Dear Amir,
These are excellent questions.
The module was written with ADNI in mind, so that’s good news for you!
Question 1: Things may have gotten better since I last looked into the high-res
T2s in ADNI, but back in the day, it was super-common to see fields of view
that did not completely c
is children aged 4-6 years old with cerebral palsy. The
T1-weighted scans are MPRAGE, 0.8mm isotropic.
Thanks,
Stefanie
From: Iglesias Gonzalez, Juan E.
Sent: Tuesday, July 1, 2025 8:35 PM
To: Freesurfer support list
Subject: [Freesurfer] Re: Recon-all clinical vs
Dear Stefanie,
What type of pathology are we discussing? What ages? What MRI sequence and
resolution?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Stefanie Bradley
Date: Tuesday, July 1, 2025 at 5:49 PM
To: Freesurfer support list
Subject: [Freesurfer] Recon-all cl
PS: the code is freely available (https://github.com/acasamitjana/uslr) it’s
just that it’s not integrated into FreeSurfer. Actually, if you try it, Adrià
and I would love your feedback 😊
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Iglesias Gonzalez, Juan E.
Date: Tuesday, July
Hi Marco,
We’ve built a wrapper method around it. We just got it published but haven’t
found the time to make it easily usable (also, the first author is on leave…).
But we’ll get there 😊
https://www.sciencedirect.com/science/article/pii/S1361841525002099
Cheers,
/Eugenio
--
Juan Eugenio Iglesia
: Iglesias Gonzalez, Juan E.
Date: Tuesday, 1 July 2025 at 05:51
To: Freesurfer support list
Subject: [Freesurfer] Re: eTIV in SynthSeg
EXTERNAL SENDER
The former!
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Marco Ganzetti
Date: Monday, June 30, 2025 at 05:57
To: freesurfer
The former!
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Marco Ganzetti
Date: Monday, June 30, 2025 at 05:57
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] eTIV in SynthSeg
External Email - Use Caution
Dear,
I would like to know how the eTIV is estimated in Synth
From: Iglesias Gonzalez, Juan E.
Date: Saturday, 28 June 2025 at 16:57
To: Freesurfer support list
Subject: [Freesurfer] Re: QC Synthseg
EXTERNAL SENDER
Dear Marco,
This QC scores represent the expected Dice of the segmentations. Of course,
computing the real Dice requires reference
Dear Marco,
This QC scores represent the expected Dice of the segmentations. Of course,
computing the real Dice requires reference segmentations, but this estimate is
pretty good at picking up bigger segmentation mistakes, insufficient field of
view, etc.
One way to use the QC scores is to rejec
ntrast.
Thanks again!
Best,
Lorenzo
From: Iglesias Gonzalez, Juan E.
Sent: Thursday, 26 June, 2025 3:29 PM
To: Freesurfer support list
Subject: [EXTERNE-EMAIL] [Freesurfer] Re: SynthSR - T1 3D post-gadoinium vs. T2
FSE 4mm
EXTERNE-EMAIL:
Klicken Sie nicht auf Links oder öffnen Sie keine
Dear Lorenzo,
Why not using the T1w images directly?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Diana Lorenzo
Date: Thursday, June 26, 2025 at 5:16 AM
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] SynthSR - T1 3D post-gadoinium vs. T2 FSE 4mm
E
Yeah the plan is to fix, recompile, and distribute the fix
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Toms, Robert
Date: Wednesday, May 14, 2025 at 12:33 PM
To: Iglesias Gonzalez, Juan E. , Freesurfer
support list
Subject: Re: Hippocampal Subvolumes Error -- segmentHA_T1_long
Dear Robert,
Sorry for the slow response.
Yes, it should be fine.
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Toms, Robert
Date: Monday, May 5, 2025 at 1:00 PM
To: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [F
Oooo I see. Yeah you’ll have to run in CPU mode on a machine with quite a
lot of RAM 😊
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Keenlyside, Andrew
Date: Wednesday, May 14, 2025 at 3:27 AM
To: Iglesias Gonzalez, Juan E. , Freesurfer
support list
Subject: Re: [Freesurfer
Dear Matthieu,
Which version did you use? The FireANTs version, I presume?
