Visual Display Interface (VDI): A MATLAB Software Library For Simulating and Processing In-Vivo Magnetic Resonance Spectroscopy and Spectroscopic Imaging Data

We describe a new comprehensive and open-source MATLAB library, called the Visual Display Interface (VDI), designed to process, model, quantify and visualize magnetic resonance spectroscopy (MRS) and spectroscopic imaging (MRSI) data. The library focuses on three major strengths: promoting reproducible research by creating comprehensive data logging and reporting and identifying outlier datasets; seamlessly combining spectral and spatial processing of both single and multivoxel data; and offering a modern, object-oriented design. VDI handles a wide range of common tasks, including spectral and spatial transforms, built-in spectral fitting, absolute quantification, and running density-matrix simulations for coupled and uncoupled spin systems to generate appropriate basis sets and aid in sequence design. VDI interfaces with the Statistical Parameteric Mapping (SPM) toolbox to carry out tissue segmentation, calculate tissue fractions within voxels, derive mean metabolite values from regions of interest defined by anatomical or functional atlases, and perform linear regression for global white matter and gray matter metabolite concentrations. The librarys workings are demonstrated for two tasks: (1) Pre-processing, fitting and analysis of single-voxel proton MRS data from healthy volunteers; and (2) Extracting region-specific metabolite concentrations from spectroscopic imaging data based on an existing cortical atlas in MNI space, and calculating average gray and white matter global concentrations using linear regression.