Dialog at the base of the application window Displaying the Task Batch To display the Task Batch the list of jobs in the Batch Processing queue click on View gt View Task Bat
63. Quadro lines 3D Labs Oxygen GVX1 and Wildcat family Matrox Parhelia Software Installation Note Installing AQX on Windows Vista from a guest account is not supported Install from an account with Administrator priviledges or as a restricted user To install complete the following procedure 1 Unplug all USB dongles then insert the software CD ROM into your CD ROM drive 2 If the software asks if you want to install Microsft NET framework 2 0 select Yes If it does not ask this it is because it is already installed on the computer 3 Follow the Installation Wizard s onscreen prompts to install AutoQuant X 5D Viewer 4 When the Installation Wizard states that AutoQuant X 5D Viewer has been installed click Finish 5 A series of HASP device driver dialogs will pop up informing you that it is uninstalling and reinstalling HASP drivers Click OK through them This will ensure that the dongle drivers are up to date January 2012 Page 14 Media Cybernetics Inc Manual amp Tutorials 6 Plug in the USB dongle and wait for Microsoft Windows to notify you that the new hard ware is ready to use 8 Double click the AutoQuant X3 icon on the desktop or in the Start menu Note If AutoQuant X3 does not automatically start the Installation Wizard browse to the CD ROM drive and double click setup bat You can find the installed folder by looking in the following directories C Program Files MediaCybernetics A
64. Quant 5D Viewer workspace OK This button Y will calibrate the images to fill in the necessary parameters in order to use the Ratiometrics feature Cancel Clicking this button will cancel the calibration process and close the Calibration Images dia log Help Clicking this button will open the Online Help file opened to the topic relating to the Cali bration dialog January 2012 Page 121 Media Cybernetics Inc Manual amp Tutorials Kd This is the affinity of a fluorescent dye to calcium Different dyes have a different affinity Below is a table listing the most popular dyes used in measuring intracellular concentration Table 1 Kg Dye Kg for Ca Fura 2 224 nM Quin 2 115nM Indo 1 250 nM Ca Green 190 nM Viscosity Factor This is the Viscosity factor which is the effect that the viscosity will have on the spectra of the dyes Viscosity is defined as the measure of the resistance to flow that a fluid exerts If the viscos ity factor is not known it is recommended that 0 7 0 85 be entered into this field PostProcessing This section allows you to clean up noisy images during processing This will result in a much more accurate analyses of the datasets Remove Spot Noise This option lets the user remove spot noise caused by a few bright pixels that may be generated during the ratio process Gaussian Smoothing This feature removes noise by using a Gaussian filter
65. R A PAR TICULAR PURPOSE OR THOSE ARISING BY LAW STATUTE USAGE OF TRADE COURSE OF DEALING OR OTHERWISE SPECIFICALLY THE SOFTWARE IS NOT GUARANTEED TO WORK TO SATISFACTION ON EVERY 3D MICROSCOPY IMAGE EVEN IF THE IMAGE IS OF A TYPE THAT THE SOFTWARE IS STATED TO BE INTENDED FOR THE ENTIRE RISK AS TO THE RESULTS AND PERFORMANCE OF THE SOFTWARE IS ASSUMED BY YOU NEITHER Media OUR DEALERS REPRESEN TATIVES LICENSORS OR SUPPLIERS SHALL HAVE ANY LIABILITY TO YOU OR ANY OTHER PERSON OR ENTITY FOR ANY INDIRECT INCIDENTAL SPECIAL OR ECO NOMIC LOSS EVEN IF WE HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES OR THEY ARE FORESEEABLE WE ARE ALSO NOT RESPONSIBLE FOR CLAIMS BY A THIRD PARTY OUR MAXIMUM AGGREGATE LIABILITY TO YOU AND THAT OF OUR DEALERS SUPPLIERS AND LICENSORS SHALL NOT EXCEED THE AMOUNT PAID BY YOU FOR THE PRODUCT THE LIMITATIONS IN THIS SECTION SHALL APPLY WHETHER OR NOT THE ALLEGED BREACH OR DEFAULT IS A BREACH OF A FUNDAMENTAL CONDITION OR TERM OF A FUNDAMENTAL January 2012 Page 10 Media Cybernetics Inc Manual amp Tutorials BREACH SOME STATES COUNTRIES DO NOT ALLOW THE EXCLUSION OR LIMITA TION OF LIABILITY FOR CONSEQUENTIAL OR INCIDENTAL DAMAGES SO THE ABOVE LIMITATION MAY NOT APPLY TO YOU Information furnished by Media in the SOFTWARE manual is believed to be accurate and reli able However no responsibility is assumed by Media for its use nor for any infringements of intellectual property rights or
66. Setting the Top and Bottom of the Sample and of the Scan eee eeeeeeeeeeeee 20 Setting the Top and Bottom of the Scan csi isc 21 Setting the Exposure Gain and Offset adi iit 22 Coote CIP Cae haces O 22 R S 1 70 Cameras Ss A ds 24 Performing the Axial SC a 27 Avoiding Backlash and Hysteresis oooooniccnnconoccnoccnconccnonoconnnonccnonoconnnnancconccnonocinos 29 Collecting Bias and Flatfield Frames sist ri dolls 29 Cooled CCD Cameras ii A i aiaia 29 S170 CAMAS A A T ts 30 Spatial APO LALOR A A a n a a a a a a eee 32 Vibration Conio l euina enn tes o 34 Collecting Confocal Datasets si ia 35 Sampling Issues ienna ahina a a AN 36 Contocal APertUre moo erar as 37 2 Photon Dataset Collection sssajec abe cages casedelqessaalaanak wove counea sudesateoveyebageSedeeuas sauce deueueeas 39 Media Cybernetics Inc Guide for Obtaining the Correct Collection Parameters ooooonccnocnnccnoccnanconccononnnonos 39 IMTROCUCTION asia 40 Whatis 3D Deconvolution ai a a ia E Oia 40 Application of the AutoQuant Deconvolution Package ooonconiccnncnoccnocococononcnncnanonncnnnons 43 What is Blind Deconvolution ii 45 Important Features of Auto Quant lt a asualec ashes weincinane aad d a eaeaiaeataneer 46 NAS A O cee totes veut 47 Basic Principles Underlying the AutoQuant System ooooconinccnocccocccononconnoconocanoconnncnnncnnos 48 User Interfaces nido 49 smooth Interface to Ed 55 Usine the Data MANITAS A a a Nal ha ote R 55 Using the
67. There are five available settings None Tiny Normal Severe and Very Severe 4 Verify that the Image Warp setting is None this is the default Alignment Method This is the frame to which all other frames are aligned e Adaptive Method This function finds the base slice which has the highest sim ilarity in features and structures to all other slices e Compare to Nearest Slice This function compares each slice to its nearest slice in order to align the dataset e Always Compare to Base This function compares every slice in the dataset to the base slice slice 1 5 Select Adaptive Method Void Region Fill This function will put a frame like cover over the areas that no longer contain data After the slices are aligned gaps and spaces are created at the edges This function chooses how those areas are filled in Note The Frame may have straight scalloped or irregular edges to it e Black This will put a black frame on the edge of the image where the aligned data has left a void This is recommended for Fluorescence and Darkfield datasets January 2012 Page 87 Media Cybernetics Inc Manual amp Tutorials e White This will put a white frame on the edge of the image where the aligned data has left a void This is recommended for Transmitted Light Brightfield datasets e Random This will place a frame on the edge of the dataset made up of random pixel values where the aligned data has left a void This is recom
68. When you select Save As if your dataset is in a Slice Viewer Projection and you have changed either the gamma brightness or flip settings a dialog box will appear with the question Save data as shown No will discard the gamma brightness and flip adjustments Selecting Yes will keep the gamma brightness and flip adjustments Saving the Application Layout The layout of the application can be saved to avoid having to customize toolbar locations in the future As all of the toolbar options are moveable this is a helpful tool Once all of the toolbar items are in their desired locations select Save Layout from the File menu January 2012 Page 63 Media Cybernetics Inc Manual amp Tutorials Editing the User Options Since multiple users may use the same workstation with AutoQuant 5D Viewer on it there is the ability to have multiple user options programmed for each user The following features can be customized to each user e Temporary File Directory This is where the temporary files created by the pro gram will be stored Default Output Directory This is where the program will default to when a data set is saved e Startup Directory When first opening a file the program will default to this directory Number of Recent Files Shown This determines how many files will appear in the recent files list on the File menu e NoCPU Intensive Operations This determines how many CPU intensive opera tions such as
69. X Y and Z refers to the axial coordinates of the respective corners Calculate FRET Efficiency in ROI This button calculates the efficiencies for a selected region of interest for the active image set January 2012 Page 117 Media Cybernetics Inc Manual amp Tutorials Save Statistics This button will save the data into a csv file which can then be copied and pasted into an Excel spreadsheet Delete This button removes an already processed region of interest removes it from the table and erases all of the efficiencies calculated for it Highlight the desired ROI s to be removed then click Delete Close kad This button will close the FRET Analysis dialogue Help s This button will open the Online Help file opened to the section regarding FRET Correct FRET Cross Talk Tutorial 1 Open a Windows Explorer and navigate to the FRET folder in the tutorial images Drag and drop Image AAa Image AAf Image DAa Image DAd Image DAf Image DDd and Image DDf into the AutoQuant X2 workspace 2 From the Analysis menu select FRET then select FRET Cross Talk Correction 3 Select the desired algorithm from the Algorithms section 4 Associate each excitation emission specimen with an opened file by clicking on the arrows in the FRET Images section selecting the file with the Donor excitation and Donor emis sion applied to the FRET specimen for the DDf heading the file with the Donor excitation and Acceptor emiss
70. a Cybernetics Inc Manual amp Tutorials shipof all copyrights mask work rights patents trademarks trade secrets and all other intellec tual property rights subsisting in the SOFTWARE documentation enhancements adaptations and any modifications thereto 5 TRANSFER RESTRICTIONS You may not distribute rent sell sublicense or lease or otherwise transfer the SOFTWARE without express written permission from an authorized Media official 6 ENFORCEMENT OF TERMS TERMINATION f you fail to fulfill any of your material obligations under this Agreement Media and or its licensors may pursue all available legal remedies to enforce this Agreement and Media may at any time after your default of this Agreement terminate this Agreement and all licenses and rights granted to you under this Agree ment You agree that if Media terminates this Agreement for your default you will within thirty 30 days after any such termination deliver to Media all software provided to you hereunder and copies thereof embodied in any medium 7 GOVERNING LAWS This Agreement will be governed by and interpreted in accor dance to the laws of the State of Maryland in the United States of America 8 LIMITATION OF LIABILITY NOTICE THE SOFTWARE IS PROVIDED ON AN AS IS BASIS WITHOUT ANY OTHER WARRANTIES OR CONDITIONS EXPRESSED OR IMPLIED INCLUDING BUT NOT LIMITED TO WARRANTIES OF MERCHANTABLE QUALITY SATISFACTORY QUALITY MERCHANTABILITY OR FITNESS FO
71. a predetermined per centage from the dropdown menu or manually enter a percentage into the textbox RO The Region of Interest icon allow the user to draw a region of interest in the colo calization image in the dialog The image in the Colocalization Preview displays the two selected channels in the colors selected and shows the colocalized areas in white On the left side of the Colocalization Preview are the controls to display and move the Coloc Viewer through the slices if it is being displayed in slice mode Checking the AlI Slices checkbox in the Operations section of the Colocalization dialog will dis play all slices and the slice scroll bar will disappear To bring the scroll bar back uncheck the 4 Slices checkbox The scroll bar can be used to move through the slices sequentially by clicking on the pointer and dragging it from side to side 2D Histogram The 2D Histogram displays the intensity distribution for the two channels The brighter an area is the higher the occurrence of that intensity The left side of the graph represents Image 1 and the bottom represents Image 2 Within the histogram are two lines forming an angle with boxes at the ends of them which act as sliders These lines create an intensity mask Only the intensities that fall between the two lines will be displayed The top line controls the displayed intensities for Image 2 in the Coloc Viewer The bottom line controls the displayed intensities for Imag
72. able in any event for incidental or consequential damages in connection with or arising out of the use of the furnishing performance or use of this documen tation the AutoQuant software or any additional software provided for use with the software To the maximum extent permitted by applicable law in no event shall Media Cybernetics Inc or its dealers be liable for any damages whatsoever including without limitation damages for loss of business profits business interruption academic prestige academic funding loss of business information or any pecuniary or other loss arising from the use or inability to use this Media Cybernetics Inc product even if Media Cybernetics Inc has been advised of the possibility of such damages Because some states jurisdictions do not allow the exclusion or limitation of liabil ity for consequential or incidental damages the above limitation may not apply to you In the event that Media Cybernetics is liable under implied warranty or incidental damages the period is limited to one year from date of purchase This notice will be governed by and interpreted in accordance with the laws of the State of Mary land in the United States of America January 2012 Page 7 Media Cybernetics Inc Manual amp Tutorials Copyright Notice Copyright 1995 2011 by Media Cybernetics Inc MANUAL All rights reserved This manual may not be reproduced stored in a retrieval system or transmitted
73. alization to perform on the dataset e Noise Smoothing Move the slider to adjust the amount of noise smoothing to perform on the dataset e Wavelet Preprocessing Move the slider to adjust the amount of wavelet pre processing to perform on the dataset Histogram The Histogram is used to segment the dataset for counting Move the Max red and Min orange sliders to adjust the intensities displayed The objective is to adjust the thresholds such that the objects to be counted are filled in with white and disconnected with other objects Count All T Count Current T These buttons will activate the Counting function with Count All T counting all of the timepoints while Count Active T will only count the currently displayed timepoint Statistics Frame At the bottom of the Counting and Tracking dialog is the Statistics frame This frame contains the statistics about all of the objects and tracks found in a dataset once the operations are performed The statistics can be saved as an xls file by clicking the Save icon Objects can also be deleted by highlighting the objects to be removed then clicking on the trash can icon 3 Clicking Auto Quant 5D Viewer icon will open the objects in AutoQuant 5D Viewer Export Objects Export Segmented Data These buttons will open a new dataset window the Export Objects button will display the objects whereas the Export Segmented Data button will display the segmented data These can be saved and
74. ame the file currentviewtest tiff then click Save 4 From the File menu select Open browse to the directory where the currentviewtest tiff file was saved select the file and then open it Note Only what is visible in the dataset window when the current view is saved will be shown and available for viewing in the tiff file that is created using the Save Current View feature If the dataset is zoomed in such a way that the edges are not displayed in the dataset window those edges will not be included in the tiff file 5 Make the original dataset active again by clicking on it 6 Click the Copy Current View Icon 7 Open a Microsoft Office document such as Word or PowerPoint 8 In the Edit menu of the Microsoft application select paste and the dataset s current view will appear Linking Slice Viewers Multiple Slice Viewers can be linked together to scroll through slices or time simultaneously and in synchronization 1 Open any 3D dataset or in the tutorial data use the colorpollenc tif dataset from the Multi Channel folder Once it displays open the same dataset again January 2012 Page 59 Media Cybernetics Inc Manual amp Tutorials 2 In each of the datasets select Slice Viewer from the Projection menu within the dataset window a 3 Click the Synchronize icon in both dataset windows 4 In the slice scroller toolbar along the left side of the dataset window click the Play icon e and n
75. and Two Dimensional Spectroscopy Los Angeles SPIE 1205 72 83 Cohen A R B Roysam and J N Turner 1994 Automated Tracing and Volume Measure ments of Neurons from 3 D Confocal Fluorescence Microscopy Data Journal of Microscopy 173 2 103 114 Conchello J and E Hansen 1990 Enhanced 3 D Reconstruction From Confocal Scanning Microscope Images 1 Deterministic and Maximum Likelihood Reconstructions Applied Optics 29 26 3795 3804 Cooper J A S Bhattacharyya J N Turner and T J Holmes 1993 Three Dimensional Transmitted Light Brightfield Imaging Pragmatic Data Collection and Preprocessing Consider ations MSA Annual Meeting Cincinnati San Francisco Press 51 276 277 Deitch J S K L Smith J W Swann and J N Turner 1991 Ultrastructural Investigation of Neurons Identified and Localized Using the Confocal Scanning Laser Microscope Journal of Electron Microscopy Technique 18 82 90 Dempster A P N M Laird and D B Rubin 1977 Maximum Likelihood from Incomplete Data via the EM Algorithm Journal of the Royal Statistical Society B 39 1 37 Elangovan M Wallrabe H Chen Y Day R Barroso M and Periasamy A 2003 Characteriza tion of one and two photon excitation fluorescence resonance energy transfer microscopy Meth ods 29 58 73 Erhardt A G Zinser D Komitowski and J Bille 1985 Reconstructing 3 D Light Micro scopic Images by Digital Image Processing Ap
76. anenos 128 a A II A S 128 Reguiremients enro Seinri ionn N tars oad cas Ardea teed Ais ida Mae een Aide 128 Interface OVERVIEW srta pai 128 O A as a a nda a a a aa aioe eats 129 F to al ccs Seca RN 130 ST VPILS SETI SS 524 cc a E E etomneadiets 130 Media Cybernetics Inc Object Parameters nu AA AAA O Ad A E 131 Tutorial rt tt id 132 NA IN 133 Keyboard HOt SY sista caiacs cents wai a ear tei E tet Seats Pal tat alah uaa tah ais 133 maes Display Controls da ni e a a 133 A E E E a ae EEEE aCe EE at ea E E 134 A tana E E tian EE A AVE AE TEA 134 A a e a a dumpiy E aa E E 137 A 139 A A O A 139 A E ame eles 141 SEED id A do ds 141 PAS dele E 141 Ad a 141 A A 142 O 142 Movie Makma ss 143 WVU COWS minan aon a ase 144 Cascade esnie OA 144 Tile Vertically a E ENR ai 144 Tile Horizontally iS ii do teadaavagiees 144 Closs AMES is 144 Clos Al Modules ern RA 144 Show Dock Windows u ae anen sae 144 Hide Dock Window se sas eves A aE E aR 144 Window List ee a sagas ohne a e a E E 145 HID saso a E ecu 146 AutoQuant 5D Viewer Hele OS 146 Check for Updates A A A tied essed eek 146 ADO A O ES a 146 Frequently Asked Questions oooocccconoccccooconcnnnannconenanconnnnos 147 INDDONIGICGS aiii 151 Table of Z Spacings Z Step Sizes A did 151 Table of Parametros laces a e an AAAA O 152 STUD AE TS A A S 153 Media Cybernetics Inc File Toolbar Icons a A ade ES rd 153 A a alude 153 Visualization Toolbar Icons reeortoinisiiac
77. athematical properties of nonnegativities that is it does not allow the tracer concentration to have negative values and smoothness suppresses snowy like noise due to low light levels 3D Blind Deconvolution is a method of deconvolution that adapts itself to the real PSF of the microscope system which can be significantly different from the theoretical PSF and from the previously measured PSFs due to specimen and instrument variations The AutoQuant Blind Deconvolution system is able to adapt to PSF changes within a specimen itself Thus the decon volved results are superior to those methods which utilize theoretical or previously measured PSFs For this example you may use your own data type Fluorescence Widefield Laser Scanning Confocal Brightfield Transmitted Light Spinning Disk Scanning Confocal or Two Photon Fluorescence images or you may follow along with the recommended dataset to use 1 Navigate to the Widefield folder under the Tutorial Data directory and open the dataset FitcDapi_crop tif This dataset has had the Standard Settings and the Expert Settings set up in the previous tutorial The information is contained in its header file and the header file is automati cally read when the dataset is loaded 2 From the Deconvolution menu select 3D Deconvolution The 3D Deconvolution box will appear Extensive research and development has made entering or even changing the settings not necessary for most uses As such le
78. ave the settings at their default setting January 2012 Page 92 Media Cybernetics Inc Manual amp Tutorials 3D Deconvolution ax 3D Blind raw tf Deconvolution Methods Adaptive PSF Blind C Fixed PSF Non Blind PSF Settings Load PSF Dataset Theoretical PSF Spherical Aberrations a None Deconvolution Settings Y Use Default 97 Adaptive PSF 10 Iterations Low noise Y Output Settings Show Progress Window J Base Name 3D Blind raw tif Save PSF Optics Information Settings Invalid Modality Widefield Fluorescence Spacings O x 0 x 0 304294 pum Obj Lens NA 1 3 Al 1 35 Emw 650 520 450 nm Subyolume Information Number of Subvolumes 1 432x320x 72 Available Memory MB 1241 00 Available DiskSpace MB 13142 Confocal Defaults Nyquist Change Defaults Z Q v qu 3 Click the OK icon Y The 3D Blind Deconvolution will start The results will be dis played in the main viewing window Close all views before continuing on to the next section January 2012 Page 93 Media Cybernetics Inc Manual amp Tutorials Performing 2D Deconvolution 2D Blind Deconvolution allows you to apply either the Adaptive PSF Blind Deconvolution or Fixed PSF Non Blind Deconvolution to a single two dimensional image the image can be Fluorescence widefield Laser Scanning Confocal Brightfield Transmitted Light Spinning Disk Scanning C
79. b field 76 D Damages 7 Dark Current 47 Data Correction 79 88 Cooled CCD 81 Data Manager 55 Deblurring Oversampled 21 44 Deblurring Data Inverse Filtering 96 98 Nearest Neighbors or No Neighbors 98 Deblurring Subfields Cropping 74 Deconvolution data types 92 widefield 92 Deconvolution Menu AutoDefault 96 Deconvolution menu Process Current Slice 98 Dendritic Spines 44 Depth Of Field 21 40 Depth of the Sample 21 DIC 85 DIC Image Settings 100 DIC Restoration 99 Dichroic Mirror 25 30 Differential Interference Contrast 85 Direct Eye Viewing 25 DOF 21 Dye Pair s Foster Distance RO FRET 117 Dynamic SubVolumes 104 E Electrophysiology Probes 44 January 2012 Page 2 Media Cybernetics Inc Manual and Tutorials ENFORCEMENT OF TERMS 10 Excitation Shutter 25 Expert Settings 101 Exposure 22 Exposure Time 22 Extend Depth 77 Extend Slices 77 File Exit 64 File Menu Open 61 82 File menu Save As 76 File Names Numeric Suffix 27 Files Recent Files List 65 First Guess 105 Flat Sheet 105 Flat Field 79 Flat Field Non Uniformity 30 Flat Sheet 105 Flatfield Correction 47 Flatfield Frame 19 30 Flatfield Frames 29 Flatfield Non Uniformity 30 Flatfield Slide 30 Fluorescence 25 Fluorescent Microspheres 45 Fluoview tif 63 Foster Distance 117 Frame Averaging 27 Frame Grabber Cards 24 Frame Grabbing 22 FRET 114 Acceptor Quantum Yield Qa 116 Dye Pair s Foster Distance RO 117 FRET Conversion Factor
80. be removed e Background Image This involves taking an image of a blank specimen which will create a dataset that is entirely background and selecting this dataset from the drop down menu the dataset needs to already be opened for this to happen The algorithm will then analyze the intensity in that image and subtract that intensity from the desired dataset 1 Click on File from the main menu and select Open Navigate to the Ratiometrics folder and open HighCal 001 2 Select Background Subtraction from the Processing menu The Background Subtraction dialog box will open 3 There are five different methods of background removal Region of Interest ROI Histo gram Peak Minimum Value Constant Value and Background Image For the ROI option create a region of interest see page 58 for instructions on creating a ROT in the image that is empty background by clicking the mouse dragging it to create a box and then releasing Make sure that nothing is inside this box The Histogram Peak option requires no input it looks at a histogram of the intensities then removes the highest occurring intensity this is typically the background in an image in which the object takes up less than half of the image The Minimum Value option also requires no input and it removes the minimum intensity from the image The Constant method allows an intensity to be entered into the text box and anything at or below that intensity level will
