search

CBICA / MUSE

Other
2 stars 4 forks source link

ERROR: Deformable Registration failed!!! #4

Open CH2-Carbene opened 2 years ago

CH2-Carbene commented 2 years ago

Hi experts，

We tried to install the MUSE in our docker. However, there's no result in target dictionary. Can you give me some advice to figure out this problem?

Best

The output of muse command shows that: Screenshot from 2022-10-24 13-20-43

When checking --------> RUNNING REGISTRATION WITH TEMPLATE13 at Sun Oct 23 05:31:14 UTC 2022 LOGFILE: /opt/sbia/dramms-1.4.1/muse_log_files/muse-dramms-registrations-Template13-23877.log

The log shows that:

Command /opt/sbia/dramms-1.4.1/lib/Deform3D -b0,0,0 -p -r3 -C0 -n5 -k10 -s0.50 -m2 -f1 -M2 -w1 -g.04444444444444444444 -e0 -F0 -S0 -u2 -a/tmp/dramms-wbzXY5/features/B_level1_mask.nii.gz /tmp/dramms-wbzXY5/features/A2B_affine_level1.nii.gz /tmp/dramms-wbzXY5/features/B_level1.nii.gz /tmp/dramms-wbzXY5/features/GaborUsed_ A2B.nii.gz DField.nii.gz failed

ERROR: Deformable Registration failed!!!
:o:o:o:o:o Aborting Operations ....

Screenshot from 2022-10-24 13-23-35

Nat-Sci commented 2 years ago

Same problem, emm ...

190865 commented 1 year ago

hey bro，I have a question, can you help me ？In the Muse_Software_Manual.pdf, Where is the muse-3.0.0-source.tar.gz in the configuration？

190865 commented 1 year ago

replace？

190865 commented 1 year ago

DeepMRSeg is better？

gurayerus commented 1 year ago

Hi, looking at the error message, this seems like due to one of the registrations failing. It's difficult to track this further. One very fast solution would be to use DLMUSE, our deep-learning-based registration that runs in seconds. DLMUSE is trained on a large set of curated MUSE labels and was validated on large samples. Please let me know if you would want to try it. We also have an AWS portal to run your data on the cloud. This may be used directly without installation

190865 commented 1 year ago

Hi, I want to try!!!!thank U!!!!!!!!

gurayerus commented 1 year ago

I will send you the link for the portal and the container tomorrow.

gurayerus commented 1 year ago

Thank you for your interest!

190865 commented 1 year ago

I receive your reply, I am particularly happy, and I am very happy that you and I recommend DLMUSE. I have another question. Is this DLMUSE a modification based on the basis of MUSE?

gurayerus commented 1 year ago

It's a DL model that was trained on labels generated using regular MUSE. It provides same set of ROIs. The paper with all details will be hopefully out soon.

190865 commented 1 year ago

Hi, are you free now?

190865 commented 1 year ago

I will send you the link for the portal and the container tomorrow.

Hi ，Are you free now？

gurayerus commented 1 year ago

Sorry, I was late to reply. Can you please use the following link: https://github.com/CBICA/niCHARTPipelines This will provide you a complete application for segmenting ROIs from a raw T1 (it includes DL based ICV segmentation + DLMUSE)

Best Regards, Guray

On Mon, Oct 23, 2023 at 8:58 AM haiou Li @.***> wrote:

I will send you the link for the portal and the container tomorrow.

Hi ，Are you free now？

— Reply to this email directly, view it on GitHub https://github.com/CBICA/MUSE/issues/4#issuecomment-1775137997, or unsubscribe https://github.com/notifications/unsubscribe-auth/AM4JINQMQ36OMCY2U6IJX7LYAZSZBAVCNFSM6AAAAAARMUIGTKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZVGEZTOOJZG4 . You are receiving this because you were assigned.Message ID: @.***>

-- Guray Erus, PhD

Director of Research Artificial Intelligence in Biomedical Imaging Lab University of Pennsylvania Richards Building, 7th Floor 3700 Hamilton Walk Philadelphia, PA 19104

190865 commented 1 year ago

I can use the link!!! I will try~Thank U~