Design and Validation of an Inertial Motion Capture System for Human Dynamic Balance Assessment

The assessment of human balance ability holds significant clinical importance for enhancing health, preventing injuries, monitoring rehabilitation progress, and optimizing sports and physical training programs. This study introduced a system with 17 low-cost inertial measurement units (IMUs) to capture human motion data and analyze balance parameters, providing essential balance analysis for clinical patients. Initially, we utilized the inertial motion capture system to collect full-body motion IMU data and fed the data into OpenSim's human body model for calculating and analyzing gait kinematics. Subsequently, we calculated kinematic parameters relevant to human balance assessment based on human motion characteristics. To validate the feasibility of our approach, we recruited two subjects for walking experiments. We computed parameters related to human gait, posture control, and joint range of motion from the IMU data collected. These parameters were compared with those from an optical motion capture system, which was considered as the gold standard. The inertial motion capture system exhibited a root mean square error (RMSE) of 0.027 m for stride length, 0.031 m for lateral swing position of the lower limb segment center of mass, 0.010 m for vertical swing position of the lower limb segment center of mass, 7.2° for knee joint flexion angles, and 5.1° for elbow joint flexion angles. The results indicate that the newly developed inertial motion capture system and method accurately estimated parameters related to dynamic balance, showing potential for practical applications and can be used as a reliable tool for gait analysis and balance assessment in various settings.