search

edickie / ciftify

The tools of the Human Connectome Project (HCP) adapted for working with non-HCP datasets
https://edickie.github.io/ciftify/
MIT License
112 stars 158 forks source link

overlays of the functional volume and the pial surface error #122

Open liuhas2004 opened 5 years ago

liuhas2004 commented 5 years ago

I'm trying to run cifti_vis_fmri with my own data, but there was something wrong with overlays of the functional volume and the pial surface, how can i locate the problems?

liuhas2004 commented 5 years ago

PialRegCor

liuhas2004 commented 5 years ago

PialRegSag

edickie commented 5 years ago

That looks like a major realignment failure during the pipeline run (i.e. it's not cifti_vis_fmri, your preprocessed outputs are actually that bad - and the functional data was not sampled to the correct surface locations).

I had a problem like this with one dataset, where the anatomical image had a high resolution and (after trying for weeks) the only way I could fix the problem was by rerunning fmriprep with the --no-submm-recon flag. (this could also be done by rerunning fmriprep_ciftify with --fmriprep-args="--no-submm-recon")

you can visualize the problem interactively by opening the files in wb_view

wb_view sub-<subid>/MNINonLinear/fsaverage_LR32k/sub-<subid>.32k_fs_LR.wb.spec

(then inside the viewer, open the fail fmri dtseries.

uwisenjeado295 commented 4 years ago

I have a problem in compiling newmesh, MSM,.... The error is default.mk, not found! Anyone who can help me?