About the usage of MSM_strain

[Git-zry]

Hi, In my thesis I want to compare MSM_strain. so I want to know how to use MSM_strain. the way I understand how to use it is as follows:

1. download the executable file https://github.com/ecr05/MSM_HOCR/releases
2. specify the config file as config/basic_configs/config_standard_MSM_strain when running the msm command Like this: msm_ubuntu_v3 --inmesh= --indata= --refdata= --conf=config_standard_MSM_strain -o
3. If I am using the HCP dataset, and only using sulc, select the configuration file config/HCP_multimodal_alignment/MSMSulcStrainFinalconf

This is my idea, is this correct please?

Also, I would like to know what the MSMAllStrainFinalconf1to1_1to3_1 and the MSMAllStrainFinalconf1to1_1to3_2 are the profiles for what application.

I hope I am not disturbing you.I'd appreciate it if you could answer my questions.

Look forward to your early reply.

[ecr05]

Hi Renato

Can you handle this.

Emma

Dr Emma C. Robinson

https://metrics-lab.github.io/

Apologies if you get emails outside normal work hours - I have small children and work flexibly. I don't expect a reply at these times.

Senior Lecturer In Medical Image Computing

School of Biomedical Engineering and Imaging Sciences

Kings College London

Office: 5th floor, Becket House, 1 Lambeth Palace Road, London SE1 7EU, UK Mail: 4th Floor, Lambeth Wing, St Thomas' Hospital, London SE1 7EH Tel. +44 (0)20 784 89557

---

From: zhang ruoyu @.> Sent: Friday, June 14, 2024 4:28 PM To: ecr05/MSM_HOCR @.> Cc: Subscribed @.***> Subject: [ecr05/MSM_HOCR] About the usage of MSM_strain (Issue #26)

Hi, In my thesis I want to compare MSM_strain. so I want to know how to use MSM_strain. the way I understand how to use it is as follows:

1. download the executable file https://github.com/ecr05/MSM_HOCR/releases
2. specify the config file as config/basic_configs/config_standard_MSM_strain when running the msm command Like this: msm_ubuntu_v3 --inmesh= --indata= --refdata= --conf=config_standard_MSM_strain -o
3. If I am using the HCP dataset, and only using sulc, select the configuration file config/HCP_multimodal_alignment/MSMSulcStrainFinalconf

This is my idea, is this correct please?

Also, I would like to know what the MSMAllStrainFinalconf1to1_1to3_1 and the MSMAllStrainFinalconf1to1_1to3_2 are the profiles for what application.

I hope I am not disturbing you.I'd appreciate it if you could answer my questions.

Look forward to your early reply.

— Reply to this email directly, view it on GitHubhttps://github.com/ecr05/MSM_HOCR/issues/26, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AD7GXPBHBO5TWZFMB4HWGL3ZHMD25AVCNFSM6AAAAABJKSCAJ2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGM2TGNRRGQ3TEOA. You are receiving this because you are subscribed to this thread.Message ID: @.***>

[Git-zry]

Sorry, I don't understand, did you send the wrong email?

[ecr05]

I forwarded it to my software engineer

Dr Emma C. Robinson

https://metrics-lab.github.io/

Apologies if you get emails outside normal work hours - I have small children and work flexibly. I don't expect a reply at these times.

Senior Lecturer In Medical Image Computing

School of Biomedical Engineering and Imaging Sciences

Kings College London

Office: 5th floor, Becket House, 1 Lambeth Palace Road, London SE1 7EU, UK Mail: 4th Floor, Lambeth Wing, St Thomas' Hospital, London SE1 7EH Tel. +44 (0)20 784 89557

---

From: zhang ruoyu @.> Sent: Friday, June 14, 2024 4:43 PM To: ecr05/MSM_HOCR @.> Cc: Emma Robinson @.>; Comment @.> Subject: Re: [ecr05/MSM_HOCR] About the usage of MSM_strain (Issue #26)

Sorry, I don't understand, did you send the wrong email?

— Reply to this email directly, view it on GitHubhttps://github.com/ecr05/MSM_HOCR/issues/26#issuecomment-2168298335, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AD7GXPFNGCNL7KXELRGSZV3ZHMFQ7AVCNFSM6AAAAABJKSCAJ2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRYGI4TQMZTGU. You are receiving this because you commented.Message ID: @.***>

[Git-zry]

Thank you for the clarification. I will wait for Renato's response. Please let me know if there is any additional information I should provide to assist with my query.

[rbesenczi]

Hi, Yes that is one possible use case. If you tell us what you are about to do, I'm sure we can help you.

The MSMAllStrainFinalconf1to1_1to3_1 files are for MSMAll in the HCP pipeline's multimodal alignment, see: https://github.com/Washington-University/HCPpipelines/tree/master/MSMAll

Please also note, that now we have a new, reworked version of MSM that runs a bit faster than the one you use. See instructions here: https://metrics-lab.github.io/newMSM/

[Git-zry]

Thanks for the reply.@rbesenczi I will describe in detail what I want to do.

I am conducting a comparative experiment on cortical surface registration (using only sulc features), where the two methods being compared are MSM_pair and MSM_strain. I am using the HCP S1200 dataset, and the goal is to register the unregistered surface files (native) in the dataset to the atlas (MSMSulc.L.refsulc.164k_fs_LR.shape.gii).

1.Data Preprocessing:

Moving Image: First, I downloaded the subjectID.L.sulc.ico6.shape.gii and subjectID.L.sphere.native.surf.gii for each subject ID from HCP. Fixed Image: MSMSulc.L.refsulc.164k_fs_LR.shape.gii and fsaverage.L_LR.spherical_std.164k_fs_LR.surf.gii

I resampled both the moving_image and fixed_image to the average 32k surface of the HCP dataset (S1200.L.sphere.32k_fs_LR.surf.gii) to obtain new shape.gii files: subjectID.L.sulc.32k.shape.gii and S1200.L.refsulc.32k_fs_LR.shape.gii.

2.Running the MSM Command: msm_ubuntu_v3 --inmesh=S1200.L.sphere.32k_fs_LR.surf.gii --indata=subjectID.L.sulc.32k.shape.gii --refdata=S1200.L.refsulc.32k_fs_LR.shape.gii --conf=MSMSulcStrainFinalconf.txt -o ~/msm_test_strain/subjectID.L.

Are there any unreasonable steps in the above procedure? I resampled the data to 32k because the FSLMSM website indicates that MSM performs well at 32k. One question that confuses me is the different versions of MSM, such as MSM_pair, MSM_strain, and MSM_HOCR. Are they simply different configuration file settings while the executable file remains the same? Can I use the newmsm mentioned in your reply to run all of them?

Look forward to your early reply.

[rbesenczi]

Hi @Git-zry,

Sorry for the delayed reply, we were super busy in the recent weeks. Your experiment sounds very interesting, and I think the steps you propose are reasonable.

If you downloaded the binary from here, the different versions are coming from using MSM with different configuration files. MSM_strain and MSM_HOCR has an better regularisation method.

You can also try newMSM, although it is still in development for a few months. Essentially it is a reimplementation of MSM with improved efficiency, negligible differences in results.

[Git-zry]

Thanks for your reply, it was very helpful!