Longitudinal recon-all –localGI after error on cross recon-all –localGI and optimal smoothing
Hi Due to aberrantly high for a few of the subjects, I ran recon-all -autorecon2-wm -randomness -autorecon3 –hemi ?h before, calculating the localGI index. However, I intended to use these data in a longitudinal stream. These subject that were reprocessed using recon-all -autorecon2-wm -randomness -autorecon3 , and should I rerun the longitudinal stream for these subjects before doing I use recon-all –long id base_id –localGI? What are the optimal smoothing when using the linear mixed models?
Best regards Knut J Bjuland
Hi Knut,
Basically as you reprocessed totally your surfaces, yes it is safer to reprocess the longitudinal stream to make sure that everything is coherent.
As for the smoothing, lGI is already inherently very smoothed, so you definitely have to take a lower FWHM than for thickness. When you run qdec, it computes for you automatically the level of "smoothedness" in your data. You'll see that with thickness this level is more or less similar to the FWHM you used, whereas with lGI it definitely is higher. So you can play with the FWHM until you reach the level of smoothedness that you want. I'd say that with linear mixed model you want the same level of smoothing that you have in equivalent cross-sectional analyses (i.e. most usually 10-15mm).
Hope this helps,
Marie
On May 29, 2014, at 1:29 PM, Knut J Bjuland knutjorgen@outlook.com wrote:
Hi Due to aberrantly high for a few of the subjects, I ran recon-all -autorecon2-wm -randomness -autorecon3 –hemi ?h before, calculating the localGI index. However, I intended to use these data in a longitudinal stream. These subject that were reprocessed using recon-all -autorecon2-wm -randomness -autorecon3 , and should I rerun the longitudinal stream for these subjects before doing I use recon-all –long id base_id –localGI? What are the optimal smoothing when using the linear mixed models?
Best regards Knut J Bjuland
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Hi Marie
Thank you for your reply. I have one additional question. Is there any difference in results when using Matlab 2012a compared to Matlab 2013b for localGI? Best regards, Knut J
On 05/29/2014 11:05 PM, Marie Schaer wrote:
Hi Knut,
Basically as you reprocessed totally your surfaces, yes it is safer to reprocess the longitudinal stream to make sure that everything is coherent.
As for the smoothing, lGI is already inherently very smoothed, so you definitely have to take a lower FWHM than for thickness. When you run qdec, it computes for you automatically the level of "smoothedness" in your data. You'll see that with thickness this level is more or less similar to the FWHM you used, whereas with lGI it definitely is higher. So you can play with the FWHM until you reach the level of smoothedness that you want. I'd say that with linear mixed model you want the same level of smoothing that you have in equivalent cross-sectional analyses (i.e. most usually 10-15mm).
Hope this helps,
Marie
On May 29, 2014, at 1:29 PM, Knut J Bjuland knutjorgen@outlook.com wrote:
Hi Due to aberrantly high for a few of the subjects, I ran recon-all -autorecon2-wm -randomness -autorecon3 –hemi ?h before, calculating the localGI index. However, I intended to use these data in a longitudinal stream. These subject that were reprocessed using recon-all -autorecon2-wm -randomness -autorecon3 , and should I rerun the longitudinal stream for these subjects before doing I use recon-all –long id base_id –localGI? What are the optimal smoothing when using the linear mixed models?
Best regards Knut J Bjuland
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Knut,
No, there is no difference in the results. At some point we had an issue with Matlab 2013, which was solved in the last FreeSurfer version. This issue makes the lGI crash, but as soon as it is solved the computation gets to the end with similar results independently on the Matlab version. So, in short, no worries about Matlab versions.
Best,
Marie
On May 29, 2014, at 8:25 PM, Knut J Bjuland knutjorgen@outlook.com wrote:
Hi Marie
Thank you for your reply. I have one additional question. Is there any difference in results when using Matlab 2012a compared to Matlab 2013b for localGI? Best regards, Knut J
On 05/29/2014 11:05 PM, Marie Schaer wrote:
Hi Knut,
Basically as you reprocessed totally your surfaces, yes it is safer to reprocess the longitudinal stream to make sure that everything is coherent.
As for the smoothing, lGI is already inherently very smoothed, so you definitely have to take a lower FWHM than for thickness. When you run qdec, it computes for you automatically the level of "smoothedness" in your data. You'll see that with thickness this level is more or less similar to the FWHM you used, whereas with lGI it definitely is higher. So you can play with the FWHM until you reach the level of smoothedness that you want. I'd say that with linear mixed model you want the same level of smoothing that you have in equivalent cross-sectional analyses (i.e. most usually 10-15mm).
Hope this helps,
Marie
On May 29, 2014, at 1:29 PM, Knut J Bjuland knutjorgen@outlook.com wrote:
Hi Due to aberrantly high for a few of the subjects, I ran recon-all -autorecon2-wm -randomness -autorecon3 –hemi ?h before, calculating the localGI index. However, I intended to use these data in a longitudinal stream. These subject that were reprocessed using recon-all -autorecon2-wm -randomness -autorecon3 , and should I rerun the longitudinal stream for these subjects before doing I use recon-all –long id base_id –localGI? What are the optimal smoothing when using the linear mixed models?
Best regards Knut J Bjuland
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
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
freesurfer@nmr.mgh.harvard.edu
-
Knut J Bjuland -
Marie Schaer