Machine Learning Aiding in the Diagnosis of Parkinson's Disease

Abstract

Parkinson’s disease (PD) is a common, progressive and currently incurable neurodegenerative movement disorder. The diagnosis of PD is challenging, especially in the differential diagnosis of parkinsonism and in early PD detection. Due to the advantages of machine learning such as automatic data analysis and making inferences for individuals, machine learning techniques have been increasingly applied to the diagnosis of PD, and have shown some promising results. Machine learning-based imaging applications have made it possible to automatically differentiate parkinsonism and detect PD at early stages on magnetic resonance imaging (MRI), resting-state functional MRI (rs-fMRI), diffusion tensor imaging (DTI) and Single Photon Emission Computed Tomography (SPECT) images. Machine learning-based SPECT image analysis applications in PD have outperformed conventional semi-quantitative analysis in detecting PD-associated dopaminergic degeneration, performed comparably well as experts, and improved PD diagnostic accuracy of radiologists. Multimodal data (such as multimodal imaging and clinical data) may further enhance early PD detection. To integrate machine learning-based computer-aided diagnostic applications into clinical systems, further optimization and thorough validation of these applications are needed to make them accurate and reliable for the diagnosis of PD. It is anticipated that machine learning techniques will further improve differential diagnosis of parkinsonism and early PD detection, which may reduce error rate of PD diagnosis, allow early neuroprotective treatment to slow down neurodegeneration progression and prevent clinical symptoms from emerging in patients with early-stage PD and allow proper treatment to effectively relieve patients from suffering.