Hello everyone,
I am seeking methodological guidance for a seed-to-atlas (thalamic nuclei) connectivity analysis using native-space MSMT-CSD FODs from HCP data. My central question is whether it is more robust to perform tractography in native space , by warping my MNI-space atlas and seed ROI to each subject’s diffusion space, or if I should first normalize all subject FODs into a common standard space . The native-space approach appears more anatomically faithful by avoiding the potential interpolation artifacts from resampling complex FOD data, a concern heightened by technical errors I’ve encountered when trying to generate a valid 4D warp field with ANTs. Therefore, I would like to ask which of these two strategies is considered the gold standard for generating the most accurate connectivity matrices for group analysis. Furthermore, if the native-space analysis is indeed recommended, what is the best practice for seeding from the warped ROI, specifically regarding the use of -seed_image versus the more precise -seed_gmwmi option to leverage the gray-matter-white-matter interface?