Hi Xiangchuan, I suspect that something else is still wrong with your volume. The "max same neighbors" flag to mri_fill is an old one which causes it to apply a rank filter (like a median) to the filled volume to remove small holes and thin strands. We haven't used it since the topology correction was implemented. Could you please send us the log file of the filling with the default value for "max same neighbors"?
cheers, Bruce On Thu, 31 Oct 2002, Xiangchuan Chen wrote: > Hi Bruce and Evelina, > > I found what the problem was. In Expert Preferences Settings, WMFill, the default >value for "max same neighbors" is 8. I changed it to 1, then the surface seemed OK. >BTW, what does this value mean? > > Thanks for your advice! > Xiangchuan > > > > ======== Start Of Original Message ======== > > Date: 2002-10-30 Time: 15:57:00 > From: Bruce Fischl ([EMAIL PROTECTED]) > To: freesurfer ([EMAIL PROTECTED]) > Subject: Re: Re: Question on segmentation > > >could you send the screen output (or log file if you're using csurf) of > >what you did? This is something we do routinely so there must be something > >wrong (do you have write permission? Check the output of mri_fill and make > >sure it's not failing). > > > >Bruce > > > >On Wed, 30 Oct 2002, Xiangchuan Chen wrote: > > > >> Hi Bruce, > >> > >> I have tried the procedure just as what you mentioned in your email, but it >didn't work. I edited the wm volume, added some voxels with lower T1 value (around >85) as the part of white matter by comparing the wm and T1 volumes, saved the wm >volume, and ran the Create Surface process again. After that, I still could not find >significant improvement on the segmentation. It seemed that Create Surface didn't use >the voxels I added. Are there any ideas on it? > >> > >> Thanks, > >> Xiangchuan > >> > >> > >> ======== Start Of Original Message ======== > >> > >> Date: 2002-10-30 Time: 14:53:00 > >> From: Bruce Fischl ([EMAIL PROTECTED]) > >> To: freesurfer ([EMAIL PROTECTED]) > >> Subject: Re: Question on segmentation > >> > >> >could you be more explicit about what you did? You should be able to edit > >> >the wm volume, save it, then rerun the filling and inflation to regenerate > >> >the surfaces incorporating your changes. > >> > > >> >cheers, > >> >Bruce > >> > > >> >-------------------------------------------------------- > >> >Bruce Fischl email: [EMAIL PROTECTED] > >> >Mass. General Hosp. NMR Center. tel:(617)-726-4897 > >> >Rm. 2328, Building 149, 13th Street fax:(617)-726-7422 > >> >Charlestown, MA 02129 USA > >> > > >> > > >> > > >> >On Wed, 30 Oct 2002, Xiangchuan Chen wrote: > >> > > >> >> Hi all, > >> >> > >> >> I found that the segmentation of the white matter near the occipital pole of >my data was not quite accurate. I tried to add some voxels to the wm volume there, >but the FreeSurfer didn't take them into account. Then, how can I improve the >segmentation at this part of brain? > >> >> > >> >> Thanks, > >> >> Xiangchuan Chen > >> >> > >> >> > >> > >> ========= End Of Original Message ========= > >> > >> > > ========= End Of Original Message ========= > >
