 |
 |
> I am currently working on a VBR (voxel-based-relaxometry) study, looking > for regions with significant differences between patient and control > groups. I am also interested in looking at global gray and white matter > T2 values, however I am not sure how to go about doing this. Any > suggestions would be greatly appreciated. Would the images have to be > segmented into gray and white matter and then summed? For the average T2 value in WM, you would sum up the T2 values in the white matter voxels and divide by the number of voxels. For a weighted average, you would sum up the voxel-by-voxel product of the T2 image, with the white matter probability map, and divide this by the sum of the values in the WM probability map. For region-wise T2 values in WM, then you could try it this way: 1) ImCalc, selecting T2 and WM. Evaluate the expression i1.*i2 2) Smooth the resulting image by x mm, and the WM image by x mm. 3) ImCalc, selecting the smoothed T2.*WM and the smoothed WM, evaluating the following expression: i1./(i2 + 1/255) Note that the 1/255 is just to prevent numerical problems in regions that hardly contain any WM. I don't actually know what value would be best to use. Best regards, -John