Development of MATLAB Kinect Skeletal Tracking System (MKSTS) for gait analysis

Current gait analysis systems are expensive and require large space with well controlled environments, which limits the measurement to laboratory settings such as ViCON motion analysis system. An economical alternative method of gait analysis system is proposed, in which MATLAB Kinect Skeletal Tracking System (MKSTS) is developed using MATLAB programming software and Microsoft Kinect Sensor to accumulate the important gait parameters based on the pixel coordinates of the body joints which requires no markers. The validity of MKSTS as an alternative method in determining the gait parameters is investigated and compared to ViCON motion analysis system. A healthy participant walked along the Kistler force plate with her lower limb kinematics data are measured by ViCON motion analysis system and MKSTS concurrently. Gait parameters of right hip and right knee joint angles were extracted from both devices and compared graphically in a period of one complete gait cycle. The results indicated that the correlation coefficients of MKSTS were 0.996 for right hip angles and 0.987 for right knee angles compared to ViCON motion analysis system as reference measurement. Therefore, MKSTS is able to measure the gait parameters, though there was a substantial error in accuracy. As a conclusion, the result of the research suggests that MKSTS is valid to be used as an alternative method in determining the gait parameters, depending on the required accuracy level. MKSTS will then further developed to be implemented into telerehabilitation system.