81. be removed from the image The Background Image option requires an image to be acquired with no object in it a blank image Select this file from the drop down menu If the background image file has not been opened open it using the Open button on the main toolbar Click on the radio button for the desired method of background removal 4 Click the OK icon v A new dataset window will open with the background removed Correcting Optical Density To correct fluctuations in the image intensity values across the depth of an image use the Correct Optical Density feature This only works on images with depth gt 1 The image intensity value often fluctuates erroneously because of random flicker from the camera shutter or the lamp insta bilities Flicker occurs with nearly all widefield microscope systems It is due to several causes Most Cooled CCD cameras have a randomness in their shutter s speed This shutter speed fluctu ation causes variations on the order of several milliseconds typically from one exposure to the next The effect can be seen best in a side view projection of a dataset January 2012 Page 83 Media Cybernetics Inc Manual amp Tutorials To determine if your dataset needs Optical Density Correction look for abrupt fluctuations in image intensity summations from one depth to the next 1 From the File menu select Open Navigate to the TLB folder and click on star 1 In the Files of type field select AU Fil
82. bject that is used to initiate a blind decon volution process Flat Sheet Select this option to use a constant array as the object guess Original Data Select this option to use a smoothed version of the original image as the object first guess Filtered Original Select this option to use an Inverse Filter as the object first guess January 2012 Page 105 Media Cybernetics Inc Manual amp Tutorials e Object Input Select this option to use the previous result from an already per formed deconvolution as the object first guess Super Resolution Activate Sub pixel Processing This option will perform deconvolution on a sub pixel grid to achieve super resolved results It will also however result in longer processing times XY Factor If the Activate Sub pixel Processing feature has been checked the XY Factor feature will become enabled This feature allows you to select the number by which the pixels will be divided The options are 1 which is for when the Activate Sub pixel Processing feature is not activated 2 and 3 with three being the highest resolution and longest processing time PSF Section In the PSF section there are two tabs the Adaptive PSF Blind and the Non Blind The Adaptive PSF Blind tab PSF First Guess This is the initial PSF estimate It contains the following PSF Guess methods Theoretical Estimate Selecting Theoretical Estimate causes the deconvo lution to generate as its first PSF
83. can minimize relative motion between microscope components Remember that two types of rigidity are relevant dynamic and static Ideally the platform should be as rigid as possible dynamically as well as statically and yet be as light as possible and respond minimally to effects such as temperature changes For example Newport manufactures a special honeycomb sandwich structure that has these desirable proper ties In most quiet climate controlled laboratories an aluminum platform with a minimum thick January 2012 Page 34 Media Cybernetics Inc Manual amp Tutorials ness of 0 5 1 inch may suffice Thicker platforms have higher static rigidity Try to minimize the length and width of the structure to the minimum necessary amount In addition to placing the microscope on a rigid platform it is desirable to isolate the platform from external vibrations For this it is useful to be aware of the sources of vibrational energy around the microscope Also vibrations usually from cooling fans can be transmitted to the microscope via unavoidable accessories such as electrical cables While most of the vibrational disturbances are vertical it is possible to have disturbances that are along the horizontal direction Finally one must consider shock control in addition to vibration control Shocks often occur in a laboratory environment due to heavy footsteps nearby when buildings are being repaired or when heavy equipment is bei
84. ces one along each of the major axes Each ortho slice can be slid along its respective X Y or Z axes They can be moved by entering selection mode El toolbutton and then clicking and dragging the slice Slice Number Allows you to manually set the location of the slice along its axis The ortho slice objects can be animated by selecting Set Start Frame from the movie tool menu moving the slice and then selecting Set End Frame from the movie tool menu When the movie is played the slice is animated between the two locations Oblique slices The oblique slice object displays a slice on an arbitrary axis To move the oblique slice enter selection mode ad toolbutton select the slice by clicking on it and then click somewhere in the middle of the slice and drag To rotate the oblique slice enter selection mode select the slice and the click and drag on the outer edge or crosshairs Click outside the oblique slice to hide the oblique slice rotation and drag controls If an isosurface or volume rendering is displayed at the same time the oblique slice will by default clip or cut the isosurface Clipping can be turned on or off for each channel of the volume or isosurface When clipping is turned on for one channel and off for another you can peel away a channel by moving the oblique slice Distance from Origin The location of the oblique slice along its axis Increase or decrease this value to manually set the location
85. ch This will open the Batch Processing Dialog along the right side of the application window 1 Navigate to the Widefield folder under the Tutorial Data directory and open the dataset FitcDapi_decon tif This dataset has had the Standard Settings and the Expert Settings set up in the previous tutorial The information is contained in its header file and the header file is auto matically read when the dataset is loaded Q From the Deconvolution menu select 3D Deconvolution The 3D Deconvolution box will appear Extensive research and development has made entering or even changing the settings not necessary for most uses As such leave the settings at their default setting 3 Click the Batch icon Q in the 3D Deconvolution Dialog The 3D Deconvolution dialog will close 4 Click the Batch icon in the toolbar This will open the Batch Viewer Dialog showing the task just added to the batch If any settings were incorrect there will be a red X in the Valid column of the Staging Area Clicking that icon will bring you to the dialog for the operation listed in this case the 3D Deconvolution dialog so that the setting can be fixed January 2012 Page 67 Media Cybernetics Inc Manual amp Tutorials Start at 08 05 2011 11 00 ae January 2012 Page 68 Media Cybernetics Inc Manual amp Tutorials 5 Select a different operation 3D Inverse Filter from the drop down menu in the Operation frame of th
86. cked throughout the day if the microscope is receiving heavy usage Make certain of the fol lowing The fluorescence excitation lamp is properly centered and focused if fluorescence is being used The trans illumination lamp is properly centered and focused if transmitted light brightfield is being used The condenser is properly centered and focused if transmitted light brightfield is being used For transmitted light brightfield Adjust the condenser iris diaphragm numerical aperture NA according to the following January 2012 Page 19 Media Cybernetics Inc Manual amp Tutorials For 3D digital imaging as performed here it is best to have the transmitted light be as incoherent as possible This implies opening the NA as far as possible For direct eye viewing on the other hand it is best to eliminate scattered and stray light in order to improve contrast and this implies closing the NA to a point just below the NA of the objective lens DO NOT DO THIS As a rule of thumb it is recommended to simply open your condenser NA as far as it will go If doing so makes viewing the sample overly difficult then lower the con denser NA until it is at about 1 2 20 higher of the objective lens NA If you are using an oil objective lens having an NA that is greater than 1 0 and using a condenser that allows usage of oil then you will need to do so 1 e you will need to place a drop of oil on the condenser in order
87. cord all exposure times and file names for the bias and flatfield images since you will need them later RS 170 Cameras The bias and flatfield frame collection procedure needs to be repeated every day and for every dataset collection session This process is used to compensate for potential misalignments in the optics which will drift from day to day and to compensate for dust that may settle in the optical train which will change from day to day Collect one bias frame and one flat field frame The bias frame is taken with light to the camera blocked This may be accomplished by shifting the dichroic mirror filter set by turning off the trans illumination lamp if transmitted light bright field is used or by shuttering the excitation lamp if fluorescence is used If you are using frame averaging average as many frames as possible As a rule of thumb average 255 frames Collect a flat field frame This will be used to calculate the flat field non uniformity of your cam era A flat field can be provided in a number of ways When using transmitted light brightfield the simplest and most reliable way is to remove the sample from the microscope and grab an image of the blank stage Another way which will take more care is to take the sample extremely out of focus to the point where in effect the camera image shows a flat field see Table 1 We recommend doing this only if it is not possible to remove the sample without disrupting your e
88. crometers between 9 5 and 10 5 micrometers With a high resolution gauge hav ing a resolution of 0 01 micrometers you may be able to verify the individual step sizes However with lesser expensive gauges having a resolution of 0 1 micrometers it will be sufficient to verify the larger step sizes of 1 micrometer or more By specifying step sizes of 1 micrometer and higher verify that these steps are indeed within 20 of the 1 micrometer within 0 8 and 1 2 micrometers However make sure that the deviation from this step size is randomly dispersed about the 1 micrometers so that the average step size is still within 5 of the 1 micrometers A position measurement device such as a Mitutoyo Aurora Illinois USA 708 978 5385 MU Checker and digital Minichecker having resolutions of 0 1 micrometers should be used to verify the average step size Heidenhain sells a number of position gauges as well A resolution in the measurement of 0 1 micrometers is sufficient because it is the average of the step sizes that is most critical value One of the common malfunctions that can occur in dataset collection is gear slippage The slippage of the microscope gears or slippage of any linkage between the motor con troller and the microscope can be a critical error in dataset collection A position measurement device will check for this error If a positioning gauge is not available then the next best way of checking the positions is to record the position of
89. d down to a size that is less than or equal to the waist of the scanning spot diameter of the Airy disk In the specimen plane for a 1 4 NA lens this spot size is around 0 25 micrometers and the recom mended confocal aperture diameter after projection onto the specimen plane is then 0 25 micrometers or less This number however does not represent the actual physical diameter of the aperture For example The Molecular Dynamics Sarastro M confocal scanning system for a 60X 1 4 NA lens a spot size of 0 25 micrometers in the specimen plane is magnified to approximately 50 micrometers in the detector plane Thus the recommended true aperture size is 50 micrometers for this particular setup Please check the specifications of your confocal microscope to determine the aperture settings needed to ensure fully confocal behavior This condition will differ among confocal manufactur ers It is well understood that often times this aperture size requirement may not be met Because of signal to noise photobleaching and other considerations the confocal aperture is often opened wider The main point to emphasize is that you should not do so casually and only if it is neces sary The dataset may still deblur fine but this will depend upon your sample and other experi mental conditions and is not extremely predictable Remember that the AutoQuant system is very robust against noise It is generally more robust against noise than it is against imp
90. dded plug in Contact your dealer for purchase infor mation In order to display and analyze the effects that changes in the sample environment have on the subject it is necessary to use the Ratiometrics feature on the two datasets each dataset being taken in a different sample environment The most commonly studied changes studied are changes in calcium concentration and changes in pH Images In this section the images to be analyzed are assigned to either the numerator or the denominator The feature will perform an image division and as such the image assigned to the numerator will be divided by the image assigned to the denominator PreProcessing Use Automatic Alignment Checking this box will align the channels using the Channel to Channel Alignment feature using its default settings For more information on the Channel to Channel Alignment see page 85 Maximum Minimum Thresholding This section has two sliders which allow you to set the upper and lower percentages of the inten sities to be analyzed Set the intensities by either clicking on the pointer and dragging it to the desired location or by clicking on the desired location to incrementally move the pointer to that spot or by entering the desired intensity in the text box January 2012 Page 120 Media Cybernetics Inc Manual amp Tutorials Use Grynkiewicz Equation for lon Concentration Select this option if you are measuring the effects of changes in calc
91. deconvolution Do not close the Inverse Filter result inv_Pollen raw dataset You will compare it to the results of the Nearest Neighbor in the Nearest Neighbor s section on page91 January 2012 Page 97 Media Cybernetics Inc Manual amp Tutorials Using the Nearest Neighbor No Neighbor Algorithms Use with Widefield Fluorescence and Transmitted Light Brightfield datasets only Confocal data sets do not work with the Nearest Neighbor No Neighbor algorithms The Nearest Neighbor algorithm is the fastest algorithm available It works by deconvolving one image slice at a time As a trade off in order to achieve this speed it is less accurate than either the Blind Deconvolution or the Inverse Filter It should be used in cases where speed is most impor tant Typical processing times are less than 1 second for a 256x256 single image slice and less than 1 minute for a 256x256x32 3D dataset Pentium III 450 Mhz AutoQuant contains two methods for running the Nearest Neighbor deconvolution on a dataset One method is called Processing Stack and will run the specified deconvolution on the entire image stack The other method is called Processing Current Slice and is run while viewing one slice of an XY Slice Viewer of the image stack To Process the Current Slice 1 Click on the Pollen deb dataset to make it the active view From the View Menu select Slice Viewer The slice currently displayed will be the slice processed 2 From th
92. e 1 in the Coloc Viewer To move the sliders place the cursor over the slider box and the cursor will turn into a hand Left click on the slider then move either up and down or left to right January 2012 Page 124 Media Cybernetics Inc Manual amp Tutorials Color Map The Color Map selection changes the color map of the 2D Histogram The options available are Hot and Cool Intensity Range The Intensity Range frame displays the intensity ranges of each image This frame is not editable Operations The operations section contains tools to toggle between slice view and Sum projection as well as to add a Region of Interest to the Colocalization Statistics All Slices Checking this box will display and calculate colocalization on the Sum projection of the dataset Unchecking this box will display and calculate colocalization on individual slices as they are dis played When in slice mode the slice can be changed via the scroll bar in the Colocalization Pre view Add ROI Once an ROI has been created in the Colocalization Preview see page 53 for more information on creating a region of interest clicking the Add ROI button will add that ROI to the Colocaliza tion Statistics section with the statistics displayed Until the Add ROI button is clicked while these statistics are displayed in the Active ROI row they are not saved and will be lost if any changes are made to the Colocalization Preview be it a change to the
93. e Batch Viewer dialog 6 In the top right part of the Batch Viewer is the Start at section This is where you can schedule a batch by entering a date and time in the future Set the time for 2 minutes ahead of the current time 7 Click the arrow to the left of where it says FitcDapi highlighting the entire row At the bottom of the Staging Area click the down arrow to move the process into the pending area 8 In roughly 2 minutes the process will move into the Launched area where the 3D Inverse Filter will begin processing Displaying a Dataset Sometimes a dataset gets closed out of the workspace but it is still left in the Data Manager To display this dataset again highlight the dataset in the Data Manager by clicking on its name then click View gt Display Dataset The dataset will now be displayed in the workspace Using the Max Min and Sum Projections There are various projections in which datasets can be displayed These are Maximum Projection Minimum Projection and Sum Projection Each of these projections can be selected from the View menu Selecting the desired projection will display the dataset in that projection without opening a new dataset window Using the Slice Viewer To display and view a 3D image one slice at a time use the Slice Viewer This function generates slices through the image optically and then displays the face of each slice in the Slice Viewer The thickness of each slice corre
94. e Deconvolution menu select No Nearest Neighbors The Nearest No Neighbor Slice Operation dialog box will appear The Haze Removal Factor the Z Kernel Width and the No Neighbor selections will have values preset for you The default value for the Haze Removal factor is 0 97 and for the Z Kernel Width it is 3 The No Neighbor algorithm and Phase Content Expected are both unchecked by default Note The Phase Content Expected should only be checked if the specimen is a Transmitted Light Brightfield dataset and exhibits significant phase characteristics e g areas of the specimen appear brighter than the background For this dataset leave the Phase Content Expected box unchecked 3 Click the Apply to Current Slice button The Nearest Neighbor deconvolution will be applied to the current slice and will be displayed in the current Slice Viewer To view the original view of the slice click the Press and Hold to View Original Image button Release the button to return to the processed view To use the No Neighbor algorithm instead of the Nearest Neighbor algorithm check the No Neighbor box on the Nearest No Neighbor Slice Operation box 4 Adjust the results by trying different values for the Haze Removal Factor and the Z Kernel Width also try the No Neighbor deconvolution by clicking in the flag box to activate and deacti vate it January 2012 Page 98 Media Cybernetics Inc Manual amp Tutorials 5 Set the Haze Removal Factor t
95. e Level found directly below If the Noise Level is set to Other then this field becomes editable and you must enter in the Gaussian Width Otherwise if you select Low Medium or High then the Gaussian Width automatically defaults to a set number This feature will be used to smooth the resultant image Noise Level Select the noise level that best represents your dataset If the Low Medium High options are not sufficient then select Other and then enter in your own Gaussian Width in the Gaussian Width section above January 2012 Page 107 Media Cybernetics Inc Manual amp Tutorials 2D Blind Deconvolution Interactive This option gives you more control over the 2D Blind deconvolution process 2D Interactive Blind Deconvolution a x 3D Blind raw tif Data Setup Channel Selection All Channels bs Iteration Setup Iterations 20 Noise Level Low Preview Control rm Iteration 0 of 0 Process Result Y Save Image 3D Blind raw tif Save PSF hod January 2012 Page 108 Media Cybernetics Inc Manual amp Tutorials To use this feature follow the steps below 1 Press the Process button 2 A preview window will appear The current view in the image display window will be pro cessed This means that if the current view in the image display is a max projection then the 2D max projection will be deconvolved If the view is a slice the 2D slice
96. e New Window 7 User Interface An all new high resolution Windows 7 look and feel with intuitive new icons for all available tools AutoQuant Connect AutoQuant Connect is a new feature that allows you to automatically import and export your data to partnering software applications for additional analysis or visualization Using AutoQuant Connect you can import a CONNECT Image Set acquired in partner software applications such as Image Pro Plus or other software partners The set includes the acquired image set the image set meta data complete deconvolution set tings instructions and completion commands Upon receiving the CONNECT Image Set AutoQuant will queue all subsequent sets into the batch processing viewer to launch the next task Each set is deconvolved using the defined settings with AutoQuant s industry leading algorithms Once the deconvolution task is complete the image set along with complete meta data is returned to the location of your choice It can be returned to the connected software partner application saved to a custom folder or displayed in AutoQuant for review visualization or analysis Gibson Lanni PSF Modeling This proven Point Spread Function PSF modeling algorithm is now available for Thoretical PSF determination as well as Spherical Aber ration SA detection and correction This results in higher quality PSF modeling Auto Detect unknown Sample Medium RI and distance from cover slip An automatic ca
97. e a large noise component Important Features of AutoQuant Resolution Improvement and Noise Reduction AutoQuant improves the resolution of the microscope in the x y and z dimensions Additionally datasets from the microscope system may be especially noisy as is often the case with confocal datasets and with fluorescent widefield data sets using an intensified CCD camera AutoQuant greatly reduces this noise Statistical Optimality Because AutoQuant is based on Constrained Maximum Likelihood Esti mation theory it is in this sense statistically optimal Intuitively the algorithm searches for the most likely 3D dye concentration which caused the collected image dataset Correct Modeling Assumptions and Constraints The AutoQuant design is based on carefully justified mathematical modeling of the optical system including Poisson statistical modeling of the noise component in the detector electronics The Poisson model not only works well with low photon count datasets it also properly constrains the probe concentration and PSF to be nonnega tive This is something that is not done by methods based on Gaussian image models e g Least squares Jansson Van Cittert The model in AutoQuant also takes into account the optical charac teristic of the dataset collection system including the sample and ensures that the estimated probe concentration and PSF agree with the physical system Advantage of an Iterative Method Because the method
98. e intensity so you can comfortably view the sample through the eyepiece If you are trying to maintain a very low light level to minimize photobleaching with a fluorescence sample this might not be possible Instead you may have to simply set the light level to a suitably low level which cannot be seen by eye 2 Set the offset value on the video camera and or frame grabber Ideally you want the offset to be adjusted so that a zero light level corresponds to a zero value on a histogram of gray values Block all light to the camera This may be accomplished either by turning the lamp power switch off for transmitted light brightfield or by closing the excitation shutter for fluorescence It may also be accomplished by sliding the position of your dichroic mirror filter set Next adjust the off set or black level position s on your camera and or frame grabber card If possible to keep mat ters simple adjust only the setting of the camera and leave the rest of the system alone Make this adjustment so that you see a spike in the gray level histogram right at the zero left most value of the abscissa as shown in Figure E Avoid a completely saturated histogram as shown in Figure F This is identified by having only a single spike at the abscissa value of zero and having no spike of finite width 3 Set the gain As mentioned you may have essentially three independent ways to control the gain camera frame grabber and light intensity neutra
99. e left is for the minimum and the Max Intensity text box on the right is for the maximum the Image Enhancement dialog will default to the minimum and maximum values present in the image The minimum must be less than the maximum Enter a new value in the Percentage boxes between 0 and 100 The Min text box on the left is for the minimum and the Max text box on the right is for the maximum The minimum must be less than the maximum Left click on the red triangle on the left side of the histogram and drag it to move the minimum intensity indicator red line to the right Left click on the orange triangle on the right side of the histogram and drag it to the left to move the maximum intensity indicator orange line on the histogram The maximum intensity indicator can only be placed to the right of the minimum intensity indicator and the minimum intensity indicator can only be placed to the left of the maximum intensity indicator The Gamma can also be adjusted by left clicking on the green triangle on the bottom of the histogram and dragging it to the left or to the right The intensity values are used in a histogram stretching intensity filter For 8 bit data this filter sets all the pixel values that are greater than the maximum parameter to 255 and all the pixel values less than the minimum parameter to 0 Pixel values which fall in between the brightest and dark est parameter values are linearly scaled between 0 and 255 Jan
