HERMES model fit error

MRS Software Osprey
1

Hello all,

I have been attempting to analyse. Some HERMES data in Osprey recently and have hit an error in the model data step.

It seems that osp_fitInitialise is unable to locate/load the HERMES basis set.

See error below:

Error using osp_fitInitialise (line 110)

There is no appropriate basis set to model your data. Please supply a sufficient basis set in

Osprey .mat format in the fit/basissets/user/BASIS_MM.mat file! Or supply a .BASIS file for

LCModel

Error in OspreyFit (line 51)

[MRSCont] = osp_fitInitialise(MRSCont);

Error in osp_onFit (line 33)

MRSCont = OspreyFit(MRSCont);

Error while evaluating UIControl Callback.

Our data is Philips 3T SDAT, having a quick look in ‘osprey/fit/basissets/3T/Philips/Hermes/gabagsh’ it seems the mat file here is basis_siemens_hermes.mat

I am not sure if this is causing the issue. See below for the options in our job file incase there is an error at this level:

%%% 1. SPECIFY SEQUENCE INFORMATION %%%

seqType = ‘HERMES’ ;

editTarget = { ‘GABA’ , ‘GSH’ };

%%% 2. SPECIFY DATA HANDLING AND MODELING OPTIONS %%%

opts.saveLCM = 1;

opts.savejMRUI = 1;

opts.saveVendor = 1;

opts.fit.method = ‘Osprey’ ;

opts.fit.includeMetabs = { ‘default’ };

opts.fit.style = ‘Separate’ ;

opts.fit.range = [0.2 4.2];

opts.fit.rangeWater = [2.0 7.4];

opts.fit.bLineKnotSpace = 0.4;

opts.fit.fitMM = 1;

%%% 3. SPECIFY MRS DATA AND STRUCTURAL IMAGING FILES %%%

files = { ‘/Volumes/Extreme2021/Hypoxia_2021/BIDS_hypoxia_2021/sub-01/ses-01/mrs/sub-01-hermes_ses-01_acq-hermes_vox-pcc.SDAT’ };

files_ref = { ‘/Volumes/Extreme2021/Hypoxia_2021/BIDS_hypoxia_2021/sub-01/ses-01/mrs/sub-01_ses-01_acq-hermes_vox-pcc_type-ref_hermes.SDAT’ }

Please let me know if you have any ideas of where the issue could be coming from?

Cheers,

Matt

2

Hi Matt,

Thanks for reaching out!

Quick question: What version are you on? Is it the current release version or the develop-branch?

One comment about the basis sets: The HERMES/HERCULES implementation uses the pulses and timing from the universal sequence which is standardized between vendors. Therefore, the basisset currently in the ‘osprey/fit/basissets/3T/Philips/Hermes/gabagsh’ folder should work for you.

Looking at your job file I can’t see any issues in there and the automated basis set picker should work. A possible easy fix for you is to add the following line to your job file to pick a certain basis set:

opts.fit.basisSetFile = '/Users/helge/Documents/GitHub/osprey/fit/basissets/3T/philips/hermes/gabagsh/basis_siemens_hermes.mat'

Replace with the path to the basis set on your computer.

For debugging purposes I would be interested in the following:

  1. Can you debug the osp_fitInitialise.m function and tell me the entry in MRSCont.opts.fit.basisSetFile after line 100?
  2. And would you be able to share the data with me, so that I can figure out at which point the wring basis set is picked?

Cheers,
Helge