Hi SPMusers,

 

This question is for SPM and marsbar users to calculate Volume of region based on Binary Image (1-tissue, 0-nothing/outside)

 

In our study we use Binary Image (typically called Binary Lesion Mask in the field of stroke/aphasia) to use it in Marsbar to calculate spared tissue substracting marsbar AAL ROI from Binary Lesion Mask. The image we feed to the Marsbar is normalized Binary Mask from SPM8. Following process is followed to normalize the Binary Lesion Mask

1-      Normalise:Write:

a.       Parameter file= seg_sn file from sepgmentation

b.      Images to Write=Binary Lesion Mask

c.       Preserve=Preserve Concentrations

d.      Bounding box=[-90 -126 -72;90 90 108]

e.      Voxel sizes= [1 1 1]

f.        Interpolation=4^th Degree B-Spline

g.       Wrapping=No wrap

 

My question is everytime we normalize Lesion Mask the boundaries of the Mask are changing which affects the ROI volume we are calculating. Following is the snapshot. White and Red are same Binary Mask normalized at two different times with same batch. But the edges of the Mask is different. Any idea why this is happening? Any responses will be really helpful.

 

Best
Kushal