Co-Registration of Siemens VOIs and Anatomicals
Analyzing magnetic resonance spectroscopy (MRS) data often requires you to register an MRS volume of interest (VOI) to a high-resolution anatomical brain image. After registration, you will be able to extract the proportion of grey/white matter in the tissue within a VOI using image segmentation tools available in imaging packages such as SPM.
Co-registration requires reading position information from the header of a spectroscopy image and applying an affine transform to rotate the VOI according to its placement. This computation is difficult to standardize, as each MR machine vendor outputs different information in these header files.
This set of scripts allows for the co-registration of spectroscopy VOIs (rda files) and anatomical images (dicoms) from a Siemens scanner. Batching across participants and VOIs, RegVOIGroup.m reads rda header files (readRDA.m), creates a mask of the registered VOI (parseVOI.m), and displays the VOI position on an anatomical image.
These scripts were developed through collaboration with Carles Falcon and Eloy Martinez-Heras at the University of Barcelona, and integrates with code from the GABA-edited MRS data analysis tool, Gannet.