Wrist Motion Analysis Based on Action Detection and Pressure Sensors

Abstract

To address the high incidence of triangular fibrocartilage complex (TFCC) injuries among fitness enthusiasts, this study proposes a wrist motion analysis method integrating object detection with flexible pressure sensing. Ulnar-side wrist pressure data were collected and, combined with Python-based key point detection, used to extract three-dimensional wrist trajectories, flexion/ulnar deviation angles, and movement velocities. A novel “pressure coefficient” metric was defined, and together with peak pressure, distribution gradient, and other features, a risk assessment model for injury was developed. A MATLAB-based biomechanical simulation system was constructed to quantify the correlation between palm pressure center shifts and TFCC injury mechanisms. This research is the first to integrate object detection with pressure sensors, enabling simultaneous motion modeling and data acquisition, providing new insights for wrist injury prevention, and enriching the theoretical framework of pattern recognition in human wrist motion analysis.