100. e menu Move the slice to another location and select Set End Frame from the movie menu The slice will be animated between the start and end positions during the movie This can be done in combination with zoom and rotation January 2012 Page 143 Media Cyberentics Inc Manual amp Tutorials Windows Cascade This feature places all the open dataset views in overlapping and offset positions from top to bot tom of the viewing window Tile Vertically This function stacks vertically the open datasets in the window Closed or minimized windows will not be tiled displayed To end the tiling click on each dataset and select the desired size from the zoom box Tile Horizontally This function stacks horizontally the open datasets in the window Closed or minimized windows will not be tiled displayed To end the tiling click on each dataset and select the desired size from the zoom box Close All Images This function allows the closing of all open views whether active or not and all docked datasets are minimized with only part of their title bar showing views all at once Close All Modules This function will close all open modules in the application Any dialogs that are open will be closed at this time Show Dock Windows This function will display all docked windows in the application Data Browser Dialogs and Operation Status If any of these is not actually opened it will not be displayed For instance if
101. e morphometry is desired Figure 3 On the left is a rendering of the 3D image of a fluorescent dye filled micropipette imaged by a laser scanning confocal microscope On the right is the same object after deblurring The axis running from left to right corresponds to the microscope axial direction January 2012 Page 41 Media Cybernetics Inc Manual amp Tutorials a b c d Figure 4 Showing the improved morphometry made possible by deblurring for the image of a known object fluorescent dye filled micropipette a Cross sec tions through the raw undeblurred dataset the sectioning positions are indicated by the white lines b Cross sections through the segmentation of the dataset at the same slices Panels c and d show cross sections through the deblurred data set Note the symmetry of the latter result which agrees well with known reality The improved axial resolution due to deblurring is illustrated in Figures 3 and 4 for the case of a known object a dye filled micropipette Deconvolution can often reveal fine details that are not directly observable Figure 5 shows an example of this fact January 2012 Page 42 Media Cybernetics Inc Manual amp Tutorials Figure 5 Example of improved detail revealed by deconvolution The upper panel shows the top and side projectional views of a portion of a neuron The lower panel shows the same views after deconvolution Notice how the spines of th
102. e neuron become observable Application of the AutoQuant Deconvolution Package The AutoQuant software package can be used to perform 3D deconvolution in the context of the following forms of microscopy 1 Transmitted Light Brightfield Microscopy TLB 2 Widefield Epi Fluorescence Microscopy WF 3 Confocal Pin Hole Laser Scanned Epi Fluorescence Microscopy CLSM 4 Spinning Disk Scanning Confocal and 5 Two Photon Fluorescence For the confocal fluorescence microscope images of adjacent planes are sequentially digitized and optically sectioned in a manner that is generally similar to the widefield case except that each frame is collected by way of a raster scanned laser spot and a photomultiplier pinhole detector Each confocal optical section is by itself already relatively deblurred in the sense that the confo January 2012 Page 43 Media Cybernetics Inc Manual amp Tutorials cal optics rejects most of the out of focus light However in spite of this improvement over the widefield microscope the confocal microscope has its own limitations 1 While it indeed rejects most of the out of focus light it by no means rejects all of it and thereby retains an obvious haze even though this haze is much reduced from that of the widefield microscope 2 Another effect of this imperfect out of focus light rejection is that the image retains sub stantial axial smearing For an image which is not diffraction limited
103. e on the edge of the dataset made up of random pixel values where the aligned data has left a void This is recommended for datasets that will be deconvolved or for images that have obvious structures at the edges of the frame White This will put a white frame on the edge of the image where the aligned data has left a void This is recommended for Transmitted Light Brightfield datasets Noise Background Set Level This option sets the noise level for the dataset Use the slider to select what percentage of the data set is noise OK Clicking the OK button Y will initiate the Channel to Channel Alignment feature and will cre ate a new aligned image that will need to be saved Cancel Clicking the Cancel button o will cancel and close the Channel to Channel alignment dialog No aligned image will be created or saved January 2012 Page 89 Media Cybernetics Inc Manual amp Tutorials Inverting Image Data In order to create a new dataset with an inverted intensity scale use the Invert Image Data feature 1 From the File menu select Open Navigate to the Small Hip folder and open SmallHip tif 2 From the Processing menu select Invert Image Data The SmallHip tif dataset will be pro cessed automatically and displayed in a new dataset window upon completion of the inversion Image Algebra Note This section applies to users of AutoQuant 5D Viewer only In order to perform a mathematical operation on two volumes
104. e size is 1024 by 1024 with 85 slices then the number of bytes shown in the directory for this file ought to be 1024 x 1024 x 85 x 2 178257920 January 2012 Page 27 Media Cybernetics Inc Manual amp Tutorials Be careful to move the plane of focus the exact same distance for each intermediate frame Check the specifications of your microscope and identify the amount of movement effected by a revolu tion of the fine focus knob i e the micrometers per revolution For example for the Olympus BH 2 the movement is 200 micrometers per full revolution of the fine focus knob These are well indicated by graduations on the knob with the knob having 100 graduations Thus the stage will move 2 micrometers for every graduation that the knob is turned It is important to have a reliable axial scanning mechanism with some way of verifying the step sizes The accuracy of the average step size is critical This should be within 5 of the specified step size The accuracy of individual step sizes is not nearly as critical As a rule of thumb these ought to be within 20 of the specified step size but their average value to emphasize the point needs to be within 5 For example suppose that you specify that the step sizes are 0 2 microme ters and that the scan will be 50 planes deep Then by using a position gauge on the stage or by monitoring the focus knob graduations you should record and verify that the stage moved within 5 of the 10 mi
105. e system which can be significantly different from the theoretical PSF see Figure 6 and from previously measured PSFs due to specimen and instrument variations The AutoQuant blind deconvolution system is able to adapt to PSF changes within a specimen itself Thus the deconvolved results are superior to those of methods that utilize theoretical or previously mea sured PSFs Figure 6 Shown on the left is a theoretical PSF for a confocal microscope On the right is the PSF as revealed by AutoQuant Direct experimental measurement of the PSF is an arcane and difficult procedure that is not straightforward Hiraoka Sedat et al 1990 for routine usage and thereby impedes the wide usage of deblurring algorithms that require it The PSF measurement method typically involves imaging a sub resolution fluorescent microsphere This approach has several limitations First noise in the imaging system also shows up in the measured PSF This can be overcome to some extent by averaging over several microspheres and or frames This is time consuming and effort intensive Second photobleaching of the microspheres limits the strength of the obtainable signal Third the actual PSF changes when the microsphere sample is removed and the biological speci men is inserted Fourth the PSF measurement must be done for each sample collection session and for every optical configuration used This is because the PSF is a function of the refractive index of the sa
106. e with a slide containing a drop of the fluorescent dye or try to defocus so that the current slide makes a good flat field target Table 1 However be especially careful that there is no out of focus remnant of the sample and make certain that for fluorescence your exposure times and gains are set so that you have enough signal level in the flat field image In selecting the camera gain or gain on the frame grabber illumination intensity or neutral den sity filter refer to the discussion given earlier under Setting the Exposure Gain and Offset When using transmitted light brightfield first try the same gain that was used for collecting the optical sections during the axial scan Adjust the gain to ensure the condition described earlier in Figure D When using fluorescence since the drop of dye will likely fluoresce at a much different intensity than the biological sample it is especially important to carefully adjust the gain to ensure the condition described in Figure D To emphasize it is important that the maximum gray value in the image is at about 75 85 of the well capacity of the camera and that there are no saturated pixels in the image as shown in Figure C Make a careful record of all gains and file names for the bias and flatfield images in your labora tory notebook Record all exposure times and numbers of frames averaged for all datasets col lected including the flatfield frames bias frames and the optical sections of t
107. ed the dimensions will automat ically be recalled for you Once a file is opened subsequent files with the same name but from a different directory will be numbered chronologically in the order they are opened For example if you have pollen deb opened from the widefield folder then you open another file name pollen deb from a different directory that file would be named pollen deb 2 and subsequent pollen deb files would be named pollen deb 3 pollen deb 4 etc Saving a Dataset This function saves an image or sequence of images as a specified file type If the FitcDapi_crop tif file is not still open from the Open tutorial open it now 1 Select Save As from the File menu and type BlueGreen as the File name Click on the Save as type drop down arrow and select STK FILE stk 2 Click the Save button This will save the file in the stk format Note Do not close BlueGreen stk Please continue on to the next section Guidelines for naming your file s Note AutoQuant and 5D Viewer follow the same naming conventions as standard Windows applications To move an AutoQuant or 5D Viewer file into and out of a UNIX computer follow the file nam ing conventions of the UNIX operating system For example using an underscore for file names greater than 8 characters and not using spaces or special characters Save As The dataset file can be saved as any of the following types Raw Binary Data File raw deb avz
108. eight 300 Depth 60 Time Point Options Crop Time Series C Crop Time Point Cropping and Reshaping 3D Blind raw tf Crop Selected ROI Crop Dimensions Reshape Dataset Swap Z T Enter the new dimensions size for the Dataset Time f Dimensions product 180 Depth o Channel a Select new dimensions order for the Dataset Channel_Time_S lice oe January 2012 Page 75 Media Cybernetics Inc Manual amp Tutorials 1 Go to File Open select the Confocal folder from the Tutorial data directory Select the Neuron deb dataset and open it 2 From the Processing menu select Cropping and Reshaping 3 If an ROI had been drawn its top left corner position and dimensions would be displayed in the Crop Selected ROI tab For this tutorial you will enter an ROI manually to crop a section like the one displayed above 4 Click on the Crop Dimensions tab In the Top Left Front Corner frame enter the following parameters X 51 Y 72 and Z 15 In the Bottom Right Back Corner frame enter the following parameters X 150 Y 272 and Z 37 Note In a time series dataset by default all time points will be cropped To crop only certain time points deselect the time points that you do not want cropped in the Data Manager section on the left side of the application 5 Click the OK icon Y The image will be cropped and displayed in the viewing window 6 From the File menu selec
109. er of sets is selected e g if the number 10 is cho sen then the intensities will be processed in 10 evenly distributed sets The recommended range is between 5 and 20 This number can either be typed into the text box or scrolled to using the up and down atrows next to the text box Note This is only available when using the Elangovan and Periasamy algorithm Result Name Prefix This text box allows the user to enter a prefix which will be used in the file name of each resultant image created by the experiment Calculate Clicking this button initiates the selected Correct FRET Cross Talk algorithm Reset Clicking this button will clear all file associations made in the FRET Images and Use Calibration Images sections There is no undo for this Close This button tJ closes the FRET dialog box Any files created using the FRET dialog will remain open only the dialog box closes Help This button w will launch the Online Help file opened to the FRET section January 2012 Page 116 Media Cybernetics Inc Manual amp Tutorials Calculate FRET Efficiency Once the Cross Talk has been corrected the analysis of FRET can begin Data for FRET Efficiency Estimation This section allows you to select the images to run statistical analysis on Do this by clicking on the dropdown arrow then selecting the correct image based on the heading above the dropdown menu In other words for the first top dropdown menu select the image
110. ers To compare multiple Slice Views simultaneously use the Slice Synchronizer feature This feature links up multiple Slice Viewers to scroll through the slices simultaneously 1 Open any 3D dataset or in the tutorial data use the colorpollenc tif dataset from the Mult Channel folder Once it displays open the same dataset again 2 In each of the datasets select Slice Viewer from the Projection menu within the dataset window Ss 3 Click the Synchronize icon in both dataset windows 4 In the slice scroller toolbar along the left side of the dataset window click the Play icon and notice that both dataset windows scroll through the slices simultaneously 5 Press the Stop icon Notice that both slice viewers stop If there are datasets that have different numbers of slices e g one dataset has 50 slices one has 30 slices and one has 90 slices all datasets will scroll through all of their slices the smaller data sets will pause at the end while the larger datasets finish scrolling Once the largest dataset has completed scrolling all datasets will start scrolling again at slice 1 Scrolling through Slices The large scrollbar on the left side of the Slice Synchronizer dialog box allows you to move through the slices in a manner similar to that of a single optical slice viewer 1 e the arrows in the scroller move one slice at a time and multiple slices can be jumped at a time by moving the slider about If the Play bu
111. erse Filtering and Maximum Likelihood Estima tion Journal of Microscopy 164 3 217 237 Janesick J R T Elliott and S Collins 1987 Scientific Charge Coupled Devices Optical Engineering 26 8 692 714 Joshi S and M I Miller 1993 Maximum a Posteriori Estimation with Good s Roughness for Three Dimensional Optical Sectioning Microscopy Journal of the Optical Society of America A 10 5 1078 1085 Kasten F H 1993 Introduction to Fluorescent Probes Properties History and Applications Fluorescent Probes for Biological Function of Living Cells A Practical Guide Academic Press London in press Kimura S and C Munakata 1990 Dependence of 3 D Optical Transfer Functions on the Pin hole Radius in a Fluorescent Confocal Optical Microscope Applied Optics 29 20 3007 3011 Krishnamurthi V Y Liu T J Holmes B Roysam and J N Turner 1992 Blind Deconvolu tion of 2D and 3D Fluorescent Micrographs Biomedical Image Processing III and Three Dimen sional Microscopy San Jose SPIE 1660 95 102 Krishnamurthi V J N Turner Y Liu and T J Holmes 1994 Blind Deconvolution for Fluo rescence Microscopy by Maximum Likelihood Estimation Applied Optics in review Lalush D S and M W Tsui 1992 Simulation Evaluation of Gibbs Prior Distributions for Use in Maximum A Posteriori SPECT Reconstructions IEEE Transactions on Medical Imaging 11 2 267 275 Lange K 1990
112. es if the folder appears empty 2 In order to appreciate fluctuations in image intensity an XZ projection needs to be gener ated From the View Menu click on Projection then select Sum Projection Then from the View menu click on Single View and select XZ View Notice the horizontal lines in the dataset 3 From the Processing menu select Optical Density Correction The Optical Density Correc tion box will appear Optical Density Correction starraw Selected C Channel 1 Max 1473 Min 386 Order of Polynomial 3 ate eT 1 552793E 08 Solid line ___ Original Data Dash line Polynomial Fit January 2012 Page 84 Media Cybernetics Inc Manual amp Tutorials The Order of the Polynomial is set at 3 This is the default value that can be used to correct for optical density The order of Polynomial can be changed by clicking the arrows in the box either up or down raising or lowering the Order of Polynomial respectively Note Increasing the order of the Polynomial will increase the correction in the image intensity to a certain point after which possible aberrations will be observed depending on the dataset 5 Change the Order of Polynomial to 4 Notice how the dotted line more closely follows the intensity profile of the image Click the OK icon Y 6 Place the star 1 tif OpticalCorrection XZ Sum Projection view below the star 1 XZ Sum Projection view and compare the di
113. essing Use Automatic Alignment Checking this box will align each set of images using the Channel to Channel Alignment feature using its default settings For more information on the Channel to Channel Alignment see page crossreference Background Subtraction This section allows the user to choose from three options regarding the Subtract Background algorithm None Minimum_ Value and Histogram_Peak Selecting None will perform no back ground subtraction on the datasets Selecting Minimum_ Value will subtract the background by eliminating the minimum value present in the dataset Selecting Histogram_ Peak will subtract the background based on the most commonly occurring intensity level There are other options which can be performed on images individually available from the Processing menu under the Data Correction option For more information on these see page crossreference The Subtract Back ground feature should be used to clean up images that have background noise Doing so will allow for more accurate results with the FRET algorithms FRET Conversion Factor G This is the relationship between the loss of donor signal due to FRET with the Donor filter set DD and the increase in acceptable signal due to FRET with the FRET filter DA set The default is 1 Note This is only available when using the Elangovan and Periasamy or Gordon and Herman algorithm Intensity Range Factor This divides the intensities into whatever numb
114. fferences between the views Notice how the Optical Density Correction has reduced the amount of horizontal lines in the cor rected dataset Note The intensity profile displayed is for a Grayscale dataset For a multi channel dataset the intensity profile displayed is Channel 1 which is generally the red channel This is the default The intensity profile for each channel can be viewed separately by making the appropriate selec tion under Selected Channel For each channel an order of the polynomial can be chosen which best fits the image intensity profile each channel can use a different Order of Polynomial Other types of cameras often have a similar problem Ordinary Video Rate CCD cameras intensi fied CCD cameras digital cameras and other types of cameras will show this flicker effect There are several factors that contribute to this effect and it is unknown to what degree each factor con tributes Some camera manufacturers may claim that their cameras do not have these imperfec tions Even so the effect is still likely to arise and is due at least in part to other factors in the optical train that the camera manufacturer cannot control Such factors include arc lamp flicker instabilities in the power supply to the lamp instabilities in the power supply to the camera fluc tuation in the 110 Volt A C power that supplies the camera electronics microscope and other electronics in the microscope system Other remote possibilities are
115. fortably through the eyepiece Next try an exposure time set ting Usually if you have done this before you can start with a setting that has worked well for you in the past If you are doing this for the first time or if an approximate setting is not known start with an exposure time around 0 05 seconds Grab and view a single frame at this setting If the picture on your screen appears too dark increase the exposure time and repeat the frame grab bing and viewing procedures Be careful not to saturate make too bright any portions of the image Saturated regions can usually be identified as bright flat white regions in the picture as seen in Figure B Figure 1 1 Figure A Figure B Figure A A successful unsaturated image Figure B A saturated image of the same sample Note the obvious bright flat white back ground Note that the highlighted background features marked 1 in Figure A are erroneously January 2012 Page 22 Media Cybernetics Inc Manual amp Tutorials eliminated in the saturated image Note that other features of the sample marked 2 are eroded in the saturated image A reliable method for detecting saturation is to compute and display the image intensity histo gram A straight vertical spike line at the end of the abscissa see illustration in Figure C is an indication of saturation which is to be avoided n D o S D 5 9 9 Oo E o E 3 zZ
116. fs NJ Poenie M 1990 Alteration of intracellular Fura 2 fluorescence by viscosity a simple correc tion Cell Calcium 11 2 3 85 91 Politte D G and D L Snyder 1991 Corrections for Accidental Coincidences and Attenuation in Maximum Likelihood Image Reconstruction for Positron Emission Tomography IEEE Trans actions on Medical Imaging 10 1 82 89 Richardson W H 1972 Baysian Based Iterative Method of Image Restoration Journal of the Optical Society of America 62 1 55 59 Roysam B H Ancin A K Bhattacharjya A Chisti R Seegal and J N Turner 1994 Algo rithms for Automated Characterization of Cell Populations in Thick Specimens from 3 D Confo cal Fluorescence Data Journal of Microscopy 173 2 115 126 Roysam B A K Bhattacharjya C Srinivas and J N Turner 1992 Unsupervised Noise Removal Algorithms for 3 D Confocal Fluorescence Microscopy Micron and Microscopica Acta 23 4 447 461 Shaw P J and D J Rawlins 1991 Three Dimensional Fluorescence Microscopy Progress in Biophysics and Molecular Biology 56 187 213 Shepp L A and Y Vardi 1982 Maximum Likelihood Reconstruction for Emission Tomogra phy IEEE Transactions on Medical Imaging 1 2 113 121 Sheppard C J R and M Gu 1994 3D Imaging in Brightfield Reflection and Transmission Microscopes 3D Image Processing in Microscopy Munich Society for 3D Imaging in Micros copy Snyder D L
117. g e g gt 100 slices This option reduces the amount of RAM required by the deconvolution process It may also be useful in rare cases where the sample thickness is so large that the PSF changes dramatically along Z In such cases Z Montage allows the blind deconvolution to find different PSF solutions for different depths XY Montage The XY Montage option allows the deconvolution application to break the dataset into sub vol umes along the XY dimensions Guidelines The default for this setting is On This option reduces the amount of RAM required by the deconvolution application It should be turned Off only if the deconvolution application is producing rigid box like artifacts in your dataset Z Montage The Z Montage option allows the deconvolution application to break the dataset into sub volumes along the Z dimensions The default for this option is unchecked Dynamic Subvolumes The Dynamic Subvolumes selection allows the deconvolution application to subdivide the dataset into the largest size the processing computer s RAM can handle The advantage of larger subdivi sions of data being processed is the increase in processing speed and therefore a decrease in the amount of time it takes to deblur a dataset Subvolume overlap Pixels The Subvolume overlap setting determines the number of pixels that the montaged subvolumes will overlap The possible values are integers from 0 to N 2 where N is the width or height of the
