Augmented Reality-Based Finger Joint Range of Motion Measurement: Assessment of Reliability and Concurrent Validity.

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Digital Health

Background:

• Accurate measurement of finger joint range of motion (ROM) is crucial for diagnosing and monitoring hand conditions.
• Traditional goniometry can be subjective and time-consuming.
• Augmented reality (AR) offers a novel approach for objective and accessible ROM assessment.

Purpose of the Study:

• To evaluate the reliability and concurrent validity of AR-based finger joint ROM measurements.
• To identify optimal camera views for AR-based ROM measurement at each finger joint.
• To assess the potential of AR technology for clinical use in hand rehabilitation.

Main Methods:

• A web application utilizing Google's MediaPipe Hands framework was developed for real-time AR hand tracking.
• Finger joint angles (MCP, PIP, DIP) were measured in flexion and extension from five camera views in 48 healthy individuals.
• AR measurements were compared against manual goniometry (reference standard), with acceptable validity defined as within 10°.

Main Results:

• Excellent test-retest reliability (ICC >0.75) was observed for AR-based ROM measurements across all joints and views in flexion, and in multiple views for extension.
• AR flexion measurements demonstrated acceptable concurrent validity (within 10° of goniometry) for specific joints and ulnar views.
• Multiple camera views consistently provided valid AR extension measurements at each joint.

Conclusions:

• AR-based finger joint ROM measurement exhibits high reliability and concurrent validity in healthy hands.
• Optimal camera views for AR measurement vary depending on the specific joint and motion (flexion/extension).
• Further validation in patients with pathological hand conditions is necessary before widespread clinical adoption.