Dear FSL experts,
I have some questions concerning my DTI processing steps and subsequent analysis. The original idea was to apply TBSS on FA and MD images and to compare mean values between patient and control groups. However, normalization to the 1mm FA template brain did not work for most subjects because of poor image quality/disturbances based on brain tumor pathology. For some reason, normalization with SPM (to MNI 2mm) incl. lesion masking worked fine, so I skipped tbss scripts and just skeletonized each subjects normalized DTI image (with tumor masked out for patient and control), applied that skeleton to the corresponding normalized MD image and compared values between groups. As far as I can see, results look reasonable besides a poorer resolution compared to the orig. tbss analysis. However, I am not quite sure if the way I analysed my data is correct and valid or if I might have missed something here? I would be really grateful for your advice!
Cheers, Kerstin.
|