Machine Learning for Parkinson’s Disease Biomarkers: A Systematic Review of Diagnostic Performance, Risk of Bias, and Clinical Translation Readiness

Background/Objectives: Machine learning (ML) is frequently proposed for Parkinson’s disease (PD) diagnosis, but reported accuracy may not reflect methodological quality, generalizability, or clinical feasibility. We compared diagnostic performance, risk of bias, external validation, and deployment burden across biomarker modalities.