Precision Balance Assessment in Parkinson's Disease: Utilizing Vision-Based 3D Pose Tracking for Pull Test Analysis

Area of Science:

• Neurology
• Biomedical Engineering
• Movement Science

Background:

• Postural instability is a frequent complication in advanced Parkinson's disease (PD), leading to falls and injuries.
• The pull test is the gold standard for assessing postural instability in PD, but suffers from low reliability.
• Objective, reliable assessment is crucial for managing fall risk in PD patients.

Purpose of the Study:

• To develop and validate a vision-based assessment tool (vPull) for the pull test using 3D pose tracking.
• To evaluate the inter-rater reliability and accuracy of vPull in distinguishing PD patients from healthy controls.
• To analyze PD-related impairments in postural response using quantitative metrics derived from vPull.

Main Methods:

• Utilized 3D pose tracking on single-sensor RGB-Depth recordings for vPull assessment.
• Validated vPull against marker-based motion capture in healthy individuals (n=15).
• Assessed vPull reliability and PD-specific postural responses in PD patients and controls (n=15 each).

Main Results:

• vPull demonstrated excellent agreement with marker-based motion capture in healthy individuals.
• vPull achieved high inter-rater reliability in assessing postural responses.
• Quantitative metrics from vPull effectively differentiated postural control between PD patients and controls, and across varying PD severity.

Conclusions:

• vPull offers an objective and reliable method for assessing postural instability in Parkinson's disease.
• This vision-based approach has the potential to improve clinical assessment sensitivity and specificity for fall risk prediction in PD.
• vPull shows promise for easy clinical implementation, enhancing patient care and management.