Continuous High-Frequency Smartwatch Sensor Telemetry Predicts Parkinson’s Motor Fluctuation Subtypes

Primary Premise (medRxiv (Neurology)): "A 24-month longitudinal cohort of 650 Parkinson’s patients monitored with wrist-worn IMU accelerometer arrays establishes automated machine learning scores correlating with MDS-UPDRS Part III subscores." Within Parkinson's Disease (PD), this work examines Digital Biomarkers, Wearables, MDS-UPDRS, Machine Learning, Motor Fluctuations in relation to α-synuclein seed amplification assays (SAA), LRRK2/GBA1 variants, DAT-SPECT, and digital motor kinematics. Authored by Lipsmeier F, Taylor KI, Postuma RB.

Study Architecture: Experimental / Clinical CNS Design (650 Patients) utilizing Accelerometer Time-series, Clinical Scores. Key reported conclusion: "A 24-month longitudinal cohort of 650 Parkinson’s patients monitored with wrist-worn IMU accelerometer arrays establishes automated machine learning scores correlating with MDS-UPDRS Part III subscores."

Secondary Analysis Applications: (1) Cross-referencing Accelerometer Time-series, Clinical Scores variables (Digital Biomarkers, Wearables, MDS-UPDRS, Machine Learning, Motor Fluctuations) against open repositories in Parkinson's Disease (PD); (2) Benchmarking feature distributions and effect sizes reported in "Continuous High-Frequency Smartwatch Sensor Telemetry Predicts Parkinson’s Motor"; (3) Informing cohort stratification and endpoint selection.

Preprint Status: This record is an unrefereed preprint; findings should be verified against subsequent peer-reviewed publication. When integrating these Accelerometer Time-series, Clinical Scores findings, account for cohort-specific inclusion criteria (650 Patients), potential site/instrument batch effects, and repository license terms (open_access).