118. g is available because it executes much faster than the Median Filter setting but it is not as accurate as the Median Filter setting 10 Click the OK icon Y to begin the data correction of the Slow Scan Cooled CCD dataset Data correction will take approximately 1 2 minutes depending upon your machine 11 Once the image has been corrected the result will open in a new dataset window Close all views before continuing on to the next section High Speed CCD The High Speed CCD selection under CCD Correction is for datasets that were collected from a standard Video camera an intensified CCD camera a standard digital camera or any other camera besides the Slow Scan Cooled CCD camera There are two tabs the Files tab and the Parameters tab January 2012 Page 80 Media Cybernetics Inc Manual amp Tutorials 1 Navigate to the Data Correction folder under the Tutorial Data directory Set the Files of type to All Files Open the Strfish tif dataset Create the ZY view of the dataset 2 From the Processing menu select CCD Correction and click on High Speed CCD The High Speed CCD Data Correction Dialog will appear 3 Next open the StrfishFfv tif file from the same directory Check the Flat Field box then select StrfishFfv tif from the drop down list 4 Open the StrfishBfv tif file from the same directory Check the Bias Field box then select StrfishBfv tif from the drop down list Note There is only one Bias Field
119. ght are larger than this value then it is resized such that the larger of the width and height are equal to this value Isosurface An isosurface is a surface that represents a constant value within the dataset The constant value or threshold is automatically selected for each channel by a background removal algorithm Each pair of voxels are examined incrementally throughout the volume If the voxel values transition from below the threshold to above the threshold one or more triangles are created between the voxels January 2012 Page 137 Media Cybernetics Inc Manual amp Tutorials The location of the surface is weighted by the voxel values If the voxel with an intensity of 62 was increased to 64 the top back corner of the surface would be moved to coincide with the location of the voxel In the previous example if the selected threshold was decreased from 64 to 60 the entire surface would move to the left If the threshold was increased to 68 or more the surface would move to the right As more voxel pairs are visited the isosurface is gradually built If the threshold is changed the geometry of the isosurface is affected This in turn will affect volumetric measurements such as surface area and 3D volume Bin factor Allows the isosurface to be generated from every gnd 3rd 4th ete voxel This limits the complexity of the isosurface For smaller bin factor values more triangles are generated and the surface appears mo
120. green check it will not be displayed Note If you have multiple datasets for which you cannot import optics data but the optics infor mation is the same for all of those sets you can load that information into AQX Then set the optics for one of those sets and copy paste the optics information into the other datasets You can do this in the data manager by right clicking on the set that has the optics information selecting Copy Optics and then selecting the remaining sets Right click on the sets where you would like the information to go and select Paste Optics The toolbar along the top of the application window provides shortcuts to several commonly used functions These will be shown below as they appear on the toolbar from left to right File Open Icon Displays the File Open dialog L E Save As Icon Displays the File Save dialog a id Save Movie Icon Opens the dialog to save the dataset as an AVI file tJ Close Dataset Icon Closes the current dataset pa Save Current View Icon Saves the current active view Ed Data Manager Icon Opens the Data Manager dialog if it is not already opened E3 Display Status Icon Displays the Status window at the base of the application window if it is not already opened Q Batch Processing Icon Opens the Batch Processing Dialog if it is not already opened gt MAX Maximum Projection Icon Displays the active dataset in the Maximum Projection view
121. h an image can be resized Dimensions Voxel Size or Resize Factor There is also the option of choosing Linear or Ideal as the method employed in the Resize function As one parameter is changed the others are automatically updated to preserve the true image size 3 In the Resize Methods section select Linear 4 In the text boxes enter and verify the following For the Dimensions New X enter a value of 250 For the Dimensions New Y enter a value of 250 For the Dimensions New Z enter a value of 100 Voxel Size Verify the Voxel Size New X is 0 108 Verify the Voxel Size New Y is 0 144 Verify the Voxel Size New Z is 0 196 Resize Factor Verify the X Resize Factor New is 2 77777 Verify the Y Resize Factor New is 2 083333 Verify the Z Resize Factor New is 2 040816 5 Click the OK icon Y When completed an XY Max projection of the newly resized dataset will be created Conversely if a dataset is too large to view easily the volume can be resized to a smaller three dimensional volume Please close all views before continuing on to the next section Performing CCD Correction The CCD Correction function performs the Flat Field and Bias Field corrections and the correc tions for the camera and lamp flicker whose principles are described in the Guidelines for Col lecting 3D Image Data section of this manual The Data Correction function is for widefield datasets fluorescence or brightfield only Do not use
122. h the XY view ZY This is a side view of the dataset or each slice It can be thought of as a vertical slice through the XY view Triple View 3D Blind raw tif 3D Max Q 100 H7 way Le sol The Triple View creates three orthogonal views of the dataset On the top left of the Triple View display is the XY view on the top right is the ZY view and on the bottom left is the XZ view of the dataset The TLBview tif XY Min Projection from the previous section should be active 1 From the View menu select Triple View The XY YZ and XZ views of the TLBview tif dataset are displayed in the Triple View box 2 The Triple View can also be used on datasets displayed in a Slice Viewer projection From the View menu select Slice Viewer The Triple View will be displayed in a Slice Viewer 3 The XY YZ and XZ Optical Slice Viewers are displayed in the Triple View box The red lines indicate where each slice is in relative position to the others in the dataset Note Do not close TLBview tif XYMin Projection Please continue on to the next section January 2012 Page 72 Media Cybernetics Inc Manual amp Tutorials Saving the View To save the view of the active dataset as a tiff file use the Save View feature This will save only the image within the dataset without the borders and toolbars Alternatively click the Save Cur rent View icon in the dataset window Copying the View To copy the image withi
123. he in plane deblurring For confocal datasets however this desired improvement may come at the cost of degrading the axial resolution depending upon how well the other rules of thumb are followed as explained below In principle so long as the confocal pinhole aperture is stopped down to a point that the microscope is considered to be fully confocal then there should not be any problem with making this in plane sampling finer In fact doing so will ordinarily improve the in plane deblurring as explained On the other hand it is very often the case arguably most of the time that the confocalist does not have his her pinhole aperture stopped down to the fully confocal position because he she usually wants to detect more light to improve sensitivity January 2012 Page 36 Media Cybernetics Inc Manual amp Tutorials Therefore as a general rule of thumb for optimal results we do not recommend making the in plane resolution any finer than as indicated in the above equation If you do so we recommend that special care be taken to be sure that your confocal microscope aperture is stopped down so that it is in the fully confocal position See the next two sections for recommended sampling when the aperture setting is other than fully confocal Confocal Aperture AutoQuant runs on the mathematical assumption that you have a fully confocal microscope This condition is true so long as the confocal pinhole photodetector aperture is close
124. he sample January 2012 Page 31 Media Cybernetics Inc Manual amp Tutorials Table 1 Dust Spots A Figure a Figure b Table 1 Figure a An incorrect attempt at a flatfield image by taking the sample well out of focus This image is not far enough out of focus An out of focus remnant of the sample is still visible Table 1 Figure b A correct attempt at a flatfield image by taking the sample well out of focus Note that no out of focus remnant of the sample is apparent This image will be used to correct the optically sectioned dataset for non uniform pixel sensitivity and for the dust spots that are highlighted Spatial Calibration Important parameters that are entered into the AutoQuant program in the Data Manager see page 49 for more information on how to enter these parameters are the x and y pixel sizes dis tance between pixel centers This may be obtained by using a microscope stage micrometer e g Edmund Scientific product number J30 593 or J30 088 Collect two images of the stage microm eter using the same lens with which you collected the optical sections You will need to perform two calibrations one for the x dimension and one for the y dimension January 2012 Page 32 Media Cybernetics Inc Manual amp Tutorials Be sure that the graduations of the stage micrometer are properly aligned with the x and y axis This is done by repetitively adjusting the angle of
125. hronization ocooconnccionnncnnoninnnnnncns 71 Selecting a Simgl e A ena a iea anai aa aie i EEE 71 Triple View nie e a a a E A E E a anc ae 72 Savia Me VACW a ais 73 Copying the VIEW Ls 73 Tutorials for Processing Menu ltemS ooooccccnnnnccnoncnnonnnn os 74 Cropping AN AIM AGE ni A li 74 Extending Slices sail 77 RESIZING AM IMAGE A A Masse auton Gy eer eset TI Performing CCD Corretto iiien ad 78 Slow Scan Co led CECD y eien eai arae iier i maa eae a h e teasers 79 Hish Speed CCD rran E 80 Correcting Atte ation te da 81 Equalizing the Background 0 ex thale aeueason sats axduivenie dinette 82 Subtracting the Background ii anand iia 82 Correcting Optical Density sofa o tended acidos 83 Alg mng an Image TN 85 A O E E T E 86 Channel to Channelen iaia irai E i anevada Ae 88 Invertino Image Dita ti ir lei ada 90 Image Algebra ona a a a dns 90 Tutorials for Deconvolution Menu Items o ccccconncnccnnncnno 92 Performing 3D Deconvolution odias 92 de 94 Performing 2D Deconvolution sota indio bd 94 Using the Inverse AA ental E E E RRR 96 Using the Nearest Neighbor No Neighbor Algorithms ooooonconnccnncnocnnoncooncononononncnncnnnons 98 To Process the Current Slice ii td 98 TO Process STACK aro 99 Performing DIC Restorations iii 99 Restoration Method a oi da 99 Explaining the DIC Image Settings a di da 100 January 2012 Page 3 Media Cybernetics Inc Expert SAA A A A E TASAS SS 101 2D Blind Deconvolution Interactive s
126. ht the first ROI by clicking on it in the Colocalization Statistics section then click the Delete button January 2012 Page 126 Media Cybernetics Inc Manual amp Tutorials 9 Click Generate Coloc Image A new image will open displaying the colocalized areas Open the Slice Viewer from the new image Scroll through the slices noticing how the colocal ized areas change 10 Close all images before moving on to the next section January 2012 Page 127 Media Cybernetics Inc Manual amp Tutorials 5D Viewer Documentation Introduction The AutoQuant 5D Viewer 5D Viewer is a rendering tool that allows you to explore your three dimensional multi channel and multi time point data You can rotate zoom color section threshold isosurface and make movies of your data AutoQuant 5D Viewer can be run as its own application or in combination with the AutoQuant X2 deconvolution software Requirements To use AutoQuant 5D Viewer you will need an nVidea GeForce 5600 or nVidea Quadro FX 2000 video card or better with 128MB of texture memory or more ATI video cards are not supported Interface Overview Guanes Dex Processing Deconvolution Visualization Analysis Window Help PBSHE ABQ we TaPPPE SD CHAAR Bww KX Qw SCLRHHO Manager 1 2D example tif SS 5D Viewer 3D Blind raw tif x cen AICA RES AS E Region of Interest Pa Scale Bar Volume ACh 1 650 nm Ach 2 520 nm ch 2 45
127. icated data drive SOOGB e OS Windows 7 64 bit e Graphics Card NVIDIA Quadro 600 Super Number Cruncher e Processor Dual Intel quad core 64 bit processors current recommended model 2X Xeon 5600 series e RAM 48GB or higher January 2012 Page 13 Media Cybernetics Inc Manual amp Tutorials e Free Disk Space 2GB on installation drive and one or more dedicated SATA 6Gb s data drivers 2TB e OS Windows 7 64 bit e Graphics Card NVIDIA Quadro 4000 Larger memory sizes are recommended for improved performance keeping in mind that the oper ating system the window interface and potential tasks of other concurrent tasks make strong demands on memory resources The sample 3D images that are supplied with the software require approximately 350MB of disk storage It is recommended that due to the inherently large size of 3D images the images be stored on large volume expandable media such as CD ROMs rewritable CDs or DVDs For rou tine use a disk drive of capacity 2GB or larger is recommended The Volume Projection Hardware selection in AutoQuant 5D Viewer requires a video card which supports OpenGL extensions It is recommended that an ATI Radeon 9600 series or higher be used for the 3D Visualization Following is an non inclusive list of consumer level video cards which support OpenGL coding ATI Radeon 8500 9000 9700 Diamond FireGL 1 nVidia GeForce 256 GeForce 2 GeForce 2 MX GeForce 3 GeForce 4 all
128. icroscope This can create for certain types of trans mitted light brightfield samples a phase character These types of samples cause a refraction of light passing through them rather than a simple absorption of light This type of sample is recog January 2012 Page 149 Media Cybernetics Inc Manual amp Tutorials nized when 1 it is a transmitted light brightfield dataset and 2 you see regions in the image which are brighter than the background surrounding the sample By checking Phase Content Expected fine detail in the specimen will be preserved Q How do I deconvolve my dataset faster A First try the Inverse Filter under the Deconvolution menu This usually provides results in a few minutes It only works with widefield and not with confocal Secondly experiment with a cropped 64x64 section before deconvolving the whole volume Also set the Deconvolution Per formance to Best Medium Fast Finally make sure that you have maximized the amount of free memory on your computer Q How is blind deconvolution different from nearest neighbors and inverse filter When should I use each algorithm A The Blind Deconvolution is an iterative constrained algorithm that in most cases not all pro vides the best resolving power including the resolving power along the z axis The Inverse Filter is a fast linear method that gives remarkable deconvolutions with widefield dataset It does not work with confocal dataset With s
129. ify that the Object First Guess is set to Original Data PSF Section Note Depending on whether Load PSF Dataset is checked or Theoretical PSF is checked in the main dialog only one of the following tabs will appear If Load PSF Dataset is checked the Fixed PSF Deconvolution tab will be shown and if Theoretical PSF is checked then the Adaptive PSF Deconvolution tab will be shown 11 12 13 14 15 16 In the Adaptive PSF Deconvolution tab Verify that the Axial Stretch Factor is 1 Verify that the PSF Waist is 1 Verify that the PSF First Guess is set to Theoretical Estimate Verify that the Disable PSF Constraints box is unchecked Click on the Fixed PSF Deconvolution tab January 2012 Page 103 Media Cybernetics Inc Manual amp Tutorials 17 Verify that the Gaussian Width is set to 1 18 Verify that the Smoothing Interval is set to 3 19 Click OK to close the Expert Settings box This sets the Expert Settings for the FitcDapi_crop tif dataset 20 Close all views before going on to the next tutorial Expert Settings Explanations Z Montage The Z Montage setting allows the deconvolution application to break the dataset into sections along the optical axis and to deconvolve these subsections separately The valid settings are On checked or Off unchecked Guidelines The default for this option is Off It should only be turned On for image stacks with a large Depth settin
130. iii ii as 108 Tutorials for Analysis Menu Options 0ccooncconcccioncanoncnnonnns 111 Co nting and Track al is 111 Track Selected DISCS as 111 PD td ol 112 A A A 112 E 112 Export Objects Export Segmented Data oooooniccnnocinococonnconnnnnncconccconoconanonncconncnnos 112 o RN 112 Please close all views before going on to the next section oooooocooccnocccocccoonoconaconncinnnoon 113 A nne gt aetna e ow a ala deseo aes ines a a a 113 Using th FRET Modul o Jad Suet Jala actos Suh elses toed Salas 114 Correcting FRET Cross Talk Re decai 114 Calc late FRET Efficiency ad is 117 Use Cross Talk Coefficients Tutorial ij ssi secratesa dd oeiseacd aiae laid sencdewuiaeenaeds 119 Calculate FRET Efficiency Tutorial tias 119 RRATIOMICITICS erinan etna N crams dis 120 A e r A taat Goat aoc dancealemadeeuatnaete Gteddu sea tealines 120 PREP POCESSING oanu iginn ea ato tas o e ED 120 Use Grynkiewicz Equation for Ion Concentration coooonnnccnocccoocnnoncconncconocannconncons 121 POSTPTOCESSIND is 122 Coloca andes nates eee toaes takes o a a aera aa a e dicey atte nnn 123 ARS oe Ae 2 aera e se dade gente yo a eclectic R E woos 123 Colocalization PES SiO 124 7 detnsen a E T as hentaueas a taggin 124 CPI A A au ola A A Aan 125 Colocalization Statistics sais 125 Del te a A 125 Save Statisties o e de 125 Generate Coloc Image rad 126 OO 126 A IR E TO tac staal tiene awe ea 126 5D Viewer Documentation oooccccconnnccnoncnconennnnconanancnnnnn
131. ill automatically resize when zooming January 2012 Page 141 Media Cybernetics Inc Manual amp Tutorials Show labels Turns on or off the X Y and Z text Show Spacing Turns on or off the numbers along each axis Camera Controls how the scene is being viewed Complexity When rotating the number of slices blended for the volume and the number of isosurface triangles are decreased for faster rendering A complexity of one represents full complexity which gives best image quality but slow performance A value close to 0 will minimize the complexity show poor image quality but give best performance e Scale Allows the zoom factor for the entire scene to be manually set Orientation X Y and Z Describe the scene s orientation vector Change this to manually orient the scene e Angle Amount of rotation along the scene s orientation vector Set in degrees Perspective Toggles between perspective and orthogonal viewing modes e Perspective Angle Changes the amount of perspective This parameter has no effect 1f perspective is turned off Scene Holder for generic scene parameters e Background color Changes the background color This is automatically set to grey when an alpha projection is selected white when a min projection is selected and black when maximum or sum projection are selected Plane Outline Turn on or off the bounding lines of the reg
132. in effect only for so long as you are in compliance with the terms and conditions of this Agreement This Agreement will termi nate if you fail to comply with any of its terms or conditions You AGREE upon termination to destroy all copies of the SOFTWARE 13 GENERAL This Software License Agreement is the entire agreement between us super seding any other agreement or discussions oral or written and may not be changed except by a signed agreement If any provision of this Software License Agreement is declared by a Court of competent jurisdiction to be invalid illegal or unenforceable such a provision shall be severed from the Software License Agreement and the other provisions shall remain in full force and effect 14 WARRANTY Media offers a 90 day warranty of the software product For this 90 day period the software may be returned by the end user to the representative for reasonable cause and must be returned with all product materials included in the shipping including the software media e g CD manual if shipped hardware key and any other product materials shipped The January 2012 Page 11 Media Cybernetics Inc Manual amp Tutorials complete set of product materials must be returned for the 90 day warranty to be valid and for a refund to be entitled The 90 day period begins on the day of shipment of the software to the end user from the Representative or from Media whoever ships the product to the end user No
133. innariocines caceria ded cdoaaesosudssdeadecnavawestasdascsouas 153 View Toolbar eons is 153 Processing LOOM ACASO aia 154 AO AA 154 SS a a 154 Analysis ool bar IGons A A AAA A R des 154 Deconvelution Toolbar Icons 1 A dace ate edie seks 154 A a A e aaa a ee einen 154 BiDHOQra phy eraron ona oo ena 155 Customer Feedback oocconcconccnoconoconccnnrnnaronacnnrnanrnnacnnnnanos 160 January 2012 Page 6 Media Cybernetics Inc Manual amp Tutorials Limitation of Liability Limitation of Liability Notice To the maximum extent permitted by applicable law Media Cybernetics Inc disclaims all warran ties either expressed or implied including but not limited to implied warranties of merchantabil ity and fitness for a particular purpose with respect to the AutoQuant software package the accompanying written materials and any accompanying hardware Specifically the software is not guaranteed to work on every 3D microscopy image even if it is of a type that the software is stated to be intended for in this manual Information furnished by Media Cybernetics Inc in this manual is believed to be accurate and reliable however Media Cybernetics Inc assumes no responsibility for its use nor for any infringements of patents or other rights of third parties which may result from its use No license is granted by implication or otherwise under any patent rights of Media Cybernetics Inc Media Cybernetics Inc shall not be li
134. ion applied to the FRET specimen for the DAf heading and the file with the Acceptor excitation and Acceptor emission applied to the FRET specimen with the AAf heading Once a set of files has been assigned and processed the file associations will be saved in the header file and only one file association will need to be made in the FRET Image section the rest will automatically fill in upon clicking anywhere in the dialog 5 Select Remove Both in the Specified Cross Talk box 6 In the Calibration for Cross Talks section associate the files with the appropriate headings as outlined in the preceding list Create a rectangular region of interest in each one of the calibra tion sets Create the region of interest around the area in which FRET is suspected of occurring by clicking and dragging on the image creating a rectangle around the desired area If no calibration images are available proceed to the Use Cross Talk Coefficients section 7 Check the Use Automatic Alignment box 8 From the Subtract Background dropdown menu select Minimum_Value January 2012 Page 118 Media Cybernetics Inc Manual amp Tutorials 9 Enter tutorialset1 for the Result Name Prefix 10 Click the Calculate button FRET will start and three new images will be generated all with the prefix tutorialsetl Use Cross Talk Coefficients Tutorial With the Gordon and Herman as well as the Media Cybernetics Inc Inc algorithms there is an opti
135. ion of Three Dimensional Imaging Properties of a Light Microscope System Partial Confocal Behavior in Epifluorescence Micros copy Biophysics Journal 57 325 333 Holmes T J Bhattacharyya S Cooper J A Hanzel D Krishnamurthi V Lin W Roysam B Szarowski D H Turner J T Light Microscopic Images Reconstructed by Maximum Like lihood Deconvolution Chapter 24 in The Handbook of Biological Confocal Microscopy 2nd Edition James Pawley Editor Plenum Press New York 1995 Holmes T J 1989 Expectation Maximization Restoration of Band Limited Truncated Point Process Intensities with Application in Microscopy Journal of the Optical Society of America A 6 7 1006 1014 Holmes T J 1992 Blind Deconvolution of Quantum Limited Incoherent Imagery Journal of the Optical Society of America A 9 7 1052 1061 Holmes T J and Y H Liu 1989 Richardson Lucy Maximum Likelihood Image Restoration for Fluorescence Microscopy Further Testing Applied Optics 28 22 4930 4938 Holmes T J and Y H Liu 1991 Acceleration of Maximum Likelihood Image Restoration for Fluorescence Microscopy and Other Noncoherent Imagery Journal of the Optical Society of America A 8 6 893 907 January 2012 Page 156 Media Cybernetics Inc Manual amp Tutorials Holmes T J Y H Liu D Khosla and D A Agard 1991 Increased Depth of Field and Ste reo Pairs of Fluorescence Micrographs Via Inv
136. ion of interest and ortho slices January 2012 Page 142 Media Cybernetics Inc Manual amp Tutorials Movie Making Quick movies that rotate around a major access can be made by selecting the movie tool Then select Quick Movies select an access and then select an amount of rotation The movie is automatically played A movie navigation bar is displayed at the bottom of the screen and a new movie scene object is added to the object tree More elaborate movies can be made by selecting a zoom and rotation select Set Start Frame from the movie menu select a new zoom and rotation and then select Set End Frame from the movie menu This allows for customer zooms and rotations Frames per second Controls the speed at which the movie is played If set too large and the graphics card can not render quick enough the movie will be played as fast as possible without skipping frames To save a movie select Save Movie from the movie toolbar Wait until the movie has completed an entire cycle A new RGB multi time point dataset is created on the AutoQuant X desktop and an Save As dialog is opened allows you to save the movie as an AVI Rock Play forward and back through the rotation Total Frames Controls the number of steps between the start and end frames of the movie Ortho slices and oblique slices can be animated in Start End frame movies Move one of these slices to its start position and select Set Start Frame from the movi
