External Email - Use Caution        

Hi Stefano,

Thank you for your reply and explanation.

The command I'm using to segment FLAIR is as follows:

run_samseg --input T2_FLAIR.nii.gz  --output samseg_outputs --threads 8 --lesion --lesion-mask-pattern 1 --threshold 0.3 --save-posteriors --save-probabilities

In some cases we're also including T1 gadolinium scans together with the FLAIR (co-registered) and use the following:

run_samseg --input T2_FLAIR.nii.gz T1_gad.nii.gz --output samseg_outputs --threads 8 --lesion --lesion-mask-pattern 1 1 --threshold 0.3 --save-posteriors --save-probabilities


Thanks again for your help,
Evie

On Wed, Mar 29, 2023 at 2:08 PM Cerri, Stefano,PHD <SCERRI@mgh.harvard.edu> wrote:
Hi Evie,

--save-probabilities and --save-posteriors will save two different things for lesions. The former is an initialization of lesions based on intensity and location constraints, and the latter is a probability map of seg.mgz after we also introduce lesion shape constraints (Cf Fig.4 of MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://www.sciencedirect.com/science/article/pii/S1053811920309563). 

The output of --save-probabilities should give you a very noisy lesion probability map with lots of false positives (at least this is my experience). In your case, it seems that too many *true* lesions are removed when the shape model is introduced. Would you be able to provide the command you're using to segment your T2-FLAIR scans? Which --lesion-mask-pattern values are you using for this input combination?

Stefano

Evie Neylon Tue, 28 Mar 2023 07:15:17 -0700

        External Email - Use Caution        

Dear experts,

I am running SAMSEG to segment MS lesions from T2 FLAIR scans. I wanted to
create a lesion mask in order to obtain lesion volumes and lesion counts,
so I used mri_binarize on seg.mgz to obtain all voxels with the value 99.
However, upon visualising this mask I see that there are holes in the mask
in regions that are quite clearly lesions based on the FLAIR image. I see
in seg.mgz that these voxels are instead assigned incorrectly to a
different tissue, as the lateral ventricle, for example.

I wondered if this misassignation of tissue was due to the chosen threshold
(0.3) so I added the --save-probabilties flag to look at the
probability maps too, and I see that actually the probability map
thresholded at 0.3 does not contain holes. The probability map actually
seems more accurate in general and picks up on more of the
hyperintensities. This is important for my desired outputs (i.e. accurate
lesion counts/volumes).

So I wondered if it was possible to explain a little why these holes might
be occurring in the final segmentation but not the probability map?
Would it be incorrect to derive the lesion mask from the probability map
instead?

Apologies I cannot attach the images of the data.

TIA,
Evie
_______________________________________________
Freesurfer mailing list
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 Mass General Brigham
Please note that this e-mail is not secure (encrypted).  If you do not wish to
continue communication over unencrypted e-mail, please notify the sender of
this message immediately.  Continuing to send or respond to e-mail after
receiving this message means you understand and accept this risk and wish to
continue to communicate over unencrypted e-mail.

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 Mass General Brigham Compliance HelpLine at MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://www.massgeneralbrigham.org/complianceline .