I agree. At some point the high resolution actually makes things much harder. For example, we are treating cortical gray matter as a single class, but at high enough resolution you start to see relatively large differences in signal intensity between deep and superficial gray, particularly where there are heavy intra-cortical myelin bands (e.g. the stria of Gennari). In general I think the better strategy is to get 1mm isotropic volumes for surface recon, then get higher res and register them for more local analysis (we have some alpha tools for this latter step).

cheers, Bruce

On Fri, 27 Mar 2009, Nick Schmansky wrote:

Rosa,

The recon-all.log file indicates that it failed the tailarich_avi step because your system does not have libg2c.dylib. Can you update to a newer version of freesurfer?

You will still have problems getting your hi-res data to be processed by the recon-all stream. I am sorting out those problems now. Basically, any step that involves an atlas or working with 'conformed' volumes (1mm^3, 256^3) needs to be specially handled.

I should point out that little is gained by trying to get surfaces using 0.8mm data versus just letting the stream conform the data and process normally.

Nick

On Wed, 2009-03-25 at 18:57 +0100, Rosa Steimke wrote:

...

Hi, Im trying to keep images at a resolution of 0.8 mm and used the ÿÿcm flag during the convertion. However i got an error message during talairach transformation in autorecon1. The logfile is attached. Do I need to use a flag here, too? I learned form the previous email that I have to also use the ÿÿnoconform flag during mri-normalization and the ÿÿnoaseg flag during autorecon2. Are there more things I will need to do? Kind regards. Rosa

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer