Deep-learning Saturation Transfer Magnetic Resonance Fingerprinting (ST-MRF) in Patients with Parkinson’s Disease

Abstract Parkinson’s disease (PD) is marked by progressive neurodegeneration in the substantia nigra (SN). This study evaluated deep-learning saturation transfer magnetic resonance fingerprinting (ST-MRF) to quantify molecular and microstructural changes in PD. We examined 23 patients with PD and 22 matched healthy controls using multimodal imaging, including ST-MRF. ST-MRF detected significant molecular and microstructural alterations in the SN of PwPD compared to HCs, including increases in magnetization transfer ratio at 3.5 ppm (MTR (3.5ppm, 1.5 µT); 0.61 ± 0.02 vs. 0.60 ± 0.02, p = 0.014), MTR (-3.5ppm, 1.5 µT); 0.61 ± 0.02 vs. 0.59 ± 0.02, p = 0.008)), and decreases in T2w (51.9 ± 3.4 ms vs. 54.5 ± 1.3 ms, p = 0.005), suggesting disrupted protein homeostasis, axonal loss, and iron accumulation. ST-MRF provides multi-parametric insights into PD-related pathology and may serve as a quantitative biomarker for future clinical trials. Further validation in larger, longitudinal cohorts is warranted.