Development of a Trunk Balance Training and Assessment System based on an Inertial Sensor Integrated with a Tilting Chair

Abstract

Trunk balance is essential for physical stability, but traditional assessment tools, such as force plates and 3D motion capture systems, can be inaccessible or unsafe for patients unable to stand. This study introduces a seated trunk balance training and assessment system that combines a tilting chair with an Inertial Measurement Unit (IMU). The system features a kinematic design that transmits the user's trunk movements to the chair's seat, allowing for quantitative measurement without the need for body-mounted sensors. A pilot study was conducted to evaluate technical feasibility by comparing a seat-mounted sensor (CS-IMU) with a reference sensor (T12-IMU) attached to the T12 vertebra. Assessment items included Range of Motion (RoM), rotation accuracy, and agility. Results indicated a very high correlation for lateral bending (r = 0.958) and rotation phase angle (r = 0.990), while flexion/extension showed a moderate correlation (r = 0.691). Additionally, agility tasks demonstrated consistent results in the lateral direction (r = 0.806). These findings suggest that the proposed system can effectively quantify multi-dimensional trunk movements in a seated position, making it a valuable tool for high-risk populations, such as patients recovering from acute stroke or spinal cord injuries.