137. it for confocal datasets January 2012 Page 78 Media Cybernetics Inc Manual amp Tutorials It is recommended that you do a CCD Correction on your dataset before processing a Widefield dataset fluorescence or brightfield Slow Scan Cooled CCD The CCD Correction feature can correct for Slow Scan Cooled CCD cameras when charge inte gration time is selectable 1 Navigate to the TLB folder in the Tutorial Data directory Set the Files of type to All Files and select star 1 Next open star bs0 star bs1 and star It0 These will be the Flat Field Bias Field 1 and Bias Field 2 images Note With your own datasets If you have not collected these fields then their selection boxes should be unchecked The recommendation is to collect and load all three Field image stacks 2 From the Processing menu select CCD Correction and click on Cooled CCD The Cooled CCD Data Correction box will appear Cooled CCD Correction ax 3D Blind raw tif Files Flat Field Bias Field 1 Bias Field 2 Exposure Times Average 0 00 sec Flat Field 0 00 sec Bias Field 1 0 00 sec Bias Field 2 0 00 sec Bad Pixel Treatment None v qu January 2012 Page 79 Media Cybernetics Inc Manual amp Tutorials 3 To load the Flat Field file 1t0 check the Flat Field box and select star lt0 from the drop down list 4 To Load the Bias Field 1 file check the Bias Field 1 box and select
138. ium concentrations in your sample To get the best results out of this equation a calibration set of images should be taken This set should consist of four images two each at two separate wavelengths one with no cal cium and the other saturated with calcium Once a calibration has been performed the values can be entered into the textboxes manually in later analyses Calibration Parameters Rmax This is the ratio of the two wavelengths for the calcium saturated specimen Rmin This is the ratio of the two wavelengths for the calcium free specimen B SI2 Sh2 This is the ratio between the calcium free and calcium saturated specimens Calibrate If the above mentioned set of calibration images are open click on the Calibrate button to assign these files to the proper setting Numerator Wavelength Assign the images associated with the numerator wavelength to this section The calcium free sample needs to be assigned to the Low Ion image and the saturated calcium sample needs to be assigned to the High Ion image Denominator Wavelength Assign the images associated with the denominator wavelength to this section The calcium free sample needs to be assigned to the Low Ion image and the saturated calcium sample needs to be assigned to the High Ion image Browse These buttons _ located to the right of each image textbox will open a Windows Explorer browser from which you can navigate to and click and drag files into the Auto
139. ive intensity values will be removed These negative pixel intensity values can come from the autoscaling function in the software or from artifacts in the dataset itself Click the OK icon Y The background adjusted dataset will be displayed Subtracting the Background Sometimes image analysis is performed more accurately with the background removed from the dataset especially if the dataset has background noise For instance the FRET or Ratiometrics features provide more accurate results with the background removed To accomplish this use the Subtract Background feature Removing the background is different than equalizing the back ground in that this feature actually removes the background rather than evening it out There are five different ways to subtract the background from a dataset e ROI This involves drawing an ROI around an area that is known to be the background The algorithm will remove all data with the same or lower intensity as the intensity within the ROI e Histogram Peak Selecting this algorithm will remove the most commonly occurring intensity level using the assumption that the majority of a dataset is background and contains no relevant data e Minimum Value Selecting this algorithm will remove the lowest intensity value from the image January 2012 Page 82 Media Cybernetics Inc Manual amp Tutorials e Constant Value This involves entering an intensity value below which all intensities will
140. l A channel can be deselected by clicking the box next to the desired channel so that the check mark in that box dis appears 1 Navigate to the Multi Channel folder in the tutorial data Open the colorpollenc tif dataset 2 Notice that when the dataset is displayed in the application workspace the filename also appears in the Data Manager 3 Click the arow next to colorpollenc tif in the Data Manager The channels will drop down in a file tree structure 4 Uncheck the box Y next to Channel 1 and notice that the red channel is no longer dis played in the dataset 5 To remove a dataset from the Data Manager highlight that dataset then click on the waste basket icon 3 A Using the Data Properties Dialog The Data Properties dialog contains crucial information about the active dataset The information is divided into five categories Summary Dimensions Channels Instrument and Info The Sum mary tab displays all of the image information contained and editable in the other three tabs The Dimensions tab contains the spacing and channel properties of the dataset The Channels tab con tains the controls for changing and naming the channels The Instrument tab contains the informa tion regarding the microscope used to acquire the dataset The Info tab contains optional information such as the name of the specimen when it was acquired and by whom and notes regarding the dataset Any of the fields that are white in any of the
141. l density filter To keep matters simple try adjusting only one of the gains If fluorescence is being used with an intensified camera we recommend adjusting only the camera gain keeping the excitation light as low as possible If transmitted light brightfield is being used with an ordinary video rate CCD camera such as an RS 170 camera we recommend adjusting only the illumination light level Open the light to the camera Adjust the gain or illumination level setting until the highest gray level abscissa on the histogram which has a non zero value ordinate is at about 75 of the range of the abscissa as shown in Figure D Avoid minimize saturated pixels using the procedures described earlier in this section January 2012 Page 25 Media Cybernetics Inc Manual amp Tutorials Saturated spike at zero Unsaturated spike near zero n D o S D E 9 9 O Q E 8 g Z Gray values Minimum Maximum possible possible gray gray value value Figure E Histogram condition for setting the offset black level of the camera frame grabber Look for an unsaturated spike at or near zero There may also be a saturated spike at zero January 2012 Page 26 Media Cybernetics Inc Manual amp Tutorials Saturated spike at zero n D S D 5 9 9 O E o S Z Gray values Minimum possible Maximum possible gray gray value value Figure F A zero saturated histogra
142. lculation is run based on the known accurate parameters to derive both val ues when the information is not available Sync Multiple Line Profiles and Measurements for comparison Com pare original image sets with deconvolved results easily by mapping a line profile to identical pix els on a second resultant image Display the line distance in microns compare the angle of the line to the horizontal auto scale the intensities and view the data in a log scale if the values are drastically different Export the comparison data in csv format for viewing in Excel ROI Deconvolution Preview It can be difficult to test the settings of your deconvolution run on larger image sets quickly This new feature available for both the 2D and 3D deconvolution algorithms allows you to preview your results using all the active settings You draw a small ROI on the image set and then the optimized acceleration preview begins Using the current settings in the deconvolution panel this feature provides an instant look into the results awaiting at the end of the full stack deconvolution e Batch Viewer Error Reporting In the event that the system or applica tion stops working the Batch Viewer can now report the details of the error Save Deconvolution and Optical settings files in cfg format Save all your active parameters in the 2D and 3D deconvolution dialogs to a file for future use on the same or a different system You can also save the typical op
143. lder Open colorpollenc tif 2 From the Analysis menu at the top of the screen select Colocalization The Colocalization dialog will open as well as the Colocalization Preview window in the workspace 3 For Image 1 select colorpollenc tif then select Channel 1 to the right of that with Red selected as the color next to it and for Image 2 select colorpollenc tif then select Channel 2 to the right of that with Green selected as the color next to it 4 In the 2D Histogram in the Colocalization Dialog change the maximum and minimum intensities displayed in the Colocalization Preview by either clicking and dragging the boxes at the bottom and left side of the axes or by clicking and dragging the line ends in the histogram area Notice how the colocalized areas change in the Coloc Viewer 5 Now move the Slice scroll bar back and forth Notice how the Coloc Viewer changes as the slices change Now click the A checkbox All slices will now be displayed as well as the colocalization in all of the slices 6 In the Coloc Viewer click on the ROI icon RO and select Square Inside the image click then drag out a rectangle and click again to complete it creating an ROI Click Add The statistics for the selected ROI will be loaded into the Colocalization Statistics section Create another ROI and click Add to load it into the statistics window 7 Click Save Statistics Name the file statistics 1 txt and click Save 8 Highlig
144. lignment box will appear Click the Advanced Settings button and then the dialog will appear as shown here Slice to Slice Alignment ax BlueGreen_OpticalCorrection Alignment Axis X and Y Axes v Use Recommended Expert Settings Change Expert Settings Expert Settings Image Warp Severity Normal Alignment Method AdaptiveMethod Void Region Fill Random Void Region Fill Noise Levl 8 0 100 Q v B 9 January 2012 Page 86 Media Cybernetics Inc Manual amp Tutorials Alignment Axis This features allows the alignment to be restricted to the X axis the Y axis or both axes The default setting is Both X and Y Axes Xx This setting will only align an image horizontally Y This setting will only align an image vertically Note Selecting both X and Y will align the image on both its horizontal and vertical axes 6 Leave the default setting of Both X and Y axes enabled Advanced Settings Unchecking the Use Recommended Expert Settings checkbox and then clicking the Change Expert Settings button will display additional Slice to Slice Image Alignment settings These set tings are already set based on the image information however if you feel that these are not as accurate as you would like them they can be fine tuned Image Warp Severity This feature corrects for warp due to motion between frames The more serious the warp the lon ger the processing time
145. lmes 1993 Developments in Three Dimensional Stereo Brightfield Microscopy Microscopy Research and Technique 24 437 451 Wilson T 1987 The Size of Detector in Confocal Imaging Systems Optics Letters 12 227 229 January 2012 Page 159 Media Cybernetics Inc Manual amp Tutorials Customer Feedback Media Cybernetics Inc greatly appreciates your candid comments regarding its products We welcome Information about your applications and your successes using this software Copies reprints of any publications reporting work performed using this software Comments regarding the usefulness and usability of the software Suggestions for improving this product Reports of any software bugs or problems that you may have encountered We will make every effort to address your comments and feedback in future product updates Media Cybernetics Inc 4340 East West Highway Ste 400 Bethesda Maryland 20814 U S A Phone 1 301 495 3305 Fax 1 301 495 5964 E mail info mediacy com support mediacy com Web http www mediacy com Copyright 1994 2011 Media Cybernetics Inc Inc All Rights Reserved Windows Windows XP Windows Vista and Window 7 are registered trademarks of Microsoft Corporation Pentium is a trademark of Intel Corporation Silicon Graphics and IRIX are trademarks of Silicon Graphics Inc R4600 is a trademark of MIPS January 2012 Page 160 Media Cybernetics Inc Manual and Tu
146. m This condition is to be avoided when setting the offset black level Performing the Axial Scan Begin grabbing images at each optical section Start by focusing onto one end of the sample Grab a frame Do this with frame averaging if this feature is available on your setup This is especially important if the datasets are noisy due to low light levels It is recommended that you average 255 frames if possible Refocus the microscope to the next adjacent frame and grab the next image and so on Refer to file formats that are compatible with AutoQuant If a TIFF format is used for instance as with some standard frame grabber cards store each frame in a file in sequentially numbered files with a numeric suffix i e files with names of the form IMAGE 1 IMAGE 2 and so forth If a raw 16 bit integer format is used for instance as with some of the cooled CCD cam eras then store all of the frames in sequential order and contiguously in the same file having an extension deb Repeat this image grabbing and storing sequence until you have scanned the entire sample Scan the entire depth that was chosen according to Figure 1 After collecting the datasets check the computer s directory for the file names that were chosen as a diagnostic check to be certain that all of the datasets were saved Check for their expected file sizes in numbers of bytes For instance if a raw 16 bit binary format is used as with a cooled CCD camera and if the imag
147. mber of iterations in half With some experience 3 trials should be sufficient to determine the number of iterations necessary for your experimental setup and then all similar samples having identical conditions sample type lens magnification z spacing should use the same number of iterations These 3 trials can be done in a few minutes by cropping a small 64x64 section and experimenting with that section before deconvolving the whole volume January 2012 Page 148 Media Cybernetics Inc Manual amp Tutorials Q How do I know the restored features are real A By looking at the raw and deconvolved images side by side compare to see if details are visi ble in both A structure if it is real will appear in the raw image although it will be more difficult to see due to having low contrast noise and blur Q When do artifacts appear A Artifacts such as ringing and mottling appear if the wrong parameters have been entered or if there is excessive noise Dead or hot pixels can also cause artifacts Using the Data Correc tion utilities can eliminate these problems Q How close together should my optical slices be A The optimal setting has them spaced equal to the depth of field for both widefield and confo cal This rule of thumb is flexible if other experimental constraints warrant a compromise Q What should I do after I load my dataset A Next go to Deconvolution Deconvolution Settings Sta
148. mended for datasets that will be deconvolved or for images that have obvious structures at the edges of the frame 6 Select Random from the drop down menu Noise Level This function allows you to set the noise threshold for which the Image Alignment feature will compensate 7 Click the OK icon Y The application will launch the alignment process and will display the results upon completion This result will need to be saved 8 Create Slice Viewers of the original and the aligned images then click on the Slice Syn chronizer icon Scroll through the datasets Notice the motion between the slices in the Unaligned avz dataset as compared to the Aligned avz dataset To add the alignment to the Batch click the Batch icon Q Close all views before going on to the next section Channel to Channel The Channel to Channel alignment feature allows the user to align multi channel datasets that have shift between the channels due to a filter cube or some other image acquisition problem To open the Channel to Channel Alignment feature go to the Processing menu option select Image Alignment then Channel to Channel The Channel to Channel Alignment dialog will appear Alignment Axis This section allows the user to select the axis along which the image needs to be aligned along the X axis along the Y axis or along both axes The default setting is to align to both axes Base Slice This section is where the Base Slice the slice to
149. micrometers graduation Ay micrometers pixel pels raduaion y If you do not have a stage micrometer available then you can use the following method to approx imate the pixel spacing From the manual for your CCD camera determine the size of the each pixel e g 6 10 microns is typical Divide this number by the lens magnification and the camera zoom you are using This will approximate the pixel spacing in your image For example a 40x lens with a 1 5x camera zoom and 6 micron CCD pixels will give you a pixel spacing for the image of 0 lum pixel The equation is camera pixel size lens magnification camera zoom 6um 40x 1 5x 0 lum pixel Vibration Control To maximize the quality of the dataset acquisition it is recommended that the imaging micro scope system be isolated from vibrational disturbances to the extent possible The best approach to vibration control is to use a professional anti vibration table air table e g see the Ernest Fullam Inc catalog Latham NY or the Newport Catalog Irvine CA Often sufficient vibration control can be achieved for video microscopy systems without these air tables The following section offers some insight into the vibration isolation Most vibration related dataset collection problems arise from relative motion between microscope components This type of relative motion can often be caused by external energy sources Mounting the instrument on a maximally rigid platform
150. mple or medium and its nonhomogeneity Gibson and Lanni 1991 Finally the PSF may vary spatially especially along the optic axis All of the above complications are compounded with confocal microscopy where light levels are especially low and where the dependency of the PSF to the sample refractive index is especially strong Visser Oud et al 1992 January 2012 Page 45 Media Cybernetics Inc Manual amp Tutorials By using blind deconvolution methods AutoQuant eliminates the need for Point Spread Function PSF measurement improving both the accuracy as well as convenience of deconvolution This is a major advantage of using this package instead of those utilizing non blind deconvolution methods Another clear practical advantage of AutoQuant is that it inherently reduces noise in the image especially in severely noisy cases This is partly because its mathematical foundations have quan tum photon noise as a fundamental assumption Intuitively this may be explained in that the microscope represents a band limited system which means that any good light signal will lie only inside the band limit Much of the undesirable noise energy lies outside of the band limit and the algorithm inherently recognizes this and automatically rejects the out of band noise while prop erly constraining the deblurred image to have only nonnegative values This advantage is espe cially important for confocal microscopy where images often hav
151. n Axial stretch factor can be used to correct for geometric distortion caused by severe Spherical Aberra tion SA in cases where SA detection does not work Axial stretch factor can also be used to reduce noise in undersampled confocal data Undersampled data is no longer quantitative as there is no way to know the true size of features in the image Over deconvolution with a stretched PSF can still be useful in the case to reduce noise and reveal the presence though not the true size of features in the image This section allows you to set how much axial stretch to allow from the the oretical first guess PSF The default for Widefield datasets is 1 and for Confocal datasets the default is 3 January 2012 Page 106 Media Cybernetics Inc Manual amp Tutorials PSF Waist The PSF Waist is the size of the narrowest part of the PSF usually measured in Airy Disc diame ters The default setting for both Widefield and Confocal is 1 Disable PSF Restraints This removes the limitations placed on the Point Spread Function PSF The Fixed PSF tab This tab contains the parameters to use with Gold s method Smoothing Interval The Smoothing Interval defines the iteration interval at which to apply noise smoothing when using Gold s method Regular applications of noise smoothing can help to control a noise over enhancement side effect associated with Gold s method Gaussian Width FWHM in pixels This field is based on the Nois
152. n the active dataset for pasting into a document for instance use the Copy Current View feature The current view of the active dataset is now in the clipboard and can be pasted into a document or image Alternatively click the Copy Current View icon in the data set window January 2012 Page 73 Media Cybernetics Inc Manual amp Tutorials Tutorials for Processing Menu Items Cropping an Image Once an ROI has been created see Creating a Region of Interest on page 58 the dataset can be cropped in three different ways Crop the Stack Slice by Slice or by entering crop dimensions Cropping the image creates a new dataset based on whatever pixels fall inside that selected region The following tutorials take you through the steps of each different cropping method Cropping and Reshaping a 3D Blind raw tif Crop Selected ROI Crop Dimensions Reshape Dataset Bounding Box of Selected ROI Top Left Front Comer X te Height File Depth B Time Point Options Crop Time Series C Crop Time Point January 2012 Page 74 Media Cybernetics Inc Manual amp Tutorials Cropping and Reshaping 3D Blind raw tif qesecscceoccsesscsscenscencvocscsasucsnecesean Crop Selected ROI Crop Dimensions Reshape Dataset A Enter the new dimensions for the image Top Left Front Corner Bottom Right Back Comer xmo n e 2h New Dimensions Width 402 H
153. ndard Settings and fill in any entries that have not been previously entered including the type of deconvolution you want to perform Then go to Deconvolution Start 3D Deconvolution to begin your deconvolution Q How are saturated pixels handled A Saturated pixels are handled by a proprietary algorithm which heuristically treats them as unknown data points A special algorithm uses the blur information surrounding the saturated pix els to infer the information that should have otherwise been provided by the saturated pixel Q When should I use optical density or attenuation correction A Use optical density correction if you notice a flicker in your optical sections You will notice this flicker by creating a XZ View View menu gt Single View gt XZ and a Sum Projection Visualization menu gt Sum Projection If it is present the flicker will be obvious so when in doubt you may assume there is no flicker Flicker looks like dark horizontal lines through your dataset Use attenuation correction if you notice that the overall brightness of your picture becomes progressively dimmer as you penetrate deeper into the sample Your dataset should be in a XZ View and the need for correction will be indicated by having a bright top and dark bottom Q When do I check Phase content expected A This is mainly used for brightfield samples which have not been stained and are viewed by closing down the condenser aperture on the m
154. ner of your screen and select Programs January 2012 Page I5 Media Cybernetics Inc Manual amp Tutorials 2 In the Media Cybernetics Folder select the name of the application AutoUpdate Media s products incorporate an AutoUpdate feature which will once a month connect to the internet to search for product updates automatically If you wish to search for an update without waiting for the default time to pass you can select Search for Update from the Help menu See the Help chapter for more information on the AutoUpdate feature January 2012 Page 16 Media Cybernetics Inc Manual amp Tutorials Uninstalling the Software Go to Add Remove Programs under the Control Panel of your Operating System The Add Remove Programs box will list all currently installed programs Find and click on the name of the program you wish to remove Once the name of the program you want to remove is highlighted click on the Change Remove button The InstallShield Wizard will remove the program from your computer January 2012 Page I7 Media Cybernetics Inc Manual amp Tutorials How to Use This Manual The Tutorials of the manual go through every menu item in the Menu Bar Depending on the product you have purchased AutoQuant AutoQuant 5D Viewer FRET Ratiometrics etc some sections may not apply to you this will be clearly indicated at the begin ning of the corresponding tutorial January 2012 Page 18
155. nes red green and blue each of which represents its respective channel color For example the red line is the relative intensity values with respect to the Max and Min intensity of the image 5 Red Channel The ori entation of the intensities is relative to the image not the start and endpoint of the line When in the Slice Viewer the Intensity Profile only displays intensities values for a single slice for the line being drawn At the bottom of the Line Profile box the Angle displays the angle in degrees between the last line drawn and the current line This works only in the Piecewise Line mode Note The distances displayed will be in microns per pixel Using the FRET Module Note This feature is only available as an added plug in Contact your dealer for purchase infor mation One method of studying the interaction between proteins is by analyzing FRET or Fluorescence Resonance Energy Transfer When doing so there are some issues to overcome in order to obtain the most accurate results possible Media Cybernetics FRET feature allows for the elimination of Cross Talk so FRET can be accurately calculated Correcting FRET Cross Talk Algorithms In the Algorithm section there are three algorithms from which to choose There is Elangovan and Periasamy Gordon and Herman and Media Cybernetics Inc Inc which is our own proprie tary algorithm For more information on the differences between the algorithms go to www aqi com and
