Self-normalized Classification of Parkinson's Disease DaTscan Images

Machine learning within the Parkinson's progression markers initiative: Review of the current state of affairs

The Parkinson's Progression Markers Initiative (PPMI) has collected more than a decade's worth of longitudinal and multi-modal data from patients, healthy controls, and at-risk individuals, including imaging, clinical, cognitive, and 'omics' biospecimens. Such a rich dataset presents unprecedented opportunities for biomarker discovery, patient subtyping, and prognostic prediction, but it also poses challenges that may require the development of novel methodological approaches to solve. In this review, we provide an overview of the application of machine learning methods to analyzing data from the PPMI cohort. We find that there is significant variability in the types of data, models, and validation procedures used across studies, and that much of what makes the PPMI data set unique (multi-modal and longitudinal observations) remains underutilized in most machine learning studies. We review each of these dimensions in detail and provide recommendations for future machine learning work using data from the PPMI cohort.

Detecting Parkinson's Disease through Gait Measures Using Machine Learning

Parkinson’s disease (PD) is one of the most common long-term degenerative movement disorders that affects the motor system. This progressive nervous system disorder affects nearly one million Americans, and more than 20,000 new cases are diagnosed each year. PD is a chronic and progressive painful neurological disorder and usually people with PD live 10 to 20 years after being diagnosed. PD is diagnosed based on the identification of motor signs of bradykinesia, rigidity, tremor, and postural instability. Though several attempts have been made to develop explicit diagnostic criteria, this is still largely unrevealed. In this manuscript, we aim to build a classifier with gait data from Parkinson patients and healthy controls using machine learning methods. The classifier could help facilitate a more accurate and cost-effective diagnostic method. The input to our algorithm is the Gait in Parkinson’s Disease dataset published on PhysioNet containing force sensor data as the measurement of gait from 92 healthy subjects and 214 patients with idiopathic Parkinson’s Disease. Different machine learning methods, including logistic regression, SVM, decision tree, KNN were tested to output a predicted classification of Parkinson patients and healthy controls. Baseline models including frequency domain method can reach similar performance and may be another good approach for the PD diagnostics.