Can you please share some screenshots of the incorrect registration?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Chourrout, Matthieu
Dat
Thanks for catching this, Robert. It’s a small bug that does not affect the
results on the cases that didn’t crash.
Jackson: I just submitted a PR. Do we need to recompile, or does this happen
automatically?
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@
Dear María,
Sorry for the slow response; I’m currently travelling.
Would you be willing to share with us a case, that gives different results
under the 2 versions?
Either way: did you use the “FreeSurfer subject” or “Any T1 scan” mode?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesi
Dear Nima,
You can of course get better labels from a 1mm iso T1w scan. If your
registration is good, then this is the way to go.
Analyzing the diffusion data will give you less accurate labels, but eliminates
registration errors. You can try both and keep the best looking one.
Having said that:
Dear Nima,
Thanks for the interest in our work.
I don’t know if I would advise trying to segment these small subregions from
such thick slices. But if you are set on doing that, I would probably run
SynthSR first, and work with the synthetic 1mm isotropic T1s. You can run
SynthSR on the T1s and
Dear Pauline,
Did the skull stripping fail? What does norm.mgz look like?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Lieshout, P.B. van
(Pauline)
Date: Thursday, March 20, 2025 at 10:40 AM
To: freesurfer@nmr.mg
ot; Is it possible to get this information by
extracting the volumes from the CA hierarchy mgz files?
Thank you!
Erin
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Iglesias Gonzalez, Juan
E.
Date: Friday, March 14, 2025 at 5:44 PM
To: Freesurfer support list
Subject: Re: [Fr
Dear Erin,
I would recommend taking the spreadsheet with the fine-grained volumes, and
simply combine the volumes (columns) as follows:
CA1 = CA1-head + CA1-body
CA3 = CA3-head + CA3-body
CA4 = CA4-head + CA4-body
GC-ML-DG = GC-ML-DG -head + GC-ML-DG -body
Cheers,
/Eugenio
--
Juan Eugenio
Dear Migueel Ángel,
Which is the problematic subregion?
Also: would you be able / willing to share a problematic case?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Miguel Ángel Rivas
Fernández
Date: Thursday, Mar
Hi everyone,
People have asked me to do this multiple times, and I finally got down to it:
I put the subregion atlases (hippocampal subfields, nuclei of the amygdala,
brainstem subregions, thalamic nuclei) in MNI/ICBM152 space.
Details here:
https://surfer.nmr.mgh.harvard.edu/fswiki/SubfieldAtlase
Dear Miguel,
Thank you very much for the interest in our work!
Your are right about *FSvoxelSpace*.mgz; these are resampled to the voxel space
of the other 1mm isotropic volumes in the mri directory, and don’t look as
pretty as the higher resolution counterparts.
The CA/FS60/HBT versions are simp
Dear Nima,
As of today, there is no longitudinal version of recon-all-clinical. It’s a bit
tricky to adapt the standard longitudinal pipeline in FreeSurfer to scans with
thick slices due to the anisotropy of the images, and also due to the deep
learning bits of the pipeline. But you can of cours
Dear Antonio,
NextBrain provides the ICV via SynthSeg 😉
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Antonio Napolitano
Date: Wednesday, February 12, 2025 at 2:42 PM
To: freesurfer@nmr.mgh.harvard.edu
Subject: [
Dear Amir,
The scripts are certainly there:
https://github.com/freesurfer/freesurfer/tree/dev/mri_histo_util/ERC_bayesian_segmentation/scripts
Maybe it’s easier just to install the development version of FreeSurfer?
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-bo
Yeah at the moment it’s only on the development version.
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Huang, Yujing
Date: Monday, January 13, 2025 at 2:37 PM
To: Freesurfer support list
Subject: Re: [Freesurfer] Issue with mri_h
Dear Nima,
We never included the CP in SynthSeg because we didn’t manage to segment it
reliably from images of any resolution and contrast. I am CCing the first
author Benjamin Billot who can further clarify (or correct me).
