Oryx-MRSI

Oryx-MRSI is a fully automated and complementary software for a comprehensive multi-slice proton magnetic resonance spectroscopic imaging (1H-MRSI) data analysis. It includes multi-slice MRSI raw data and LCModel .coord file output visualizations, tissue fraction calculation, chemical shift correction, metabolite maps generation, registration onto MNI152 brain atlas, and atlas-based ROI analysis.

Features

• Main Page The user needs to provide parameters for the multi-slice 1H-MRSI data. Required parameters:

a. Exclusion criteria for fCSF, SNR, FWHM, CRLB

b. RF bandwidhth of the system for chemical shift correction

c. Cut-off value for the probabilistic binary map after registration

d. Chemical shift correction is On or Off

e. RFOV dir is RL or AP

f. Chemical shift dir (AP) is A or P

g. Chemical shift dir (LR) is L or R

h. Chemical shift dir (FH) is F or H

i. Reference metabolite

• Load Data Reads the raw 1H-MRSI data and LCModel .coord otput files for raw data and Coord file visualization of spectra.

• Co-registration Coregisters FOV,Press-Box(VOI), all voxels of spectra considering chemical shift correction if chemical shift correction is 'On'.

• Segmentation Calculation of WM, GM, CSF fractions in each voxel of all metabolites considering chemical shift correction.

• FWHM-SNR Visualization of FWHM and SNR values for all voxels.

• Spectral Quality Visualization of included voxels into the 1H-MRSI data analysis after exclusion criteria values considering FWHM, SNR, CRLB, and fCSF.

• Metabolite Map All metabolite results are used to create metabolite maps including:

  • concentration map,
  • concentration map to Ins ratio,
  • concentration map to Cr+PCr ratio,
  • CSF corrected concentration map,
  • CSF corrected concentration map to Ins ratio,
  • CSF corrected concentration map to Cr+PCr ratio.

• Registration Generates MNI152 brain atlas-Registered metabolite maps including the outputs created in the previous module.

• ROI Analysis Region of interest (ROI) analysis at multiple brain atlases like Schafer2018 100/400 Parcels on 7 resting-state (rs-fMRI) networks or MNI thr 0/25/50 brain regions.

Prerequirements

• MATLAB R2020b
• FSL for FLIRT
• SPM12
• GUI Layout Toolbox
• Widgets Toolbox
• Oryx-MRSI is tested on MAC (2.9 GHz Quad-Core Intel Core i7, 16 GB 2133 MHz LPDDR3, Radeon Pro 560 4 GB Intel HD Graphics 630 1536 MB ) and Ubuntu 18.04.4 LTS (Memory 32GIB, Processor Intel Core i7-9800X CPU @3.8GHzx16, Graphics GeForce RTX 2070/PCle/SSE2)

Installation

Oryx-MRSI uses FSL-Flirt function so using FSL from MATLAB should be ready.

If you want to install FSL into your computer, check this link

If you use MAC, check this link (Advance Usage part-Using FSL from MATLAB)

If you use LINUX, check this link (Using FSL from MATLAB)

Plase download SPM12 using this link

Download Oryx-MRSI from Github repository,

Addpath Oryx-MRSI with subfolders.

Addpath SPM12 with subfolders.

Please make sure that FSL usage from Matlab command window installation is completed properly. Before running a data analysis using Oryx-MRSI, let's check that FSL usage is from Matlab is done.

Please open matlab and run check_fsl_usage_from_matlab.m script which is given under Oryx-MRSI Github repo.

If there is no error, FSL usage from Matlab is completely installed.

If you get an error, plese check these:

If you use MAC, check this link (Advance Usage part-Using FSL from MATLAB)

If you use LINUX, check this link (Using FSL from MATLAB)

How to get started and User Documentation

Oryx-MRSI Documentation

Developers

Sevim Cengiz

Muhammed Yildirim

Abdullah Bas

Esin Ozturk Isik

Should you publish material that made use of Oryx-MRSI, please cite the following publication:

Cengiz S, Yildirim M, Bas A, Ozturk-Isik E. ORYX-MRSI: A data analysis software for multi-slice 1H-MRSI. International Society for Magnetic Resonance in Medicine. Virtual Meeting, May 15-20, 2021. (Digital Poster)

Release

• Version 1.0

Help and Support

• There isn't known any bug or issue up to now.
• If you see any bug or issue, please submit a topic in issues, or contact: sevim_cengiz@icloud.com
• If you support or contribute the code, most welcome to Oryx-MRSI Github Repository.

License

• MIT License

Acknowledgement

• This project was funded by TUBITAK 115S219. We thank all open-source MR and MRS tools.
• Oryx-MRSI uses some functions of FID-A, check this link for license.
• Oryx-MRSI uses some functions of Gannet.
  • Edden RAE, Puts NAJ, Harris AD, Barker PB, Evans CJ. Gannet: A batch-processing tool for the quantitative analysis of gamma-aminobutyric acid-edited MR spectroscopy spectra. J. Magn. Reson. Imaging 2014;40:14451452. doi: 10.1002/jmri.24478)
• Oryx-MRSI uses some functions of Osprey, check this link for license.
  • G Oeltzschner, HJ Zllner, SCN Hui, M Mikkelsen, MG Saleh, S Tapper, RAE Edden. Osprey: Open-Source Processing, Reconstruction & Estimation of Magnetic Resonance Spectroscopy Data. J Neurosci Meth 343:108827 (2020).
• Oryx-MRSI uses some functions of MRSMRIlibs, check this link for license.
• Oryx-MRSI uses some functionf of NIFTI-Matlab, check this link for license.
• Oryx-MRSI uses some functions written by Jamie Near (McGill University)
• Oryx-MRSI uses some functions written by H.Ratiney (CREATIS-LRMN)
• Oryx-MRSI uses Schaefer2018100/400Parcels7NetworksorderFSLMNI152_2mm.nii. See Github link, for license.
• Oryx-MRSI uses MNI152T12mm_brain.nii.gz, MNI-maxprob-thr0/25/50-2mm.nii.gz acquired from FSL, for license.
• If there is any function that I forget to mention about name/link/citation, please let me know.