156. ng moved Keep in mind that vibration isolation equipment must be considered not only for the microscope system itself but also for laboratory vibration sources such as computer and electronic systems containing fans and relays It is important to note that vibrations are best controlled at the source if at all possible Practical and Inexpensive Solutions Inexpensive vibration isolator components are available based on a variety of principles ranging from pneumatic through those based on the use of viscoelastic polymers see Edmund Scientific Barrington NJ catalog no A35 264 Sorbothane Vibration Mounts Such vibration mounts may be simply placed under a heavy pallet made of wood ceramic Plexiglas slate aluminum or some other convenient preferably heavy material Many laboratories have mounted tennis balls on such pallets as vibration isolators We recom mend first deflating the tennis balls by drilling a small hole on one end Otherwise we have expe rienced that inflated tennis balls are too rigid and do not sufficiently dampen building vibrations Collecting Confocal Datasets When collecting confocal datasets that will be used in conjunction with deblurring the following rules of thumb apply We emphasize that these are recommended rules of thumb for optimal operation Owing to the difficult trade offs among experimental conditions of signal to noise photobleaching photo damage field size and many other considerations
157. nser 19 Condenser NA 20 Eyepiece 25 Fine Focus Knob 28 Fluorescence Excitation Lamp 19 Graduations 28 Horizontal Vibrations 35 Iris Diaphragm 20 Lamp Brightness Control 24 Micrometers Per Revolution 28 NosePiece Focusing Microscope 29 Objective NA 20 Relative Motion 34 Rigid Platform 35 Stage Focusing Microscope 29 Trans Illumination Lamp 19 Microscopes Confocal 43 Transmitted Light Brightfield 43 Widefield Epi Fluorescence 43 Microsphere 45 Modeling Assumptions 46 Movie Making 143 N Nearest Neighbors 98 Nearest Neighbors Method 46 O Neutral Density Excitation Filter 24 New Dimension 77 No Neighbor 98 Noise 46 Noise Reduction 46 Noise Threshold Automatic Alignment 88 Noisy 148 Nonnegative Image Values 46 Number of Slices 21 Numerical Aperture 21 Object Tab Dynamic SubVolumes 103 XY Montaging 103 Z Montaging 103 Oblique Slices 139 Offset 22 24 25 Optical Density Correction 83 Optical Density Correction Also known as Shutter Lamp Flicker Correction 83 Adverse laboratory condition 85 Arc Lamp Flicker 85 Instabilities in Power Supply 85 Optical Sections 40 Optical Slices 149 Optics Drift 29 30 Dust 29 Misalignments in the optics 29 30 Optical train 29 Optics Alignment 19 Ortho Slices 139 Out of band noise 46 Overview Collecting Images 19 OWNERSHIP OF SOFTWARE AND ME DIA 9 P Performing Image Enhancements 56 Phase content expected 149 Photobleaching 25 45 Photomul
158. nt 80 Settings 80 Bad Pixel Treatment Settings Median 80 Bandlimit 46 Basic Principles Underlying the AutoDeblur System 48 Batch 147 Bias 47 Bias Fields 79 Bias Frame 19 30 Bias Frames 29 Bio Rad PIC pic 63 Black Level 24 25 Blind adaptive deconvolution 94 January 2012 Page 1 Media Cybernetics Inc Manual and Tutorials Blind Deconvolution Methods No PSF 46 Blur Nonisotropic 44 Bottom of the Scan 20 Bottom Slice 20 Cascade 144 CCD Camera Shutter 29 Charge Integration Time 29 Cooled CCD Camera 29 Cooled CCD Cameras 29 Dark Current 29 Flatfield Non uniformity 29 Frame Averaging 24 RS 170 Cameras 30 Well Capacity 23 CCDCorrection Cooled CCD Data Correction 78 Channel to Channel Alignment 88 Check for Update 146 Close 62 Close All 144 Collecting Optical Sections 20 Colocalization 123 Coloring 134 Confocal image 41 Confocal Microscopes Molecular Dynamics 37 Sarastro 37 Constrained Maximum Likelihood 46 Contents 146 Cooled CCD Cameras 22 Cooled CCD Correction Setting Exposure Times tab 80 Cooled CCD Data Starfish Data 80 Cooled CCD Data Correction Browse 81 Cooled CCD Settings Dimensions 80 Exposure Times 80 Copying Restrictions 8 Copyright 1995 2001 8 COPYRIGHT NOTICE 11 Copyright Notice 8 Correcting Cooled CCD Data 81 Data Correction 81 Launch 81 Correcting Data Cooled CCD 79 Correcting for Depth Attenuation 81 Crop Cropping a Sub field 76 Crop Icon Cropping a Su
159. o follow this rule of thumb In that case scan the depth of the sample plus any additional amount allowed by your free hard disk space Setting the Top and Bottom of the Scan For best results the rule of thumb in setting the axial distance between slices is to have this dis tance equal to the Rayleigh depth of field DOF as given by the following equation A AZs DOF 0 5 am NAM Zslic 4n sin 0 5 sinr NAMN where A is the wavelength if unknown use a value of 0 5 micrometers n is the refractive index of the immersion medium 1 0 for air 1 33 for water 1 515 for oil and NA is the numerical aper ture of the objective lens The total number of slices required then will be equal to the thickness of the scan selected as described earlier divided by the chosen axial distance between slices If you have limited disk storage space you may not be able to follow the above rule of thumb In that case you may set your axial distance between slices equal to a number larger than the depth of field AutoQuant will still work in that case However as a trade off you can expect to experience some degrada tion in image quality The extent of the degradation will depend upon the axial sampling distance chosen relative to the DOF Deblurring is best carried out when the datasets are seemingly oversampled In other words best results are achieved when the pixel or optical section spacing is finer than what would be used normally withou
160. o 0 98 the Z Kernel Width to 2 and leave the No Neighbor check box unchecked These values produce satisfactory results for the Nearest Neighbor or No Neighbor algorithm 6 Click the Close button bd to exit from the Nearest No Neighbor Slice Operation box The algorithm chosen Nearest Neighbor and the values selected for the Haze Removal Factor and the Z Kernel Width are now saved for this specific dataset Note Do not close the Pollen deb dataset Please continue on to the section To Process a Stack 1 Click on the Pollen deb XY MAX Projection to make it the active view From the Decon volution menu select No Nearest Neighbor 2 The Nearest No Neighbor Parameters box appears The settings for the Haze Removal Factor 0 98 the Z Kernel Width 3 and the Nearest Neighbor algorithm were determined above in the Processing Current Slice section 3 Click Start Processing Stack The deconvolution will be performed on each slice of the entire stack A status bar will indicate the progression A new dataset window will appear with the results 4 Compare the results of the Nearest or No Neighbor deconvolution to the original dataset and note how the deconvolution removes haze and sharpens features Examine the Maximum pro jections and the Slice Viewers of the original Pollen deb dataset the Inverse Filter and the Near est Neighbor results in both the XY and XZ views Close all datasets before continuing on to the next section
161. o Expert Settings button The Expert Settings dialog will appear January 2012 Page 101 Media Cybernetics Inc Manual Tutorials Deconvolution Expert Settings ax BlueGreen_OpticalCorrection Subvolume Y XY Montage Sub volume overlap Pixels 14 Z Montage XY Guardband Pixels 14 hale Z Guardband Pixels Minimum Intensity Removal Object First Guess Fittered Original v Super Resolution Activate Sub pixel Processing XY factor 1 v Fixed PSF Deconvolution Gold s Method Parameters Gaussian Width FWHM in pixels fi Smoothing Interval Iterations January 2012 Page 102 Media Cybernetics Inc Manual amp Tutorials Subvolume section 4 5 This section of Expert Settings affects the handling of the object data The default values are as follows XY Montage is selected enabled Z Montage is deselected disabled Dynamic Subvolumes is selected enabled Leave these fields at their default values Verify that the following fields are set to their default values The Subvolume overlap Pixels is 10 The XY Guardband Pixels is 10 The Z Guardband Pixels is 6 Pre processing section 6 Verify that the Intensity Correction box is checked 7 Verify that the Minimum Image Intensity Removal box is checked 8 Verify that the Accelerated SA Detection box is checked 9 Verify that the Pre Condition Imported PSF box is checked 10 Ver
162. ome types of samples especially those that have fine low con trast staining it gives even better results than the blind deconvolution Use the Inverse Filter if speed is the most important consideration or if the sample has fine low contrast detail that is important to see The Nearest Neighbor is the fastest deconvolution Its main advantage is speed It provides images typically within seconds It is a linear method Use the Nearest Neighbor algo rithm for quick feedback or for enhancing pictures for presentations but be careful about interpre tation of the images because it is not as robust against artifacts and noise as are the Blind Deconvolution and Inverse Filter Q Why do all the features in my deconvolved dataset look darker than before A The deconvolution will increase the dynamic range of the dataset as a result of removing the blur When an image is displayed on screen it is auto scaled between the maximum and mini mum intensities Therefore the background features will appear darker on screen Use the bright ness and gamma controls to better observe the features in the dataset Q Why won t my images load into Photoshop A This is because you may have saved the images in a format that is not compatible with Photo shop Save the images as TIFF and when asked by the program if you want to export as multiple files answer Yes January 2012 Page 150 Media Cybernetics Inc Manual amp Tutorials A
163. on for Z and sample as you otherwise normally would do along x y If the experi mental objectives do not call for an improved z resolution but they do call for an improved x y resolution then follow the Sampling Criterion for X Y and sample as you otherwise normally would do along z The modality setting of AutoQuant which is found in the Data Manager see page 49 for more information on using the Data Manager both on the Summary tab and the Instrument tab This is because the optical PSF of the typical 2 photon microscope theoretically assuming the detector aperture is wide open is equivalent to that of a confocal PSF except that it is determined by the excitation wavelength rather than the emission wavelength January 2012 Page 39 Media Cybernetics Inc Manual amp Tutorials Introduction What is 3D Deconvolution Also known as 3D image restoration deconvolution is a computational technique for removing out of focus haze from stacks of optical sections The out of focus haze can be mathematically modeled as a point spread function PSF Deconvolution methods can therefore be thought of as methods for inverting the unavoidable and natural blurring effect of the point spread function PSP For example suppose we focus a microscope having a fine depth of field DOF of about 0 4 micrometers onto a plane in a microscope sample which may be anywhere from a few microme ters e g chromosomes or epithelial cell
164. on to enter the Cross Talk Coefficients rather than loading calibration images It is still highly recommended that calibration images be used however the Use Cross Talk Coefficients tab exists for when these images are not available 1 Select Gordon and Herman from the Algorithms section 2 Click on the Cross Talk Coefficients radio button on the top right side of the Calibration for Cross Talks section 3 Enter the estimated Cross Talk Coefficients for each of the 6 entries on the tab 4 Enter the FRET Conversion Factor G into the text box If this value is not known the default setting of 1 can be used or an estimation between 0 and 1 can be entered if using the Gor don and Herman algorithm a value higher than 1 can be entered G is defined as the ratio of acceptor that is excited due to FRET versus the donor that is quenched due to FRET 5 Click Calculate The application will create 3 new files with the results 6 Changes can be made to the Cross Talk Coefficients settings Once the changes have been made to create new files click the Calculate button Calculate FRET Efficiency Tutorial Once the algorithm has run on the images and the new files are opened analyses can be run on the images 1 At the bottom of the dialogs section click on the Calculate FRET Efficiency tab 2 In the Data for FRET Efficiency Estimation section if the images are not already loaded in assign the appropriate images to the C
165. onfocal and Two Photon Fluorescence without requiring previous knowledge of the microscope parameters or the image parameters The 2D Deconvolution algorithm is also capable of improving the resolution of an image for the restoration of features at a sub pixel reso lution level You may process multi frame time series image sets individual color channels or intensity images The 2D algorithm is able to suppress noise while retaining quantitative accuracy total number of photons in the image thus allowing you the ability to make valid quantitative measurements 1 Navigate to the Demo Data folder and open 2D Data tif 2 From the Deconvolution menu select 2D Deconvolution The 2D Deconvolution Dialog will appear January 2012 Page 94 Media Cybernetics Inc Manual amp Tutorials 2D Deconvolution Ax 3D Blind raw tif 1 o Deconvolution Methods Adaptive PSF Blind Fixed PSF Non Blind PSF Settings C Load PSF Dataset v Theoretical PSF Derived PSF Deconvolution Settings Use Default 9 Algorithm Settings Total Iterations fio Save Interval 10 Noise Level Low y gt Noise Value 2 Performance Y Faster Processing Reduced Resolution l Use Gold s Method Y Iteration Acceleration Continue From Previous Result yy Use Recommended Expert Go to Expert Settings Settings a ein othe Y Output Settings Y Show Progress Window J Base Name 2D Blind raw tif
166. orrected FRET Corrected Donor and Corrected Accep tor images by clicking the dropdown arrows and selecting the correct image from the list 3 Select a region of interest to process by clicking the ROI icon in the dataset window selecting the rectangular ROI then clicking and dragging on one of the images around the area to be analyzed creating a rectangle around it then clicking the mouse button again January 2012 Page 119 Media Cybernetics Inc Manual amp Tutorials 4 Select the dye pair used in the images in the Dye Pair s Foster Distance Rg section by clicking on the dropdown arrow and selecting the appropriate pair from the list Note If the dye pair that was used in the images is not listed but you know the Forster s distance you can enter that distance in the text box as well Check the Use New RO box then enter the known RO into the text box below the check box 5 Select a percentage range of intensities to analyze by adjusting the Fixed Threshold sliders to the desired percentages The minimum must be lower than the maximum 6 Click Calculate Efficiency in ROI The efficiencies will be calculated and displayed in the table in the dialog box The statistics can be saved as a csv file by clicking Save Statistics and then specifying a file name and directory to which they will be saved 7 Close all images before moving on to the next section Ratiometrics Note This feature is only available as an a
167. otice that both dataset windows scroll through the slices simultaneously 5 Press the Stop icon we Notice that both slice viewers stop January 2012 Page 60 Media Cybernetics Inc Manual amp Tutorials Tutorials for File Menu Items For this tutorial you may use your own data type Fluorescence Widefield Laser Scanning Confocal Brightfield Transmitted Light Spinning Disk Scanning Confocal Two Photon Fluorescence or 2 Dimensional images or you may follow along with the recommended dataset to use Opening a Dataset To open a dataset click on Fi e on the menu and select Open This will open a list of directory of files from which a chosen file can be loaded into AutoQuant and AutoQuant 5D Viewer for pro cessing For a list of file types that can be opened see Guidelines for naming your file s below Note When opening a folder in any of the following tutorials if the folder appears empty select All Files in the Files of type field 1 From the File menu select Open You will see a list of available file formats Supported Formats ims raw deb avz ids ics lif tif tiff stk 1nd bmp Ism zvi ipl iplab lei oif oib nd2 se AutoQuant X Dataset xml Legacy AutoQuant Dataset aqh Bitplane IMS File ims Raw Binary Data File raw deb avz Image Cytometry Standard File ids ics Leica LIF File Jif TIFF Image File
168. p and Bottom 20 Setting the Exposure Gain and Offset 22 Setting the Top and Bottom of the Scan 21 Shock Control 35 Sources 35 Single View 67 Slice Selection 100 Slice Viewer 69 Smoothed raw image 105 Software Copying Restrictions 8 Duplication 11 Software License Agreement 8 SOFTWARE USE RESTRICTIONS 9 Spatial Calibration 32 Specified Cross Talk FRET 116 Statistical Information displayed 112 Statistical Optimality 46 Statistics 112 STK stk 63 Sub resolution Microspheres 45 January 2012 Page 5 U Media Cybernetics Inc Manual and Tutorials Subtract Background FRET 116 Sub Volume Overlap 104 Subvolumes 103 Super Resolution 106 Table of Parameters 152 Table of Z Spacings Z Step Sizes 151 TERMINATION 10 The Brightness and the Darkness values 57 TIFF tif 62 Tile Horizontally 144 Tile Vertically 144 time lapse 147 TRANSFER RESTRICTION 10 Trans Illumination Lamp 30 Triple View 72 Tutorials for File Menu Items 61 Tutorials for PreProcessing Menu Items 74 Tutorials for the View Menu Items 67 Uninstalling 17 United States copyright laws 8 Unsaturated Image 23 Update AutoUpdate 146 US and International laws 8 RIGHTS 11 V Vibration Control 19 34 Vibration Sources 35 Vibrations Air table 34 Anti Vibration Table 34 From Cooling Fans 35 Relative Motion 34 Shock 35 Video Gain 24 Video Rate CCD 80 View 71 XY 72 XZ 72 ZY 72 Viscosity Ratiometrics
169. plied Optics 24 2 194 200 January 2012 Page 155 Media Cybernetics Inc Manual amp Tutorials Fay F S W Carrington and K E Fogarty 1989 Three Dimensional Molecular Distribution in Single Cells Analysed using the Digital Imaging Microscope Journal of Microscopy 153 2 133 149 Gerchberg R W and W O Saxton 1974 Super Resolution Through Error Energy Reduc tion Optica Acta 21 709 720 Gibson S F and F Lanni 1991 Experimental Test of an Analytical Model of Aberration in an Oil Immersion Objective Lens Used in Three Dimensional Light Microscopy Journal of the Optical Society of America A 8 10 1601 1613 Gordon G Berry G Liang X Levine B and Herman B 1998 Quantitative fluorescence reso nance energy transfer measurements using fluorescence microscopy Biophysical Journal 74 2702 2713 Grynkiewicz G Poenie M and Tsien RY 1985 A new generation of Ca2 indicators with greatly improved fluorescent properties Journal of Biological Chemistry 260 3440 3450 Hebert T R Leahy and M Singh 1988 Fast MLE for SPECT Using an Intermediate Polar Representation and a Stopping Criterion IEEE Transactions on Nuclear Science 35 1 615 619 Hiraoka Y J W Sedat and D A Agard 1987 The Use of Charge Coupled Device for Quan titative Optical Microscopy of Biological Structures Science 238 36 41 Hiraoka Y J W Sedat and D A Agard 1990 Determinat
170. ppendices Table of Z Spacings Z Step Sizes Table 1 Numerical Step Size for Oil Step Size for Water Step Size for Air Aperture NA n 1 515 n 1 333 n 1 00 0 2 18 83 16 61 12 37 0 3 8 3 7 33 5 43 0 4 4 6 4 08 2 99 0 5 2 95 2 57 1 87 0 6 2 02 1 76 1 25 0 7 1 458 1 256 0 875 0 8 1 094 0 935 0 625 0 9 0 824 0 713 0 443 1 0 0 663 0 552 0 250 1 1 0 528 0 429 Undefined 1 2 0 424 0 330 Undefined 1 3 0 339 0 242 Undefined 1 4 0 267 Undefined Undefined 1 5 0 192 Undefined Undefined DOF Depth of Field AZ slice DOF Wavelength A 0 50um A January 2012 Page 151 Media Cybernetics Inc Manual amp Tutorials Table of Parameters Table 2 File Name Numerical Refrac Modality X Y and Z X Y and Z Spacing Aperture tive Dimensions Pixels Microns Index FitcDapi_crop tif 1 4 1 515 Widefield 90 120 50 0 07 0 07 0 15 FitcDap_crop2d tif 1 4 1 515 Widefield 90 120 1 0 07 0 07 0 15 Blue_Green stk 1 4 1 515 Widefield 90 120 50 0 07 0 07 0 15 FitcDapi decon tif 1 4 1 515 Widefield 90 120 50 0 07 0 07 0 15 Malaria Red tif 1 3 1 518 Widefield 202 243 60 0 1299 0 1299 0 3043 Malaria _Green tif 1 3 1 518 Widefield 202 243 60 0 1299 0 1299 0 3043 Malaria_Blue tif 1 3 1 518 Widefield 202 243 60 0 1299 0 1299 0 3043
171. r icon in the Docking Controls is the Close icon x Clicking this icon will close the dialog The Data Browser dialog can be reopened by going to View gt Data Browser Depending on which dialog was open on the right side the dialog can be reopened from its source menu e g the 3D Deconvolution menu would be opened by going to Deconvolution 3D Decon 2 Dialog Box The Dialog Box appears on the right side of the application window and is opened either via a menu item or a toolbar icon If there are multiple dialogs open e g 3D Decon 5D Viewer and Statistics a tab will appear for each of the open dialogs If they are docked always displayed with a vertical tack icon the tabs will appear at the bottom of the dialog box If they are not docked not displayed all the time with a horizontal tack icon the tabs will appear along the right side of the application window Clicking on these tabs will display the tab that has been clicked on 3 Operation Status The Operation Status displays the status of all open operations in the appli cation window Any operation can be right clicked on and canceled Right clicking on a process also gives the option to show details if it is a process that has sub processes as well as to clear all completed processes This is helpful when the Operation Status becomes cluttered with various processes many of which are completed 4 Image Enhancement Dialog The Image Enhancement dialog contains the control
172. re detailed Bin factors of one or two are not recommend for large datasets because there may not be enough memory to hold all the triangles generated A large number of triangles also can slow down the rendering time and make rotation and interac tion with the viewer cumbersome e Clip If true and the oblique slice is present the oblique slice will cut the isosur face This is useful in showing a cut away view of the isosurface Display sides Renders an isosurface one the sides of the volume e Max threshold Picks the upper value at which the isosurface will be generated For fluorescent data this usually remains at the maximum intensity of the dataset and no isosur face is generated using this threshold For brightfield data decrease this value to remove back ground and dim feature Min Threshold Picks the lower value at which the isosurface will be generated For fluorescent data increasing this will remove background and dim features January 2012 Page 138 Media Cybernetics Inc Manual amp Tutorials Use Smoothing Applies a 3x3 two dimensional Gaussian smoothing filter to the data before generating an isosurface This removes noise and yields smoother surfaces This is the recommended default Wire Frame When true shows the triangles which make up the isosurface Ortho slices If an ortho slice is added to the scene an Ortho Group is added to the object tree with three child ortho sli
173. reated This l DOPE ROI is done using the Region of Interest icon in the dataset window 1 Open a dataset January 2012 Page 58 Media Cybernetics Inc Manual amp Tutorials 2 Click the Region of Interest icon a drop down menu will appear showing the available modes for drawing a region of interest Square Circle Shape Free hand as well as options to Delete Selected Delete All Select Objects and an option to draw lines 3 Select the Square region of interest 4 Click and drag a region of interest When the ROI is of the desired shape and size click again to finalize the ROI After doing that a region of interest will appear on the dataset 5 Repeat steps 2 4 several times creating several regions of interest 6 Click on one of the regions of interest in the dataset 7 Click the Region of Interest icon again and select Delete Selected The region of interest will be deleted 8 Click the Region of Interest icon again and select Delete All and all regions of interest will be deleted Saving and Copying the Current View There are two ways to save the current view of a dataset One is to save the current view to a tiff file the other is to copy the current view to the clipboard so that it can be pasted into a document 1 Open any dataset or from the tutorial data use colorpollenc tif from the MultiChannel folder LJ 2 Click the Save Current View Icon E 3 Select the location and n
