Validity and Reliability of Kinect v2 for Quantifying Upper Body Kinematics during Seated Reaching

Area of Science:

• Biomechanics
• Rehabilitation Engineering
• Neurorehabilitation

Background:

• Upper limb kinematic analysis is crucial for monitoring stroke recovery but lacks clinical feasibility.
• The Kinect v2 offers a potential markerless solution for kinematic assessment.

Purpose of the Study:

• To evaluate the validity and reliability of the Kinect v2 for upper limb reaching kinematics analysis.
• To compare Kinect v2 performance against a gold-standard motion capture system (VICON).

Main Methods:

• Twenty-six healthy participants performed seated reaching tasks with added load.
• Simultaneous 3D motion capture of upper limb and trunk using Kinect v2 and VICON.

Main Results:

• Kinect v2 demonstrated moderate to excellent reliability for trunk compensations, hand range of motion, movement time, and mean velocity.
• Poor to moderate reliability was found for elbow/shoulder range of motion, time to peak velocity, and path length ratio.
• Instantaneous hand and elbow tracking lacked precision for velocity peaks and peak hand velocity assessment.

Conclusions:

• The Kinect v2 is suitable for clinical routine use in semi-automated quantitative diagnostics for upper limb rehabilitation.
• Its markerless and user-friendly nature aids in guiding individualized stroke rehabilitation.
• Limitations in tracking precision necessitate awareness for clinicians and engineers.