Impact of Video Quality on Parkinson's Disease Digital Assessment

315 videos of Parkinson's disease patients performing finger-tapping tasks, collected by figshare admin karger and last updated in April 2026. The dataset includes 227 high-resolution clinical recordings and 88 patient-recorded home videos, used to quantify how video parameters affect pose estimation and clinical scoring accuracy. It supports research into remote monitoring and digital biomarker reliability.

Use Cases

• Benchmarking human pose estimation algorithms based on video quality parameters like frame rate and resolution.
• Developing guidelines for remote patient monitoring systems based on hand visibility and lighting condition analysis.
• Training models for automated UPDRS scoring based on finger-tapping event detection from video.
• Studying the robustness of digital biomarkers for Parkinson's disease in uncontrolled home environments.

Strengths

• Dataset size of 5.2 MB with 315 total videos from two distinct settings (clinical and home).
• Includes systematic analysis of key video parameters: resolution, frame rate, lighting, and hand visibility.
• Performance is measured using specific metrics: MPJPE, PCK, and ELFF.

Limitations

• Row count and specific column-level documentation are unknown, limiting suitability assessment.
• Data may reflect bias inherent to the specific patient cohorts and recording conditions used in the study.
• Description metadata is limited; actual data structure and content require manual inspection after download.

Provenance

Source

figshare

Collection Method

Videos were collected from Parkinson's disease patients performing finger-tapping tasks in controlled clinical and home settings.

Freshness

Last updated 2026-04-18 07:55:21