Machine Learning of T2W FLAIR MR Images Can Detect Parkinson Disease

Background: Parkinson disease (PD) is due to an abnormal aggregation of α-synuclein and the loss of dopaminergic neurons in the substantia nigra (SN) and striatum. When motor symptomatology occurs, reportedly there is already a loss of 40% to 60% of the nigral dopaminergic neurons and almost 80% of the synaptic function. Screening for PD prior to symptomatology using machine learning (ML) analysis of MRI findings may allow earlier intervention, which might slow the progression of this disease. Objective: To determine whether MRI findings using routine T2-weighted fluid-attenuated inversion recovery (T2W FLAIR) images evaluated by ML techniques could accurately identify evidence of PD. Design: Retrospective cohort study. Participants/Methods: Both PD and healthy control (HC) subjects evaluated from January 2019 and October 2023 were recruited. Information from MRI images of subjects from the primary in-hospital system was randomly assigned to the training and internal test sets. more...