MEGA-semiLASER: fsl_mrs sequence description in .json

Analysis / Modeling Basis Sets
1

Hello all,

I am new to MRS field and currently trying to fit phantom spectra recorded with MEGA-semiLASER sequence. For basis spectra simulation a sequence description as .json file is required. I wanted to ask, if anyone could share such description for Siemens 3T Prisma or advice me a guide on how to do that? Is it possible to infer it based only on existing data (which are mrs .nii files with .json sidecars)?

Best regards,
Igor

2

Hi Igor,

Sorry to keep you waiting. I’m cc’ing @wclarke in since he has authored FSL-MRS. Where did you get the MEGA-sLASER sequence from - CMRR?

Best,
Georg

3

Hi @Igor,

Sorry I missed this. Georg is right to ask what sequence this is. It is quite possible that a sensible basis set already exists, e.g. an LCModel one will work well.

Otherwise do you have an accurate description of the sequence? For this you need to know the RF timings (and optionally the pulse shapes) in the sequence. I’m afraid it’s not possible to infer directly from just the data.

Will

4

Hi, @wclarke !

Thank you for your reply!
The MEGA-sLASER comes from CMRR and is used in Siemens 3T Prisma Fit. I searched the MRS software like Osprey and MRspa, but there are no appropriate basis spectra.

I have the sequence description as .pdf output from scanner software, but I am not sure how to make pulse shapes correctly. That’s why I’m interested if an already computed basis spectra/seq.description already exists in free access in order not to reinvent the wheel.

Best regards,
Igor

5

Hi Igor,

The pulse waveforms are usually vendor-proprietary, which is why you won’t find them publicly available. If the CMRR sequence uses default Siemens pulses, we should be able to make appropriate simulations based on the sequence timing (we have the waveforms). Otherwise we might have to ask Dinesh Deelchand whether he’d be willing to help out. He sure has LCModel basis sets, but I don’t know how he simulates them.

Best,
Georg

6

Again, I agree with @admin. Contacting CMRR to see if they have suitable basis sets already available will be the easiest and potentially quickest route for you.

If not you could request a detailed description of the sequence, and then I can help you interpret it into the required format.

Else, do you have access to the Siemens IDEA simulation environment? It isn’t too much work to generate an accurate description by running the sequence in the simulator and extracting the timings.

1 Like