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
Fr
Thanks for your message, Mark! Is this for a single case, or a more general
issues?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Mark Wagshul
Date: Monday, December 9, 2024 at 6:36 PM
To: freesurfer@nmr.mgh.har
Dear Nárlon
Thanks for the interest in our work!
You have several options:
1. If it’s only a couple of subjects, you could bite the bullet and run them
on the CPU. It’ll be slow, but hey, easy to run (just add the --cpu flag).
2. Changing the resolution at which the segmentations are genera
Hi Melissa,
What version of FreeSurfer are you using?
Kind regards,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Melissa Marie Rundle
Date: Wednesday, October 30, 2024 at 4:13 PM
To: freesurfer@nmr.mgh.harvard.edu
Subje
Dear Ian
That sounds like a RAM issue. Did you try using the crop option?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Johnson, Ian
Date: Monday, October 7, 2024 at 11:41
To: freesurfer@nmr.mgh.harvard.edu
Subje
Dear Woo-Suk,
Thanks for the interest in our work! It seems like you’re looking at the
whole-structure volumes from SynthSeg? There should be another text file with
all the other volumes.
Also: I would recommend the “simplified” parcellation; it does a better job
with the brainstem / midbrain re
Hi Jacob,
0.9 vs 1mm shouldn’t change much. If you want to use the T2, then definitely
provide it as “additional volume” eg with segmentHA_T2.sh
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Jacob Bunyamin
Date: T
Dear Annchen,
Do you get this error with a more recent stable version, e.g., 7.4.1?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Annchen Knodt
Date: Tuesday, September 3, 2024 at 1:57 PM
To: freesurfer@nmr.mgh.h
Dear Alice:
1. Can you please send another capture without the mask, or with the option
“show label boundaries only” activated? It’s hard to tell, without seeing
what’s behind 😊
2. You can read the paper for free here:
https://www.sciencedirect.com/science/article/pii/S1053811915003420
Ch
Dear Yongoo
These are different modules with different atlases; rather than worrying about
absolute differences, I would worry if the correlation was low.
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Yoonho Hwang
It’s always a good idea to divide the volumes by ICV (or add ICV as a
regressor). This is very unlikely, but what if your controls happened to have
smaller heads than your patients? For peace of mind, I’d correct by ICV.
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From:
Dear Naveen,
WMH-SynthSeg is currently only available in the development version of
FreeSurfer.
I have updated the wiki with this information.
Apologies for the inconvenience.
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Byripud
Dear Barbara,
you could in principle use recon-all-clinical for the enhanced T1 but the
surface reconstruction would probably fail.
Instead, I’d probably run SynthSR to synthesize 1mm T1 intensities while
inpainting abnormalities and then run the standard recon-all.
Cheers,
/Eugenio
--
Juan Euge
Dear FS experts,
We get a segfault when I try to resample images with the command:
mri_convert input.mgz output.mgz -rl reference.mgz
when reference.mgz has multiple frames (e.g., a RGB volume). Any idea of why
this is happening?
Cheers,
/Eugenio (&Stefano & Pranav)
___
advancing my research, and I wish to
express my sincere appreciation for your assistance.
Warm regards,
Edward
From: freesurfer-boun...@nmr.mgh.harvard.edu
On Behalf Of Iglesias Gonzalez, Juan E.
Sent: Saturday, April 6, 2024 3:39 AM
To: Freesurfer support list
Subject: Re: [Freesurfer] Easyr
Dear Sam,
The native images live in one space (typically anisotropic), whereas the
recon-all-clinical files always live in their own 1x1x1mm isotropic grid (which
is the whole point!). I am not familiar with the Enigma QC procedure but I
wouldn’t be surprised if it’s not compatible with these va
ing this objective.
Thanks in advance!
Best regards,
Edward
Sent from my iPad
> On Apr 5, 2024, at 9:57 AM, Hoffmann, Malte,PhD
> wrote:
>
> What are you trying to achieve?
