Real-Time Motion Correction in Magnetic Resonance Spectroscopy: AI solution inspired by fundamental science

Abstract

Magnetic Resonance Spectroscopy (MRS) is a powerful non-invasive tool for metabolic tissue analysis but is often degraded by patient motion, limiting clinical utility. The RECENTRE project (REal-time motion CorrEctioN in magneTic Resonance) presents an AI-driven, real-time motion correction pipeline based on optimized GRU networks, inspired by tagging and fast-trigger algorithms from high-energy physics. Models evaluated on held-out test sets achieve good predictive performance and overall positive framewise displacement (FD) gains. These results demonstrate feasibility for prospective scanner integration; future work will complete in-vivo validation.