Parkinson’s disease is a common neurodegenerative disorder characterized by the degeneration and death of dopaminergic neurons in the substantia nigra of the midbrain. PET/MR is a novel imaging technology that has been developed in recent years to enable the simultaneous acquisition of metabolic and structural images. This technology holds great value in the early detection of clinical markers for Parkinson’s disease. Artificial intelligence models have been extensively employed in clinical settings to assist physicians in diagnosis，yielding promising outcomes. In this paper，we aim to review the application of machine learning and PET/MR in the diagnosis of Parkinson’s disease，focusing on their potential to enhance diagnostic accuracy and facilitate early detection.