Hi there!
I’m coming back to some work I did 3-4 years ago, and wanted to check if there’s been any major changes to pipeline or methods I should know of. I am looking at longitudinal changes in connectivity between ROI’s, comparing pre/post ablative neurosurgical procedures. The method I’m using is:
- Preprocess using mrtrix pipeline
- Dwinormalise within each subject (normalise between their preop and postop scans)
- Use Tournier method for generating a response function for each subject, and then take the mean response function for each subject (mean of pre/post)
- Generate 20 million tracts per subject/timepoint.
- Perform tcksift
- tckedit down to the tracts of interest using ROI’s on either end
Does this method still seem somewhat valid?
Thanks for your consideration!