174. resis that is inherent in the microscope s focusing mechanism Collecting Bias and Flatfield Frames Cooled CCD Cameras Bias and flatfield dataset frames are used by AutoQuant to automatically calculate the cooled CCD camera s dark current distribution pre amplifier bias and flatfield non uniformity This pro cedure needs to be repeated every day or preferably for every dataset collection This process compensates for potential misalignments in the optics which may drift from day to day and to compensate for dust that may settle in the optical train which also changes from day to day Collect two bias frames and one flatfield frame The bias frames are taken with the camera shutter closed or with light to the camera blocked Select a charge integration time for the first bias frame The accuracy of the bias frame dataset will increase with increasing charge integration time As a rule of thumb use 60 seconds for the first bias frame The charge integration time for the second bias frame needs to be twice that of the first bias frame so as a rule of thumb use 120 seconds for the second bias frame January 2012 Page 29 Media Cybernetics Inc Manual amp Tutorials Collect a flatfield frame This will be used to calculate the flat field non uniformity of your cam era A flat field sample can be provided in a number of ways When using transmitted light brightfield the simplest way is to remove the sample from the micro
175. return or refund is entitled after the 90 day period has expired Upon such a valid return the dealer is obligated to refund the price of the product less shipping costs The Representative will return the product materials to Media and Media will refund the discounted price of the product less ship ping costs to the Representative January 2012 Page 12 Media Cybernetics Inc Manual amp Tutorials Installation Transferring a License Uninstalling and Hardware Requirements Sales Information Media Cybernetics Inc 4340 East West Highway Ste 400 Bethesda MD 20184 E mail info mediacy com Website www mediacy com Ph 301 495 3305 or 800 263 2088 North America only Fax 301 495 5964 Technical Support Internet techsupport mediacy com Ph 301 495 3305 or 800 263 2088 North America only Fax 301 495 5964 Software and Hardware Requirements Windows Systems AQX supports the following operating systems e Windows XP 32 bit e Windows 7 32 and 64 bit You will also need Minimum Requirements e Processor Intel dual core processor e RAM 3 GB Memory e Free Disk Space 2GB on installation driver free space for images 20GB e OS Windows XP 32 or 7 32 or 64 e Graphics Card NVIDIA Graphics Card for 5D Viewer performance Recommend Requirements e Processor Intel quad core 64 bit processor current recommended model Core 17 2600 e RAM 8 GB Memory e Free Disk Space 2GB on installation drive ded
176. roper confocal settings so if you have a choice between fully confocal behavior and noise for deblurring pur poses it is better to sacrifice signal level noise for a proper confocal behavior Remember that AutoQuant thinks that you have a fully confocal microscope so it will try to reconstruct a fully confocal PSF As you open the pinhole aperture your microscope behavior begins to lean towards widefield behavior and the true PSF begins to have widefield character such as a flaring hour glass like shape This behavior in a sense confuses the AutoQuant software and it will still attempt to reconstruct a fully confocal PSF so you should avoid this condition if possible If this condition cannot be avoided AutoQuant still has the robustness that it may provide a nice deblurred result but this will be dependent on your sample and other experimental setups and is less predictable than with fully confocal settings Consider the signal to noise requirements explained below January 2012 Page 37 Media Cybernetics Inc Manual amp Tutorials Less Than Fully Confocal Conditions If you find it unavoidable to have the aperture setting opened wider than the fully confocal posi tion then we recommend likewise broadening your XY sample spacing by the same amount For instance if your confocal aperture is broadened to twice the diameter of its fully confocal posi tion then you should broaden your XY sample spacing microns per
177. s If it is not possible to obtain the calibration images DDd DAd AAd DDa DAa and AAa the Cross Talk Coefficients tab can be used to estimate the Cross Talk See the Use Cross Talk Coeffi cients Tutorial for more information on this Specified Cross Talk Donor Only Acceptor Only Donor and Acceptor This section allows you to select which datasets to correct for cross talk Typically if both donor and acceptor calibration images are available it is best to select Donor and Acceptor However if only one or the other set of calibration images are available for instance only the donor calibra tion images then select that set to correct in this instance select Donor Only Calibration Images Checking this button allows you to load in calibration images with which to perform FRET By default this button is checked Cross Talk Coefficients Checking this button allows you to enter cross talk values in lieu of using calibration images For each image donor and acceptor there are three values to be filled in These should be between 0 0 and 1 0 Ideally settings should be set between 0 0 and 0 5 Settings of 0 5 and 1 0 are accept able but not ideal The only way to verify is to visually inspect the resultant images For this rea son unless you know what these coefficients are from previous experiments it is ideal to have calibration images January 2012 Page 115 Media Cybernetics Inc Manual amp Tutorials PreProc
178. s to on the order of 50 to 200 micrometers e g neurons or brain slices The acquired image will contain both sharp in focus features originating from the plane of focus as well as hazy blurry features originating from out of focus planes that are above and below the plane of focus After capturing this image electronically we can refocus the micro scope to an adjacent plane that is on the order of a depth of field DOF away and acquire a sec ond image This process can be repeated until the entire specimen has been scanned The resulting dataset will be a 3D dataset or in other words a 3D image However it will be a rather poor image because it will contain all of the out of focus haze and blur The purpose of deconvolution is to remove the haze and blur and to restore sharpness and clarity to the image as illustrated for a neuron image below The effects of deconvolution are usually most pronounced in the axial direction z axis as illus trated for a confocal image of a rat neuron in the image that follows January 2012 Page 40 Media Cybernetics Inc Manual amp Tutorials a Top View b Side View c Top View d Side View Figure 2 Confocal image of a CA3 hippocampal rat neuron Panels a and b show the projections of the dataset from the top and side respectively Panels c and d show the same dataset after deconvo lution The improved axial resolution becomes especially important when accurat
179. s to adjust the maximum and minimum intensities by moving the red and orange triangle sliders at the bot tom of the graph and the gamma by either entering a value into the textbox labeled Gamma or by adjusting the green triangle slider at the bottom of the graph 5 Data Browser The left side of the application window contains the Data Browser in which is contained the Data Manager Data Properties Summary and Image Enhancement Dialog The Data Browser can be docked or undocked as described in the Docking Controls section above 6 Data Properties The Data Properties section contains the image information for the selected dataset Included in this information are the image dimensions image spacings channel informa tion number of slices modality information as well as space to enter notes on the dataset January 2012 Page 51 Media Cybernetics Inc Manual amp Tutorials 7 Data Manager The Data Manager section displays the opened datasets in a file tree format With multiple datasets open the active dataset can be selected by clicking on that dataset in the Data Manager or by clicking on the dataset display in the application window Additionally the display of a dataset can be adjusted by selecting or deselecting channels slices and or time points to be displayed This can be done by clicking on the box next to each channel time point or z slice If there is a green check that item will be displayed if there is no
180. scope and grab an image of a blank stage A rather simple method with transmitted light brightfield is to take the biological sample extremely out of focus to a point where the image shows a completely flat field When using fluorescence the most reliable way is to prepare a flatfield slide by placing a drop of flu orescent dye onto a microscope slide and sealing it with a coverslip In case any debris particles may be contained in this droplet it is best to grab this image with the droplet well out of focus Some care must be taken to ensure that no out of focus remnants of debris appear in the image and that no other particles on the slide happen to come into focus See Table 1 Figure a and Fig ure b for an illustration of these conditions In setting the exposure time when using transmit ted light brightfield first try the same exposure time that was used in collecting the optical sections during the axial scan Adjust the exposure time to ensure the conditions described in Fig ure C and Figure D When using fluorescence since the drop of dye will likely fluoresce at a dif ferent intensity than the biological sample it is especially important to carefully adjust the exposure time to ensure the conditions described in Figure C and Figure D To emphasize it is important that the maximum gray value in the image is at about 75 85 of the well capacity of the camera and that there are no saturated pixels in the image Carefully re
181. sed by entering a positive number by the Constant Value number 3 Select Subtraction from the Operation drop down menu and click the OK icon Y 5D Viewer will subtract the Pollen deb dataset from the PollenPlus5 deb dataset creating a new file 4 The resultant dataset PollenPlus5_AlgebraResult may now be saved as PollenPlus4 deb 4 Close all views before going on to the next section January 2012 Page 91 Media Cybernetics Inc Manual amp Tutorials Tutorials for Deconvolution Menu Items Performing 3D Deconvolution Note This section applies to users of AutoQuant Blind Deconvolution is considered to be the most accurate algorithm available with AutoQuant Blind Deconvolution does not require the calibration and measurement of the Point Spread Func tion PSF Blind Deconvolution can be used with Fluorescence Widefield Laser Scanning Confocal Brightfield Transmitted Light Spinning Disk Scanning Confocal or Two Photon Fluorescence In addition structured illumination images from the Optigrid Structured Light system and the Quorum Angstrom systemn have been successfully deconvolved using the confocal modality 3D Blind Deconvolution is an iterative and constrained algorithm It is iterative in the sense that it repeats the same computational operations many times while converging to the enhanced image solution It is constrained in the sense that it only accepts deconvolved images that have the cor rect m
182. see if the wavelength values are reasonable January 2012 Page 110 Media Cybernetics Inc Manual amp Tutorials Tutorials for Analysis Menu Options Counting and Tracking The Counting and Tracking feature can be used to count objects within a dataset track those objects if it is a time series dataset and give statisitcs on the objects and if applicable their move ments and changes over time 1 From the File menu select Open Navigate to the Counting and Tracking folder click on Nitellal tif Check the Use Sequence Detection box and then select Time from the drop down menu that appears 2 From the Analysis menu select Counting and Tracking 3 Adjust the threshold of the image in the Counting and Tracking Preview Window by mov ing the minimum red slider and maximum orange slider sliders in the histogram in the Count ing and Tracking dialog e The goal is to adjust the thresholding such that the objects to be counted appear as separate white areas 4 Once the dataset is thresholded properly click the Count All T button to count all of the objects in all time points The section at the bottom of the Counting and Tracking dialog will con tain all of the statistical information about each object found and beneath that grid will be dis played the total number of objects found 5 Click the Export Objects button This will open a new window with just the segmented objects displayed This image can be saved
183. sponds to the Z spacing or step size Viewing in the Slice Viewer is a good way of examining the degree of optical sectioning in the dataset It is advisable to examine datasets in a Slice Viewer because it allows for viewing of subtle and not so subtle changes from slice to slice Make the Pollen deb XY Max Projection dataset active by clicking within its view 1 From the View menu select Slice Viewer to generate an optical slices projection of the dataset from the XY perspective Note When a projection operation is selected the resultant view formed depends upon the cur rently active view For example in the previous step the XY Max projection was the active view Thus when the Slice Viewer menu item was selected optical slices were created from an XY per January 2012 Page 69 Media Cybernetics Inc Manual amp Tutorials spective 2 To step through the optical slices move the slider on the left side of the dataset window This can be accomplished by clicking on the button and dragging it up and down 3 The Slice Viewer also has the ability to cycle through all the slices automatically To do this click the Play icon w on the left side of the dataset window 4 Clicking the Stop button m will halt the Slice Viewer Note Do not close Pollen deb XY Max Projection Please continue on to the next section January 2012 Page 70 Media Cybernetics Inc Manual amp Tutorials Synchronizing Slice View
184. successful when the sample spacing is finer than 1 2 the Rayleigh width If in plane deblurring is desired we will sample at 0 1 um or finer when the 1 2 Rayleigh width is 0 25 um Axial deblurring is substantial with widefield optics regardless of the z sampling The rule of thumb for confocal datasets without deblurring is to have the optical sections spaced according to the confocal spot size along z which is the reso lution element along the z direction calculated by the following equation Webb et al 1995 A 1 44n NA where A is the wavelength in microns um n is the refractive index of the immersion medium and NA is the numerical aperture of the objective lens If Az is 0 5 um we will sample at 0 5 um or finer per slice The noise suppression mentioned above requires this type of fine sampling January 2012 Page 44 Media Cybernetics Inc Manual amp Tutorials What is Blind Deconvolution Blind deconvolution is a term that is used to describe methods of deconvolution which do not require the point spread function PSF of the system to be explicitly known prior to the deconvo lution AutoQuant is based on blind deconvolution Indeed a reconstructed estimate of the PSF is produced by AutoQuant concurrently with the deconvolved image dataset There is no need to provide AutoQuant with a PSF of any kind The important practical implication of blind deconvolution is that it adapts to the real PSF of the microscop
185. t 5D Viewer by going to Visualization 5D Viewer 3 Display an isosurface as the current view 4 In the object tree turn the volume rendering on 5 Turn the volume s child object named Ch 2 520 nm off 6 Turn the isosurface child object named Ch 3 520 nm off This will display the blue channel as a volume rendering and the green channel as an isosurface Insert picture here Time series When a multi time point dataset is loaded into AutoQuant 5D Viewer the time series scrollbar is shown at the bottom of the display window Time 4 41 a Dl gt The scroll bar can be used to move between time points one at a time using the scrollbar or the play forward and play back buttons The time series can be continuously played using forward and back play buttons There will be a rendering delay the first time a time point is visited The volume must be loaded from disk preprocessed and textures generated This information is cached for each time point January 2012 Page 130 Media Cybernetics Inc Manual amp Tutorials visited The next time the time point is displayed it is instantaneously rendered Object Parameters When an object or child object is select in the object tree the options for that object are displayed in the bottom of AutoQuant 5D Viewer dialog If an object containing child objects is selected and its parameters are changed these changes are also applied the child objects January
186. t Save As 7 When prompted for a file name to save the cropped image type Neuron_crop1 in the File name field and click the Save button You may close the Neuron_cropl view January 2012 Page 76 Media Cybernetics Inc Manual amp Tutorials Note Do not close Neuron deb Please continue on to the next section Extending Slices In order to protect meaningful data in the outermost slices of a 3D dataset the image can be extended have additional optical slices added to it along its Z axis using the Extend Slices fea ture This function allows you to generate additional image slices based on linear extrapolation and appends them to the top and bottom of an image It is not the same as adding blank slices to the top and bottom of a stack as mentioned in 1 below An extended dataset will have false slices attached to its top and bottom These false slices serve the purpose of a buffer to keep the guardband region from overlying meaningful features thereby keeping the meaningful data from being obscured in the deconvolution process This tutorial continues with the Neuron deb dataset from the Crop tutorial above 1 From the Processing menu select Extend Slices The Extend Slices Dialog will appear You have the option of extending the data by using false slices or by using the Zero fill option which when selected adds blank slices to the Top and Bottom of the dataset devoid of any infor mation The Extend Depth indicates
187. t deblurring In plane deblurring is most successful when the sample spacing is finer than 1 2 the Rayleigh width If in plane deblurring is desired sample at 0 1 um or finer when the 1 2 Rayleigh width is 0 25 um Axial deblurring is substantial with widefield optics regardless of the z sampling The rule of thumb for confocal datasets without deblurring is to have the optical sections spaced according to the confocal spot size along z which is the resolu tion element along the z direction calculated by the following equation Webb et al 1995 _ 1449 AZ 43 where is the wavelength in microns um n is the refractive index of the immersion medium and NA is the numerical aperture of the objective lens If Az is 0 5 um sample at 0 5 um or finer per slice The noise suppression mentioned above requires this type of fine sampling January 2012 Page 21 Media Cybernetics Inc Manual amp Tutorials Setting the Exposure Gain and Offset Cooled CCD Cameras With cooled CCD cameras the exposure time is usually adjustable Consult the manufacturer s manual to determine how to change the exposure time To set the exposure time for collecting optical sections use the following procedure Focus onto a point near the center of the sample see Figure 1 Adjust the illumination light to a setting that you desire If you are using a transmitted light brightfield microscope adjust the light so that you can view the sample com
188. t the load If you have a stage focusing micro scope where the stage moves up and down as the focus knob is moved then always step the focus in the direction where the stage is moving upward This will cause you to scan the sample in the direction from top to bottom If you have a nosepiece focusing microscope then always step the focus in the direction where the nosepiece is moving upward This will cause you to scan the sample in the direction from bottom to top When adjusting the focus from one sequential slice to the next move the knob very slowly and cautiously Also be careful to move the knob only in one direction during the scan Once you have started grabbing frames never reverse the direction of the fine focus knob not even a miniscule amount and not even just once Doing so by even a very small amount will ruin the scan due to backlash and hysteresis of the focusing system To compensate for erroneous hysteresis effects do the following before grabbing the first frame First focus into the specimen If you have a stage scanning microscope first move the stage down by 20 micrometers or more below the position where you wish to grab the first frame Then being careful to move the knob in only one direction as mentioned earlier once you are in this stage of operation never reverse the direction of the knob slowly and carefully move to the posi tion where you want to grab the first frame This operation counteracts a hyste
189. tabs can be edited either by clicking within the desired field and entering text or numbers into the field or by selecting an item from the dropdown menu within the field January 2012 Page 55 Media Cybernetics Inc Manual amp Tutorials Using the Image Enhancement Tool Depending on the display of the machine running the software images may appear too dark or in other cases too bright To compensate for these differences it is possible to perform a Gamma correction on a particular view 1 From the File menu select Open Navigate to the SmallHip folder open the SmallHip avz dataset and generate an XZ view 2 Towards the bottom of the Data Browser it says Image Enhancement To display the Image Enhancement dialog click the expand button Y the Image Enhancement dialog will now be displayed 3 Enter a value of 1 2 in the Gamma field Notice that the triangle located in the center of the histogram moves slightly as well as a change in the image The Gamma can also be adjusted by clicking on that triangle on the histogram and dragging it to the left or right Note A Gamma value of 1 0 restores the image to its original state Values greater than 1 brighten the image whereas values below I darken the image Gamma values may be between 0 1 and 10 0 4 You may also adjust the darkness and brightness of the image in any of several ways Enter a new value in the Intensity text boxes The Min Intensity text box on th
190. tates the scene right counter clockwise around the y axis Left arrow rotates the scene left clockwise around the y axis PgUp increases the minimum threshold by 5 PgDn decrease the minimum threshold by 5 Image Display Controls The image display controls affect all view objects volumes isosurfaces ortho slices and oblique slices When these tool buttons are pressed they change the associated parameters of each object The exact value can be viewed and changed in the parameter list of each object The increase decrease threshold tool buttons 1 4 can be used to change the minimum threshold of volume renderings isosurfaces ortho slices and oblique slices This reduces the amount of background so that bright features are easier to see If a volume is displayed as a minimum projection the threshold buttons affect the maximum threshold instead January 2012 Page 133 Media Cybernetics Inc Manual amp Tutorials 4 The increase decrease brightness tool buttons can be used to change the brightness of the volume rendering ortho slices and oblique slices Use these buttons to brighten up dim data Coloring Channel coloring is done through the Summary and Coloring tabs in the Data Manager When the mouse hovers over a channel line on the summary tab a drop down combo box becomes visible 402 x 300 x 60 Tim Paints Spacinas 0 13 x 0 13 x 0 304 pum MM Chi yy 650 nm ua Che 520 nm MM
191. the Data Browser had been closed it will not reopen when Show Dock Windows is selected Hide Dock Windows This function will dock all displayed windows in the application Data Browser Dialogs and Operation Status January 2012 Page 144 Media Cyberentics Inc Manual amp Tutorials Window List This feature lists all current views on the screen January 2012 Page 145 Media Cybernetics Inc Manual amp Tutorials Help AutoQuant 5D Viewer Help This feature provides a help window opened to the table of contents Check for Updates Media Cybernetics products incorporate an AutoUpdate feature It is configured to check for updates once a month if your machine is connected to the internet If you would like to check for an update before that select Check for Update from the Help menu You will be prompted to click OK to search the internet for updates If you do not have a LAN connection but are using some other internet connection you will be prompted to click Configure to configure your internet set tings Any internet connections available on your machine will be listed in the Use Dial Up Con nection drop down menu Select the service you wish to use then click OK This will take you back to the original dialog box at which point you can click OK again AutoUpdate will now search for any available updates About This feature shows the version and copyright information about the product Version X3
192. the scatter from deep tissues In practice the z resolution is not good compared to the x y resolution Resolution along the z axis is typically 0 5 to 3 microns This means that structures January 2012 Page 147 Media Cybernetics Inc Manual amp Tutorials are elongated along the z axis Applying AutoQuant to a confocal image stack improves the z resolution dramatically often achieving beyond the theoretical 0 5 micron limit The signal to noise ratio of the deconvolved images is greatly improved and poor signal to noise is otherwise common with confocal microscopes especially in live cell imaging and where photobleaching is a problem Q What types of image files are supported by AutoQuant A AutoQuant supports most common image formats including 8 12 16 bit TIFF s Universal Imaging s STK BioRad s PIC Leica s native files BMP s AVI s Olympus Fluoview raw 8 12 16 integer 32 bit floating point data and others Q Can AutoQuant handle the saturated portions of my image volumes A Yes Not only can AutoQuant handle images that contain saturated areas it can often recover information from within these saturated sections This is possible in situations where there are non saturated areas nearby the saturated areas The blur in the nearby non saturated region contains information about the saturated region which AutoQuant restores The degree to which this is effective depends on the size of the saturated area
