-
1
Parkinson's disease is characterized by motor symptoms like bradykinesia and gait instability, with no approved disease-modifying therapies available.
-
2
Current symptom assessment methods, like the MDS-UPDRS, have limitations in sensitivity, reliability, and granularity for individualized therapy.
-
3
A deep learning model using smartphone videos can accurately assess gait impairments in Parkinson's disease, outperforming individual clinician assessments.
-
4
Smartphone-based gait assessment improves access to care by enabling remote monitoring and shifting assessments to primary care settings.
-
5
Video-based approaches address barriers of wearable devices, fitting into existing workflows, but are limited by episodic data collection and video quality issues.
