Hi Marina

there is a tutorial on fixing defects you can look at here:

https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TopologicalDefect_freev...

I find the best way to do this is to load the lh.orig.nofix and lh.inflated.nofix (or the rh), then load the file lh.defect_labels onto both. You can then see the defects that are listed in the topology correction (note that the defect_labels = topology correction labels+1). If you select the lh.orig.nofix there is a checkbox to color the edges by the overlay value, which will then let you see them both in the slice view and in the 3D surface rendering.

In your case you have a giant posterior defect, probably caused by a bunch of skull/muscle not removed by the stripping. You'll need to erase it in the wm.mgz I suspect.

Feel free to upload you data and we will confirm this if you want

cheers Bruce

On Thu, 25 Apr 2019, Marina Fernández wrote:

External Email - Use Caution

Dear freesurfer experts, We run recon-all in a specific subject and we can see in the terminal every commands until the following line: CORRECTING DEFECT 21 (vertices=36576, convex hull=4845, v0=64547)

When the procedure is in this line it takes a lot of time (days) and it doesn't advance. When we open the lh.inflated and [l/r]h.orig.nofix files we can see that this subject has several errors (see attached images) but we don't know how we can repair it to finish the initial recon-all.

Does anyone know how to solve this situation? What can we do?

We will appreciate any help. Thank you very much in advance.

Best regards,

Marina Fernández Álvarez, PhD-student.

Laboratory of Functional Neuroscience

Pablo de Olavide University