193. this smearing appears akin to a linear motion blur in the axial direction For an image which is diffraction limited as with empty magnification this smearing appears as a substantial nonisotropic blur or nonisotropic spatial resolution with most of the smearing along the axial direction 3 Owing to this light rejection far fewer photons are detected so a rather substantial quan tum photon noise component may be seen in the raw dataset This noise component routinely causes limitations in protocols where very fine structures need to be seen For instance it limits our ability to detect a void in the fluorescence concentration that is often useful to pinpoint impalement sites of electrophysiology probes Turner Szarowski et al 1991 and it limits our ability to determine if a dendritic spine is present It is difficult to judge whether a void in the image intensity is due to a fine structure that is really there or if it is due to erroneous random fluctuations caused by the statistical nature of quantum noise In the case of confocal fluorescence microscopy it is the purpose of deblurring algorithms to reduce these three undesirable effects and to thereby improve the utility of the microscope Deblurring is best carried out when the dataset is seemingly oversampled In other words best results are achieved when the pixel or optical section spacing is finer than what would be used normally without deblurring In plane deblurring is most
194. tics parameters that you use most January 2012 Page 47 Media Cybernetics Inc Manual amp Tutorials often Enter the system parameters once and then simply recall the settings each time you load an image into AutoQuant Sample RI and Distance from Coverslip added to Data Manager These two parameters in the Data Manager are now visible so that you can apply them to the core of the deconvolution algorithm This improves accuracy which leads to better results Faster Deconvolution Times This version of AutoQuant uses optimized math libraries fewer temp files and improvements to various algorithms which should result in faster processing times for certain systems Faster loading of images into the batch processing queue The batch queue has been optimized to reduce the loading time for batch processing The larger the image set the easier it is to queue up and begin the process e New Additions to the Objective Camera and Dye lists Many new entries have been added to update these lists with the latest hardware and dyes available in the market Basic Principles Underlying the AutoQuant System The basic mathematical and conceptual principles underlying the AutoQuant system have been widely published in scholarly journals and conference proceedings worldwide Constrained Maximum Likelihood Estimation may be thought of in two ways First consider a 3D fluorescent sample We may think of the true 3D probe concentration as an
195. tif tiff MetaMorph STK File stk MetaMorph ND File nd Windows Bitmap File bmp Carl Zeiss LSM File sm Car Zeiss ZVI File zvi Scanalytics IPLab File ipl iplab Leica LEI File lei Olympus OIF OIB File oif oib Nikon ND2 File nd2 ImagePro Sequence File seq BioRad PIC File pic All Files 2 From the Tutorial Data directory select the Widefield folder Select the dataset FitcDapi_crop tif and click on Open Note Various types of files including AutoQuant file AutoQuant 5D Viewer file 12 bit data 16 bit data Bio Rad PIC TIFF File 8 bit character data 32 bit float and STK File can be handled If your dataset is stored in a series of slice files the software will recognize the series and ask if you wish to load the entire series or simply the file that you selected If your dataset was stored as 12 bit data on a UNIX machine and you are running the software on a PC you must select 12 bit swapped for UNIX as the file type This is also true for 16 bit data and when transferring 12 bit and 16 bit data from a PC to a UNIX machine January 2012 Page 61 Media Cybernetics Inc Manual amp Tutorials 3 The FitcDapi_crop tif dataset will be loaded and displayed in the XY Max Projection view Note If it is necessary to enter the dimensions of a particular dataset the first time it is loaded you will be prompted to do so The next time that dataset is select
196. tiplier 43 Piecewise Line 114 Pixel Sizes 32 Planeoffocus 40 January 2012 Page 4 Media Cybernetics Inc Manual and Tutorials Point Spread Function 40 45 92 point spread function 147 Poisson Modeling 46 poor signal to noise 148 Pre amplifier Bias 29 PreProcessing Menu Attenuation Correction 84 Crop 76 Depth Attenuation 84 Optical Density Correction 84 Probe Concentration 48 Processing Current Slice 98 Processing Many Data Sets 47 Processing Stack 99 PSF Experimental Measurement 45 Missing Cone 46 PSF settings 103 QuantumPhoton Noise 44 RO 117 Ratiometrics 120 B SI2 Sh2 121 Gaussian Smoothing 122 Grynkiewicz Equation for Ion Concentration 121 Kd 122 Rmax 121 Rmin 121 Viscosity 122 Rayleigh Depth Of Field 21 Rayleigh Width 21 44 Refractive Index 21 Regions of Interest 141 Resize Factor 78 Resize Methods 78 Resizing an Image 77 Resolution 46 Resolution Improvement 46 Reverse Document Contrast 90 Different from Invert 90 Rmax Ratiometrics 121 Rmin Ratiometrics 121 Roughness Penalty 103 RS 170 Cameras 24 30 SA Detection 105 Sample Depth 21 Nonhomogeneity 45 Refractive index 45 Saturated Image 23 Saturated Pixels 25 saturated portions 148 Saturated Regions 22 Saturation Detection 23 Save As 62 Save Current View 63 Saving Individual Channels As Separate Stacks 62 Scalebar 141 Scan Actual Axial Distance of 20 Reversing Direction 29 Thickness of the Scan 21 To
197. to ensure this condition If you are using an oil objective lens for which the NA is higher than the highest available NA of the condenser for example a 1 4 NA objective lens and a 1 25 NA con denser then use the highest available NA that your condenser will allow and use a drop of oil on the condenser assuming your condenser allows you to use oil with it Collecting Optical Sections Setting the Top and Bottom of the Sample and of the Scan You need to select the top and bottom slice of the sample By performing a through focus opera tion with either direct eye or video camera viewing you should judge where the top and bottom of the sample are Note The top and bottom of the scan are not necessarily the same as the top and bottom of the sample The rule of thumb for selecting the top and bottom of the scan is described below using the terms illustrated in Figure 1 below o Top of Scan Top of Sample 7 Bottora of Sample Bottora of Scan Figure 1 Shows the geometry and illustrates the terms used for 3D widefield dataset collection Note The actual axial distance that is scanned is about twice that of the sample The top and bot tom are over scanned by approximately half the thickness of the sample The above rule of thumb January 2012 Page 20 Media Cybernetics Inc Manual amp Tutorials provides the best results If you have limited disk or memory space then you may not be able t
198. torials Index Numerics 2D Blind Deconvolution 108 2D Deconvolution 94 2D images 147 2 Photon Data 39 Sampling Criterion 39 3 D Data Collection Avoiding Backlash and Hysteresis 29 Bias and Flatfield Frames 29 Bottom of Sample 20 Confocal Data 35 RS 170 Cameras 24 30 Setting Exposure Gain Offset 22 Setting Number of Slices 21 Signal to Noise Ratio 38 Spatial Calibration 32 Top of Sample 20 Vibration Control 34 3 D Data Set 40 3D Deconvolution 92 5D Tutorial Time series 130 5D Viewer 128 Axis 141 Camera 142 Grid 141 Image Display Controls 133 Interface Overview 128 Introduction 128 Keyboard hotkeys 133 Object Parameters 131 Requirements 128 Scene 142 Toolbar 129 Tutorial 130 132 Volume rendering 134 5d Viewer Navigation 133 A About 146 About AutoDeblur 146 Acceptor Quantum Yield Qa FRET 116 Accurate deconvolution 147 Adaptive PSF 96 106 Advanced DIC Image Settings 100 Algebra 90 Alignment 19 Channel to Channel 88 Manual 88 Application of AutoDeblur Deconvolution Package 43 Attenuation Correction 85 Depth Attenuation 81 Excitation Light Absorption 81 Fluorescent Emission Light Absorption 81 Photobleaching 81 AutoDeblur Installation 13 Software and Hardware Requirements 14 Starting 15 Automatic Alignment Noise Thresholding 88 AutoQuant Incidental Damages 7 Warranties 7 Axial Sampling Distance 21 Axial Smearing 44 B S12 Sh2 Ratiometrics 121 Backlash 29 Bad Pixel Treatme
199. ttenuation Correction The Attenuation Correction will be performed automatically on the dataset and the result will be displayed in a new dataset window January 2012 Page 81 Media Cybernetics Inc Manual amp Tutorials 4 Compare the raw XZ Sum Projection dataset to the newly corrected XZ Sum Projection dataset The bright to dark effect has been diminished in appearance Attenuation Correction is most often applied to Confocal datasets where there is frequently a decrease in image intensity due to attenuation light scatter and spherical aberration as the depth into the sample is increased Apply Attenuation Correction to the raw dataset before processing 5 Close all views before going on to the next section Equalizing the Background Note This section applies to users of 5D Viewer only In order to equalize a dataset that has an uneven background use the Background Equalization feature This function allows you to remove unnecessary background from an image 1 From the File menu select Open Navigate to the SmallHip folder and select SmallHip tif 2 From the Processing menu select Background Equalization The Background Equaliza tion box will appear Adjust the Background by typing in the Background Equalization box or by scrolling the bar until the Maximal Equalization value equals 45 3 Check the Remove Negative Intensities box to enable it When this box is checked any background points that have negat
200. tton na is clicked the views automatically scroll through their slices Adding and removing Datasets from Slice Synchronization Datasets can be added and removed from Slice Synchronization by clicking the Slice Synchro nizer icon in that dataset window If a dataset is synchronized a blue outline will appear around the Slice Synchronizer icon If it is not synchronized there will be no outline around the icon Selecting a Single View There are several views in which the dataset is presented The dataset can be viewed from three orthogonal perspectives The choices available are XY XZ and ZY where X represents the image width Y the image height and Z is the depth or may be thought of as the optical axis in microscopy applications When a choice is made the single view that is generated will be for the currently active projection For example if a Sum Projection is the active view selecting ZY will generate a ZY Sum Projection of the view January 2012 Page 71 Media Cybernetics Inc Manual amp Tutorials 1 Navigate to the TLB folder and open the TLBview tif dataset Select XZ from the Single View option under the View menu Note Transmitted Light Brightfield datasets will open in the Min Projection view XY This is the view of the face of the dataset or the front view of each slice e XZ This is an edge like view of the dataset or of one of its slices It can be thought of as a horizontal slice throug
201. uary 2012 Page 56 Media Cybernetics Inc Manual amp Tutorials Image Enhancement may be done on a single color channel Red Green Blue or on All Chan nels by activating the appropriate choice under Channel 5 Close the Image Enhancement box by clicking the up arrow button Q Note Only in a Slice Viewe will the maximum and minimum values for Brightness and Darkness Level be the true maximum and minimum values of the entire dataset In a time series dataset the changes can be applied to all time points by clicking the Apply to all button and conversely they can be reset by clicking the Reset all button Status Display The Status Display feature will show the status of all current processes within the AutoQuant 5D Viewer platform The display is at the bottom of the application window To display the Status Window select Display Status from the View menu Within the Status Display processes can be canceled by right clicking on that process in the Status Display and selecting Cancel Addition ally the Status Display window can be cleaned up by right clicking and selecting Clear Com pleted All completed or canceled processes will be removed from the Status Display Lastly the details of a process can be displayed by right clicking on that process and selecting Show Details This will open a new tab in the Status Display showing the details of the processes that are occurring or have occurred in that process Using the
202. up of the 3D Inverse Filter dialog This control specifies the spherical aberration correction from 15 to 15 no correction is 0 The values used for SA calculations are wavelengths You may obtain an estimated value using the Auto Detect method the Manual method or the Calculate SA method This will fill in the SA value with the estimate which then can be used in conjunction with the deconvolution or inverse filter If you have already performed an SA estimate calcuation for the conditions under which the image was collected you may enter that previously calculated SA value directly Auto Detect SA This option extracts the SA correction factor from the image data using a pat ented blind optimization technique and displays the results in the in Spherical Aberration control so that you can see if the wavelength values are reasonable It is recommended that Optical Den sity Correction be performed before SA detection Manually Detect SA When the Auto Detect Manual option is checked a previous SA estimate calculation for the conditions under which the image was collected should be entered manually using the Wavelengths of Phase Error edit box or scroll bar January 2012 Page 109 Media Cybernetics Inc Manual amp Tutorials Calculate SA This option calculates the spherical aberration based on Obj Lens immersion RI Specimen Embedding RI and Distance from Coverslip It displays the results in the SA control to that you can
203. utoQuant X3 Dongle Installation and Requesting a Permanent License All Media products will run for one month following their initial installation on your computer During this time you will need to contact Media for a permanent license so that you can continue using your software beyond the initial trial period Once you have purchased a permanent license you will receive a CD and dongle at which point you can follow the instructions below Updating the Dongle If you have purchased additional functionality or an extension to your maintenance or license agreement you will need to update your dongle to reflect the changes in you license 1 Click Start gt Programs gt Media Cybernetics gt AutoQuant X3 gt Dongle Update 2 You will be presented with the dongle update screen 3 Contact Media Cybernetics Inc Provide the User Registration ID and Dongle ID shown in the Dongle Update Utility dialog By Email customerservice mediacy com e By phone 301 495 3305 4 If the change that must be made has been approved you will receive an update code in the form of a long string of letters and numbers Enter this string into the Dongle Update Code text box and click Update Dongle 5 Upon successful update you will receive confirmation and the update will take effect the next time you run your AutoQuant product Starting the AutoQuant Software Windows Systems To start 1 Choose the Start button at the lower left hand cor
204. ving power in time lapse movies Q Can I process many images as a batch A AutoQuant contains functions that permit you to set up a batch of images that will then be sequentially processed After setting up the parameters for your deconvolution you may choose to place the image into a queue that you launch at a later time This is useful for laboratories who do a lot of imaging and for researchers performing time lapsed observations of 3D volumes These laboratories can set up their images to process a batch of images overnight and have them all ready for analysis in the morning Q How can I easily inspect my images before and after deconvolution A Not only does AutoQuant contain five different types of deconvolution adaptive 3D blind nearest neighbors no neighbors inverse filtering and 2D deconvolution but it also provides many different ways to visualize the raw and processed images AutoQuant enables a 3D dataset to be visualized using several different projections and orientations Each volume can be viewed from XY XZ and ZY orientations as a maximum minimum summed or voxel gradient projec tion or as individual optical slices No other package currently on the market combines these powerful processing tools with an environment for visualizing the results Q Can I process images that were acquired with a confocal microscope A Yes Confocal microscopes conventional two photon and spinning disk are effective at reducing
205. which all others will be aligned is chosen The default setting is to let the application select the base slice Deselecting Set Automatically will enable the slice scroll bar with which the user will then need to scroll to the desired base slice Alternatively the user can enter the slice number into the text box to the right of the scroll bar January 2012 Page 88 Media Cybernetics Inc Manual amp Tutorials Advanced Settings Deselecting the Use Recommended Expert Settings and then clicking the Change Expert Settings button will open the Advanced Settings section It is recommended that the default settings be used Clicking this button again with the Advanced Settings dialog showing will close the Advanced Settings section Image Warp Severity This feature corrects for warp due to motion between frames The more serious the warp the lon ger the processing time There are five available settings None Tiny Normal Severe and Very Severe Void Region Fill This function will put a frame like cover over the areas that no longer contain data After the slices are aligned gaps and spaces are created at the edges This function chooses how those areas are filled in Note The Frame may have straight scalloped or irregular edges to it Black This will put a black frame on the edge of the image where the aligned data has left a void This is recommended for Fluorescence and Darkfield datasets Random This will place a fram
206. will be deconvolved As the 2D deconvolution proceeds the preview window will be updated with the results 3 After the 2D deconvolution is finished the use can use the Preview Control Iteration scroll bar to move through and examine the results of the 2D process at each iteration 4 You should set the scroll bar at the iternation that contains the best results Then press the Apply All button At this point the 2D deconvolution will be performed on the entire dataset not just the current view using the selected number of iterations Notes on Spherical Aberration Detection Spherical Aberration SA Spherical Aberration occurs when peripheral rays have a different focal length than rays near the center of the lens Spherical aberration is the dominant aberration in microscopy Spherical aberration causes asymmetry in the PSF which distorts the data Detect ing spherical aberration before deconvolution will improve the accuracy of the theoretical PSF calculation The theoretical PSF can be used in both fixed deconvolution and as a first guess to blind deconvolution Improving the accuracy of the theoretical PSF calculation will in turn improve deconvolution results This control is available on both the 3D Deconvolution dialog and the Inverse Filter dialog The control is hidden by default and can be made visible by pressing the Spherical Aberration drop down button in the PSF Settings group of the 3D Deconvolution dialog or the Input PSF gro
207. xperiment setup Otherwise one should avoid this method because it is more prone to errors and misjudgment For example Table 1 Figure a shows a common mistake which is quite often made This occurs when the sample is brought well out of focus but an obscure out of focus rem nant of the sample is still present Avoid this situation It is extremely important to have a flatfield image that represents a totally flat field as shown in Table 1 b When using fluorescence the most reliable way to prepare a flat field sample is to place a drop of dye onto a slide and to seal it January 2012 Page 30 Media Cybernetics Inc Manual amp Tutorials with a coverslip Although in principle it may be possible to take the lens out of focus with the sample in place for a flat field we do not recommend doing so with fluorescence Doing so is prone to problems With fluorescence it is much more difficult and prone to errors to judge that the field is completely flat Also it is difficult and often impossible to collect an out of focus flat field picture that has sufficient signal level because the signal level decreases as the sample is taken out of focus This signal level problem may be compensated for by increasing your video gain or by increasing your exposure time doing so is prone to errors and is best avoided To summarize For a flat field brightfield image remove the sample For a fluorescence flat field image replace the sampl
208. y open unsaved datasets you will be prompted to save them January 2012 Page 64 Media Cybernetics Inc Manual amp Tutorials Recent Files List This feature displays five of the most recently loaded and closed datasets This list is found in the File menu underneath Exit To open one of these most recent files select the file from the list by clicking on it Please close all views before going on to the next tutorial AutoQuant Connect AutoQuant Connect is a new feature that allows you to automatically import and export your data to partnering software applications for additional analysis or visualization Using AutoQuant Connect you can import a CONNECT Image Set acquired in partner software applications such as Image Pro Plus or other software partners The set includes the acquired image set the image set meta data complete deconvolution settings instructions and completion commands Upon receiving the CONNECT Image Set AutoQuant will queue all subsequent sets into the batch processing viewer to launch the next task Each set is deconvolved using the defined set tings with AutoQuant s industry leading algorithms Once the deconvolution task is complete the image set along with complete meta data is returned to the location of your choice It can be returned to the connected software partner application saved to a custom folder or displayed in AutoQuant for review visualization or analysis Export to Image Pro
209. you may find it neces sary to bend these rules These are recommended conditions and not hard rules You should go ahead and bend them if necessary Because of the robustness of AutoQuant your deblurred results may still turn out fine but this will depend on many factors including your sample type and is less predictable than if you follow the rules To emphasize keep in mind the following adage garbage in garbage out AutoQuant is designed to be robust against breakage of these guidelines Breakage of say one or two of them may cause no problem However with each addi tional guideline that is broken your chances of achieving satisfactory experimental results January 2012 Page 35 Media Cybernetics Inc Manual amp Tutorials decreases As a general rule do not break any of the guidelines for sake of convenience Break only those guidelines for which your experimental design restrictions absolutely prevent you from following these guidelines In other words only break a guideline if you have no choice but to do so Sampling Issues As a rule of thumb the in plane sampling ought to be equal to the resolution element resel size or spot size of the image According to Webb et al _ _ 0 42 Ad lresel WA xy and Ad Lahn NA where Ad and Ad are the spot sizes in micrometers for the in plane XY and axial z dimen sions respectively and where and NA are the wavelength and numerical aperture respectively
210. yzing datasets from a confocal optical microscope so long as that confocal microscope is licensed in conjunction with U S Patent No Re 34 214 Most confocal microscope manufacturers have such a license It is the responsibility of the customer to determine if their confocal microscope is licensed in conjunction with U S Patent No Re 34 214 which may be done by enquiring with their confocal microscope manufacturer This software may not be used with confocal microscopes that do not have such a license Media Cybernetics will not be respon sible for any infringement nor appearance of such infringement This restriction does not in any way imply any acknowledgment by Media of the validity of U S Patent No Re 34 214 4 OWNERSHIP OF SOFTWARE AND MEDIA You agree and acknowledge that Media transfers no ownership interest in the SOFTWARE or in any SOFTWARE copy to you under this Agreement or otherwise and that Media and its licensors reserve all rights not expressly granted to you hereunder After you pay the applicable license fee and or purchase price for the SOFT WARE you will own the media on which the SOFTWARE was originally provided to you here under and on which you subsequently copy the SOFTWARE duly exercising the restrictions noted in Section 2 above Media and its licensors shall retain ownership of all SOFTWARE and copies of the SOFTWARE or portions thereof embodied in or on such media including owner January 2012 Page 9 Medi

Download Pdf Manuals

Related Search

Related Contents

APC Smart 750VA LCD 230V  AFG3000 Series Quick Start User Manual  Samsung Washer/Dryer 10kg Wash/7kg Dry (WD1102XVM) User Manual  fongicide Switch vs oïdium ds concombre de serre  Manual de instalación  Audiovox AXWD2002 User's Manual  Práctica de captura - Wireshark - Universidad Carlos III de Madrid      

Copyright © All rights reserved. Failed to retrieve file