Early and accurate diagnosis of Parkinson's disease (PD) remains challenging due to reliance on clinical assessment of motor symptoms. This retrospective study compared the diagnostic performance of neuromelanin (NM)-sensitive MRI and free water (FW) imaging in 247 patients with early PD and 78 controls from a single tertiary center, with independent external validation in a separate cohort acquired at another institution. NM volumes were measured in the substantia nigra pars compacta and its functional subregions (limbic, associative, and sensorimotor), while FW values were extracted from corresponding substantia nigra regions using diffusion tensor imaging. A combined NM model incorporating subregional volumes was developed through multivariable logistic regression. NM volumes were reduced across all regions in early PD patients compared to controls, while FW values showed no significant group differences. NM measurements consistently outperformed FW imaging across all regions, with the combined NM model achieving superior diagnostic accuracy, and NM-based measures also demonstrated higher diagnostic performance than FW in the external validation cohort. These results highlight the clinical utility of NM-MRI for early diagnostic assessment and its potential integration into emerging multimodal biomarker frameworks for PD.
Roh et al. (Fri,) studied this question.