>
> ____
> From: Iglesias Gonzalez, Juan E.
> Sent: Thursda
h.harvard.edu
On Behalf Of Iglesias Gonzalez, Juan E.
Sent: Wednesday, April 3, 2024 4:16 AM
To: Freesurfer support list
Subject: Re: [Freesurfer] Easyreg > cortical parcels missing after SynthSeg
with --parc option ?
Ha! I understand that sharing the images may be tricky, but do you think
esurfer/python/lib/python3.8/site-packages/numpy/linalg/linalg.py",
line 89, in _raise_linalgerror_singular
raise LinAlgError("Singular matrix")
numpy.linalg.LinAlgError: Singular matrix
From: freesurfer-boun...@nmr.mgh.harvard.edu
On Behalf Of Iglesias Gonzalez, Juan E.
Sent: Monday, April 1,
Dear Edward,
Try deleting:
/home/chang/GBM_Data/T1_template/Template_MNI_Normal122_sc_age19-40_tirq_1.0mm_T1_scalp_trip_synthseg.nii
and
/home/chang/GBM_Data/GBM_original/20231223_1/Pre/Kuo/Brain_only_synthseg/042/pre_042__synthseg.nii.gz
And then run the same command. Easy_reg will create the
Dear Katharina,
First of all, and to avoid any confusion, the bug only affected the script to
gather the volumes from different subjects (“quantifyThalamicNuclei.sh”), and
not the segmentations themselves.
Would you mind sharing some images, showing how/where the LGN is oversegmented,
please? It
Dear experts,
My collaborator Martina Bocchetta is encountering a weird problem with
recon-all, specifically with mris_remesh after the topology correction (see
below).
Any insight?
Thanks in advance,
/Eugenio
mris_remesh --remesh --iters 3 --input /path/to/subject/surf/rh.orig.premesh
--outpu
I’m not taking sides on this one, but I’ll nevertheless pinpoint that you can
also get a sbTIV using SynthSeg, hehe.
Looking forward to reading people’s opinions on this one!
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Jackson Le
Dear Joao,
I’ll defer to our software specialists, but another option may be trying the
Python version:
https://surfer.nmr.mgh.harvard.edu/fswiki/SubregionSegmentation
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of João
I’m pretty sure that the Matlab code does not reorient.
Cheers,
Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Huang, Yujing
Date: Tuesday, January 23, 2024 at 08:49
To: Freesurfer support list
Subject: Re: [Freesurfer] MR
I would totally give it a try! It runs pretty fast and gives you a segmentation
along the way.
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: Èlia Pedregosa
Date: Monday, January 22, 2024 at 8:07 AM
To: Iglesias Gonzalez, Juan E. ,
freesurfer@nmr.mgh.harvard.edu
Cc
Dear Francesca,
Our method cannot reliably segment these thin/faint structures from clinical
scans of arbitrary resolution and contrast, which is why they don’t get
segmented.
If you want to segment the choroid plexus, you could take then intensities
under the ventricles and split them into cere
Dear Hengameh,
Synthesis of T2 and FLAIR images is on the to-do list, and should happen in the
next couple of months 😉
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of hmarz...@student.ubc.ca
Date: Thursday, January
Dear Ali,
Making a SynthSeg version of the tool is on the to-do list but has not happened
yet. For now, you need to run recon-all. The good news is that you get a ton of
additional useful information on the brains!
Kind regards,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From:
FREESURFER_HOME too?
Best,
Eva
Van: freesurfer-boun...@nmr.mgh.harvard.edu
Namens Iglesias Gonzalez, Juan E.
Verzonden: vrijdag 22 december 2023 15:28
Aan: Freesurfer support list
CC: Brouwer, I. (Iman) ; Gool, J.K. (Jari)
Onderwerp: Re: [Freesurfer] SUBJECTS_DIR writing permissions to run
hippocampal
Dear Eva,
The subfield temporary directory is under SUBJECTS_DIR, not FREESURFER_HOME. So
you only need SUBJECTS_DIR to be local.
