Automatic, Online Reconstruction of Advanced Single Voxel and Spectroscopic Imaging Sequences

Motivation: MRS and MRSI provide valuable insight into cellular metabolism, however conventional offline data processing and analysis presents a significant practical barrier to widespread adoption. Goal(s): We propose a complete inline MRS and MRSI processing and analysis workflow, using Siemens FIRE to run advanced reconstruction on SVS and non-cartesian MRSI sequences. Approach: We adapted our FID-A and LCModel pipelines for MEGA-PRESS, SPECIAL, and Rosette MRSI to be used on the scanner. Processed and quantitative data are converted to DICOM format. Results: Processed spectra and LCModel results are viewable on the scanner. All results are saved and can be brought offline for future use. Impact: We demonstrate the benefit of automatic reconstruction and analysis pipelines for spectroscopy, integrated on-scanner with FIRE. The proposed framework benefits collaborative work as complete containerized environments for processing and analysis can be shared between research sites.