Multi-Metabolite AI Boosted CEST-MRF in an MPTP Mouse Parkinson’s Model

Motivation: Previous glutamate CEST-weighted studies were unable to separate the confounding effects stemming from different proton pools. Goal(s): To develop a rapid and quantitative CEST MR Fingerprinting (MRF) pipeline for glutamate, amide, and semisolid MT imaging and validate it using a Parkinson's disease (PD) mouse model. Approach: An AI-based CEST-MRF framework was designed and used for imaging glutamate phantoms and MPTP PD mice (n=8). Results: In-vitro maps were in excellent agreement with ground truth (r=0.95, p<0.0001). The in-vivo maps exhibited significant increases in the glutamate and MT concentrations following MPTP treatment, in agreement with MRS. Impact: A multi-metabolite CEST-MRF quantitative approach was developed and validated using a mouse PD model. The extensive semisolid MT, amide and glutamate proton-exchange maps are expected to aid in the diagnosis and characterization of Parkinson's disease.