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Heese, E. van (Eva)
Date: Friday, De
Dear Chuyue,
It’s very posible that you ran out of RAM. I think you need about 32GB to run
this module. Do you think that could be the problem?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of 赵楚越
Date: Thursday, De
Dear Amrita,
Please insert your subject directory between your subject name(sub-19253779)
and your FGATIR scan (sub-19253779/mri/FGATIR.nii.gz).
If you running your code from the subjects directory, which seems to be the
case, it would simply be:
segmentThalamicNuclei.sh sub-19253779 . sub-19253
Dear Joshua,
This is a bit of an empirical question. I’d totally run both on a few subjects
and compare the results.
If you do so: I would selfishly/greatly appreciate it if you shared your
impressions with me :-)
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesur
Dear Pauline,
Thanks for the interest in our new pipeline!
You can only run one scan at the time (but that scan may be of any resolution /
contrast).
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Lieshout, P.B. van
(Pau
Dear Victor,
Thanks for letting us know. We’ll look into this ASAP.
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Zeng, Victor (BIDMC -
Keshavan - Psychiatry)
Date: Wednesday, November 8, 2023 at 00:37
To: freesurfer@n
I’ll defer to the software engineers for the first issue!
For the second: does it work if you copy norm.mgz into orig/001.mgz? :-D
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Maya Yablonski
Date: Tuesday, October 31, 2023 at 1:
Dear Maya,
You can use then -synthseg flag with recon-all to use SynthSeg instead of the
standard algorithm 😉
For fmriprep: which is the offending volume? There are very minor differences
with the standard recon-all output.
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: fre
Dear Dawn,
Integrating the hippocampal subfields into SynthSeg in on the to-do list, but
it’s tricky due to differences in resolution.
In terms of robustness, we are making progress – but I like your suggestion of
adding defacing to the list of “artifacts”! We’ll take it into consideration 😊
Che
Hi Jasmine,
This is tricky because the “absorption” happens at the discrete segmentation
level (rather than using the soft probabilities).
At that point, you’re probably better off counting voxels and multiplying by
their volume.
Doug: is there a way to do this with FreeSurfer programs, as oppose
JgDuCjLKOCEBQpCSEDa4xKQ/https%3A%2F%2Fscholar.harvard.edu%2Floggia>
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Iglesias Gonzalez, Juan
E.
Sent: Tuesday, October 3, 2023 8:52 PM
To: Freesurfer support list
Subject: Re: [Freesurfer] SynthSeg volume
External Email - Use Caution
Dera Lauren,
I am a bit confused: is /Applications/freesurfer/7.3.2/MCRv97 pointing to
/Applications/MATLAB/MATLAB_Runtime/R2019b/MCRv97?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/1X-6jqhKciXd6EPCb1LpoJ_D-PycfnYplnrld0ymBS
External Email - Use Caution
Thanks, this is super weird. When did you download this dev version? Also:
would it be possible to share the CSV file, please?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/17R5U4dEBJCcdG3QUUTbhXq98ahPm8wK3gHT_Kp2MyRltQVDu9UikD
External Email - Use Caution
Dear Kaibo,
How much RAM does your machine have?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/1f4DYOHf8Rr4YRZ9OT7geQMSYd2V5bpiQti0xpXTDLLzxMI7NosKCSWMzsvo08dZ8aEYauyNz8JGo2myKBy3e4PsLDLFEJCB0iaw8i3ezyCW698CCDeAg98tINFdWmSLWzhSE
External Email - Use Caution
Dear Julian,
I believe this is a problem with SynthSeg not being compatible with the new
microprocessor in MAC computers. Our software engineers are looking into
solutions to this.
Sorry for the inconvenience,
Eugenio
--
Juan Eugenio Iglesias
http://
External Email - Use Caution
Thanks for the interest in our work, Anton!
We don’t have a template in MNI space, sorry (it’s on the to-do list, but
buried among many other items). But you can always segment the template with
the atlas 😃
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
h
External Email - Use Caution
Dear Fatima,
You can use dcm2niix to convert Hyperfine to Nifti, and then use our FreeSurfer
tools directly on those (mri_synthsr, etc).
