Hello Dear FSL Experts,
We are trying to use FSL to perform probabilistic tractography on user-defined seedpoints (6 segments of the corpus callosum saved in Analyze format and then converted to nifti_gz for use in FSL). We have tried using our own DTI maps as well as DTIFIT outputs for the input of Bedpost.
Because Bedpost runs without showing any error, we use the Bedpost output directory as the input of ProbTrackx. In ProbTrackx, we upload our seedpoint (nifti_gz) as a single mask and use the following settings: number of samples = 5000, curvature threshold = 0.2, step length of 0.4, fiber volume fraction threshold of 0.0, and max number of steps = 2000. While there is no error when running Probtrackx, when we open the data file (nifti) and fdt_path file of the seedpoint (output of Probtracks) in FSLview, no object or fibers are visible (i.e. max and min values are both 0.0) even though the FA map is clear and the corpus callosum is visible. We use display as RGB, modulation as dti-FA, and view seedpoint as red-yellow in FSLView.
What steps do you recommend for us to take in order for FSL to recognize the fibers present in our ROI?
Thank you,
|