Probe Repeatability and Comparison with Target Design - BUNPC/AtlasViewer GitHub Wiki

After importing individual probe geometries into AtlasViewer, the user could examine the experimental repeatability of probe position across subjects.

(1) First prepare a group-level directory folder you would like to analyze, which should contain a subfolder for each subject and each subject folder should have the digpt.txt file created for that subject. The original SD file should also be contained in the root group-level directory folder.
In this example as shown below, we have a ‘GroupA’ folder that contains the Probe.SD file, S1 subfolder, and S2 subfolder. Each subfolder has a digpt.txt file containing 3D coordinates of all sources, all detectors, and five landmarks.

(2) Launch AtlasViewerGUI and go to the \GroupA\S1 directory.

(3) Click “Tool --> Register Atlas to Dig Points,” then click “Register Probe to Surface.”



After that, save work for this subject using “File --> Save Viewer State". Then the atlasViewer.mat file will be generated and stored in the \GroupA\S1 folder.

(4) Change to the \GroupA\S2directory using “File --> Change Subject” and Repeat the step 3 for S2. Then the atlasViewer.mat file for S2 will be generated and stored in the \GroupA\S2 folder.

(5) Change the current directory to GroupA where the design probe (.SD file) is stored using “File-->Change Subject.” A message will pop up asking ‘AtlasViewer has detected dig points in the current subject’s sub-folders. Do you want to load the mean of the group dig points?’ Click NO.

(6) Import the Probe.SD file into AtlasViewer using “Tools--> Import Probe,” then click “Register Probe to Surface.”

(7) Click “Tools --> Probe Placement Variation.” You’ll be asked to select a directory. Select the current directory, i.e., /GroupA in this example.

(8) A dialog will pop up and ask the user to make one of two selections: Inter-subject variability OR Probe fabrication error.



Inter-subject variation plots the optode locations for multiple subjects and generates the mean and standard deviation for each optode position. It displays these stats using fixed-size color circles and a table of numbers. This is useful for determining how consistently the SD channels are positioned over the targeted cortical areas.

Click the “Inter-subject variation” option then two figures and one table will display, as shown below.

Figure 1 shows the digitized optode positions, color-coded by subject.



Figure 2 represents the standard deviation in millimeters, plotted at the mean of the optode positions.



The table shows the mean of the optode positions and can be copied to the operating system’s clipboard.

(9) Click “Tools --> Probe Placement Variation” again and select the “Probe fabrication error” option. NOTE that you could do this step if you are in the same AtlasViewer session. Otherwise if starting a new AtlasViewer session, you need to repeat the step 6 (i.e., import SD file and register probe to surface) before this step.

The Probe fabrication error option will run an analysis of how closely the digitized optode locations from individual subjects match the original probe design. This is helpful for evaluating deviations from the original probe design and detecting both single-point and systematic errors in probe fabrication.

Two figures and one table will display after clicking “Probe fabrication error”, as shown below. Figure 1 shows digitized probe optode positions color-coded by subject. Black dots are the locations of the optode locations from the registered probe design.



Figure 2 represents the mean geometric error from the original SD locations. Original optode locations are plotted in black while the digitized optodes are represented as ellipses plotted at the mean optode locations. The size of the ellipse radii is determined by the standard deviation across subjects, along the x and y axes. The color represents the mean-geometric error that is the distance between the mean digitized optode location and the design optode location.



The table gives 3D coordinates of the original design and mean digitized locations as well as the mean geometric error.