Kind regards,
Eugenio
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/1biIzD2_G4nFSdmA5Ydn6CcUOvXhaPhM2Ck-Z0o
External Email - Use Caution
Dear Haley,
The Matlab version of SynthSeg writes nifty files. You could run the FreeSurfer
command to convert to mgz (and rename to aseg, along the way):
mri_convert synthseg.nii.gz aseg.mgz
Now: I don’t know if it’s possible to simply replace FreeS
/1DRP30tSXMppp0KyzN8sBiVb_rbxRjFPmjNRpJ3ep6UGsmPgdFdL1PHpTB2HiZikT7mhDnrOSuMMNSyfWBC9ExkKF0npewjTGTJgEMsFLawYntgwWgMNuZX9F5v8wQsflOvhLxItA2W7jYwGI8Cq2vXFN4aiKieIWckGqFLsw8C0LJXSmsoAoKcQtCUEb549gkPIJD-OI9wJ_b9xR4nL2-29eow6o6Uz_ZrSOtwJGbb8aoGTToKo_15rmCrhx2wKl/http%3A%2F%2Fwww.jeiglesias.com
From: Eric BARDINET
Date: Thursday, August 24, 2023 at 11:53 AM
To: Iglesias Gonzalez, Juan E. , Freesurfer
support list
Subject: Re: [Freesurfer] Easyreg
-Hjz58DwOlAqvsv0QtYWZbj0obWyvYKUSBsYkOBbSGZCgVjrd6KvPVp4G9dmPqfdWQRoGAmacG4ezf1nh2KJjy9DHaSDROF3vvRKeZUeRV-MA/http%3A%2F%2Fwww.jeiglesias.com
From: Eric BARDINET
Date: Wednesday, August 23, 2023 at 9:32 AM
To: Iglesias Gonzalez, Juan E. , Freesurfer
support list
Subject: Re: [Freesurfer] Easyreg > cortical parcels missing after SynthSeg
w
External Email - Use Caution
Dear Eric,
Thanks for the interest in our tool!
Can you please send us the exact command line that you used, and the full
output of mri_easyreg?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/1U33kJUEZ__sfspXazQKQX0hyZheFOuaPY4
External Email - Use Caution
Sorry for the slow response.
1.1: the results are bound to be different (since they are different methods)
but they should be very highly correlated. For the eTIV, please take a look at
Figure 4 of
https://secure-web.cisco.com/1pF9TRM2P6hqsAScfzof5e0
External Email - Use Caution
Dear zlseu,
We currently do not have an “official” release of the probabilistic atlas in a
reference space. It is try that some of the smallest nuclei may not appear in
the hard (discrete) segmentation, but you can always use the option to write
down
External Email - Use Caution
Dear Andrei,
The script requires that you’ve run the main FreeSurfer longitudinal pipeline
(see Longitudinal Processing in the wiki).
Don’t worry about the files, they have different names and can coexist without
issues 😉
Cheers,
/E
--
Juan Eugenio
External Email - Use Caution
Uuummm I don’t think this would work, as a lot of our intermediate files (eg
norm.mgz) are not comparable to their counterparts in the standard recon-all
stream. Sorry!
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/1R9UOepVdGkzRhS_JbSZ7bkBYP0O
External Email - Use Caution
I’ll defer to our software engineering experts, but in the meantime, you could
maybe try the Python version?
https://secure-web.cisco.com/1Cv_49AxchVd_2vIYynJ4qP6kfmnE5Dy5Tw3AuuGeUNzzOEqT7zCC8a1jFBudPBWmbpay8jscwU4QvGJDaMi2Cexs8SJBYFvrJeX-LR2OR0B-B8vsf
External Email - Use Caution
Dear Nadia,
It seems like many centroids are missing. Did you visually check the
segmentations of the reference and floating image, e.g., with Freeview? Do they
look good?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/1b2Y7tMk
External Email - Use Caution
Dear Amir,
We don’t have a script to make templates, but I don’t think it’d be too hard to
do it yourself. You can start from eg MNI (not a problem to start from an
existing atlas since your first iterations will be linear anyway), and use the
usual
External Email - Use Caution
Dear Chris,
This is a pretty weird one!
