Thanks, Marco.

The volume of the LGN in our atlas is ~215mm3. Volume discrepancies with other atlases or manual tracing are common, particularly for such small nuclei with faint contrast (and also due to discrepancies in criteria eg to determine the posterior / medial boundary with the rest of the thalamus).

Now, for this particular subject, the right thalamus is considerably oversegmented. In fact, it’s one of the worst failure I’ve seen (if not the worst!). Does this over segmentation happen in other subjects in your dataset? If so, we definitely need to investigate the issue.

Cheers,

/Eugenio

Juan Eugenio Iglesias Senior research fellow CMIC (UCL), MGH (HMS) and CSAIL (MIT) http://www.jeiglesias.com

On Sep 16, 2021, at 16:48, Marco Aiello <marcoaiello1978@gmail.commailto:marcoaiello1978@gmail.com> wrote:

External Email - Use Caution

Dear Eugenio,

Thank you.

I've uploaded via FTP a zip file (g1120_5.zip) containing a recon-all folder of a subject, including thalamic nuclei segmentation.

Best,

Marco

Il giorno lun 30 ago 2021 alle ore 10:17 Marco Aiello <marcoaiello1978@gmail.commailto:marcoaiello1978@gmail.com> ha scritto: Dear Freesurfer experts,

I'm running (FS 7.1.1) segmentThalamicNuclei.sh on T1 scan from recon-all. On a sample of about 100 subjects I've found an average volume of 250 mm^3 and 290 mm^3 for LGN-left and LGN-right, respectively. Consistent with literature, I would expect an LGN volume around the order of 100 mm^3. Do you have any suggestions to address this issue?

Best,

Marco

An example of left thalamic nuclei stats on a subject:

# Left Thalamic nuclei volume statistics as created by segmentThalamicNuclei.sh 1 1 0 179.160119 AV 2 2 0 89.639229 CeM 3 3 0 40.504433 CL 4 4 0 298.704524 CM 5 5 0 29.204221 LD 6 6 0 323.610060 LGN 7 7 0 172.804240 LP 8 8 0 16.699212 L-Sg 9 9 0 370.292762 MDl 10 10 0 1028.359432 MDm 11 11 0 99.816768 MGN 12 12 0 16.307111 MV(Re) 13 13 0 4.583480 Pc 14 14 0 46.055844 Pf 15 15 0 8.414055 Pt 16 16 0 274.769665 PuA 17 17 0 276.059756 PuI 18 18 0 197.470981 PuL 19 19 0 1329.257389 PuM 20 20 0 540.769724 VA 21 21 0 48.461021 VAmc 22 22 0 879.500725 VLa 23 23 0 1143.223136 VLp 24 24 0 33.606403 VM 25 25 0 1224.126516 VPL 26 26 0 8671.400807 Whole_thalamus

-- ------------------------- Marco Aiello, PhD Researcher IRCCS SDN | Synlab Via Gianturco 113 / Via Brin 55, Naples, IT phone (0039) 0812408299 marco.aiello@synlab.itmailto:marco.aiello@synlab.it _______________ _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer