Issues with combining subspectra

[Daniellalouisejones]

Hello,

I am a PhD student using the FID-A MEGA-PRESS processing pipeline to analyse MRS data collected from 4x brain regions on a 3T Siemens MR system.

I have been using your run_megapressproc_auto.m script to acquire these spectra, which has successfully worked for 3 out of 4 regions. However, running the script on the final region produces the following error:

I tried to run 'op_fourStepCombine.m' prior to 'op_combinesubspecs.m', unfortunately with no success. I then ran the non-automated MEGA-PRESS script and the same error occurred after the SpecTool GUI was used.

Please can you advise me on how I might be able to overcome this problem? I'd really appreciate your input.

Kind regards,

Daniella

[jamienear]

Dear Daniella, The error indicates that your dataset has four sub-spectra. This is a rare circumstance; it might be the case if you were using an ISIS localised spectral editing sequence (MEGA-SPECIAL) or a Hadamard encoded sequence (HERMES or HD-SPECIAL). But in the case of MEGA_PRESS data, you should only have two sub-spectra (edit-on and edit-off). It’s also unusual for the script to work fine for some voxels and then fail for others (assuming they were all acquired with the same sequence). Maybe something went wrong on the acquisition side? A few things you can try: 1. Load each dataset using out=io_loadspec_twix(’filename’);, and then look at the dimensions of the data arrays (out.fids and out.specs), and the labels on each dimension (out.dims). Do this for each of the four voxels to see if the one that failed is different to the others. 2. If possible, you could send me your data files and I could have a look at it. Hope this helps. Jamie On Sep 26, 2022, at 8:22 AM, Daniella Louise Jones ***@***.******@***.***>> wrote: Hello, I am a PhD student using the FID-A MEGA-PRESS processing pipeline to analyse MRS data collected from 4x brain regions on a 3T Siemens MR system. I have been using your run_megapressproc_auto.m script to acquire these spectra, which has successfully worked for 3 out of 4 regions. However, running the script on the final region produces the following error: [Screenshot 2022-09-26 at 12 53 51]<https://user-images.githubusercontent.com/56583587/192273016-492d2e15-7ded-41c5-b4c9-1d0a6296ca69.png> I tried to run 'op_fourStepCombine.m' prior to 'op_combinesubspecs.m', unfortunately with no success. I then ran the non-automated MEGA-PRESS script and the same error occurred after the SpecTool GUI is used. Please can you advise me on how I might be able to overcome this problem? I'd really appreciate your input. Kind regards, Daniella — Reply to this email directly, view it on GitHub<#100>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACHTOQNGVRS7GBBCYZR5A5DWAGIQRANCNFSM6AAAAAAQVXXYOY>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[Daniellalouisejones]

Hello,

Thank you for getting back to me, and for your suggestions, Jamie.

To clarify, the first three voxels were scanned using a jn_svs_special sequence, and the fourth using a MEGA_SPECIAL sequence. I initially ran 'run_specialproc_auto' on all four voxel regions, however, the same error occurred just for that final voxel. As a result, I tried to see what would happen if I used 'run_megapressproc_auto.m' instead, and seemed to get a little further before the error showed.

As per your initial suggestion, I have loaded and compared all four voxels and can see that 'out.dims' are the same for all, however 'out.fids' and 'out.specs' are different in the last voxel, compared to the three before.

If it is ok with you, can I share the data with you? It's too large to upload here - my email is d.l.jones@surrey.ac.uk if required.

All the best,

Daniela

[jamienear]

Thanks Daniela (sorry I spelled your name wrong before). The result makes sense, given you are using MEGA_SPECIAL for the fourth voxel. The processing script ‘run_megapressproc_auto’ will not work on MEGA-SPECIAL data… only MEGA-PRESS data. The ‘run_specialproc_auto' function is the correct function for the jn_svs_special acquisitions. Unfortunately, FID-A does not include a processing script for MEGA-SPECIAL data, but I do have one locally that I can share with you. I will contact you offline at the email you provided below. Best wishes, Jamie On Sep 26, 2022, at 10:29 AM, Daniella Louise Jones ***@***.******@***.***>> wrote: Hello, Thank you for getting back to me, and for your suggestions, Jamie. To clarify, the first three voxels were scanned using a jn_svs_special sequence, and the fourth using a MEGA_SPECIAL sequence. I initially ran 'run_specialproc_auto' on all four voxel regions, however, the same error occurred just for that final voxel. As a result, I tried to see what would happen if I used 'run_megapressproc_auto.m' instead, and seemed to get a little further before the error showed. As per your initial suggestion, I have loaded and compared all four voxels and can see that 'out.dims' are the same for all, however 'out.fids' and 'out.specs' are different in the last voxel, compared to the three before. If it is ok with you, can I share the data with you? It's too large to upload here - my email is ***@***.******@***.***> if required. All the best,https://docs.github.com/github/writing-on-github/getting-started-with-writing-and-formatting-on-github/basic-writing-and-formatting-syntax Daniela — Reply to this email directly, view it on GitHub<#100 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACHTOQJFOTCHHZWY7F2MD53WAGXLLANCNFSM6AAAAAAQVXXYOY>. You are receiving this because you commented.Message ID: ***@***.***>