Question: what files are there in the temporary directory
subjectDir/subjectBase/tmp/hippoSF_T1_long.v22_left/ ?
Cheers,
/E
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/1FmaPIPrt8VADXFs3ezSt_rWHcBh_tWsK4
External Email - Use Caution
I am with Yujing; if you copy-pasted the command from somewhere, can you please
make sure that what you have in front of the ‘s’, ‘i’, and ‘all’ are all
dashes? They look different in the capture…
--
Juan Eugenio Iglesias
http://secure-web.cisco.com/
Dear Minh,
Maybe user mri_binarize on lh.hippoSfLabels-T1.v10.FSvoxelSpace.mgz to create a
mask, and then use mri_mask to extract the hippocampus?
Cheers,
/E
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: on behalf of Hoang Gia Minh
Reply-To: Freesurfer support list
Date: Tuesday,
Dear Emily,
These boundaries are mostly geometric and based on the atlas.
For the atlas building:
* the head/body boundary was place at the level of the uncal apex.
* For the body/tail boundary, I’ll defer to Jean Augustinack (CCed)
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jei
day, March 17, 2023 at 16:49
To: "Iglesias Gonzalez, Juan E."
Cc: "freesurfer@nmr.mgh.harvard.edu"
Subject: mri_synthsr with CT images
External Email - Use Caution
Dear Juan:
I tested mri_synthsr using three different CT images with the threshold (80
value) that you tol
It totally looks like it. How much RAM do you have on the machine you’re using?
Cheers,
Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: "Fischl, Bruce R.,PHD"
Date: Thursday, March 16, 2023 at 14:04
To: Freesurfer support list , "Iglesias
Gonzalez, Juan E
Correct! Switching the orders of the inputs will switch the assignment of the
fwd and bak fields, which will otherwise be the same in both cases.
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: on behalf of Codrin Mironiuc
Reply-To: Freesurfer support list
Date: Thu
Hi again, Codrin.
mri_easyreg currently does not support part to whole registration, unless that
part covers most of the brain.
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: on behalf of Codrin Mironiuc
Reply-To: Freesurfer support list
Date: Thursday, March 16
Dear Codrin,
There was a bug incode, that caused mri_easyreg (and mri_synthseg, and
mri_synthsr) to crash if absolute paths were not used. This is now fixed in the
the development version.
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: on behalf of Codrin Mironiuc
Dear Fidel,
This is really weird… that’s essentially the last step. I you eg didn’t have
writing permissions, you’d get an error.
Would it be possible to share the data, so I can try to reproduce the issue on
my end?
Cheers,
/Eugenio
--
Juan Eugenio Iglesias
http://www.jeiglesias.com
From: on
Dear Asmin,
The discrepancies arise from differences in anatomical protocols. You can look
at the definitions in the corresponding papers and keep the one that adapts
better to your needs. The results should be fairly highly correlated.
Kind regards,
Eugenio
--
Juan Eugenio Iglesias
http://www.j
: Wednesday, March 8, 2023 at 13:56
To: Benjamin Billot
Cc: "Iglesias Gonzalez, Juan E." ,
"freesurfer@nmr.mgh.harvard.edu"
Subject: Re: SynthSeg with CT images.
External Email - Use Caution
Hi Benjamin:
I tested with the clipping, and it works ok with a CT image.
Since
http://www.jeiglesias.com
From: Gonzalo Rojas Costa
Date: Wednesday, March 8, 2023 at 12:04
To: "freesurfer@nmr.mgh.harvard.edu"
Cc: "Iglesias Gonzalez, Juan E."
Subject: SynthSeg with CT images.
External Email - Use Caution
Hi:
I tried SynthSeg using CT images
1 - 100 of 333 matches
Mail list logo
