 |
 |
Is there a straightforward way to create hard segmentations like those in fast4's _seg file output from the _pve files? I tried to generate one using a max value approach (see below), but it looks slightly different from the original _seg. Do I need to apply some kind of smoothing to get exactly the same result? Thanks. # T1 weighted image named a01 # a01_seg and a01_pve_? generated with fast4 # Generate a file containing the maximum of three pve values per voxel fslmaths a01_pve_1 -max a01_pve_2 -max a01_pve_3 max # Extract from each individual pve those voxels that match the maximum # and set them to a value reflecting the tissue class fslmaths max -sub a01_pve_1 -abs -bin -sub 1 -abs t1 fslmaths max -sub a01_pve_2 -abs -bin -sub 1 -abs -mul 2 t2 fslmaths max -sub a01_pve_3 -abs -bin -sub 1 -abs -mul 3 t3 # Combine the classes for comparison with a01_seg fslmaths t1 -add t2 -add t3 -uthr 3 tseg # Histogram comparison fslstats tseg -h 4 4088262.000000 249205.000000 721571.000000 516642.000000 fslstats a01_seg -h 4 4088262.000000 282874.000000 707927.000000 496617.000000 # Difference image -- screenshot uploaded to # http://www.soundray.org/pve-seg-screenshot.png fslmaths a01_seg -sub tseg diff -- Rolf A Heckemann, PhD Research Associate Division of Neuroscience and Mental Health MRC Clinical Sciences Centre Imperial College London Hammersmith Hospital Campus Du Cane Road London W12 0HS United Kingdom 1213617743