Gait analysis plays a crucial role in evaluating motor function in children with intellectual disabilities. Traditional gait measurement methods, such as the Zeno electronic walkway, provide high-precision data but are costly and have limited accessibility. Recently, 2D video-based gait analysis using computer vision techniques like OpenPose has emerged as a low-cost alternative. This study aims to assess the feasibility of using OpenPose for gait analysis by comparing its extracted parameters with those obtained from the Zeno walkway. Method: A total of 33 children with intellectual disabilities participated in this study. Following a practice session, each child completed three gait trials on the Zeno electronic walkway. To ensure consistency in gait patterns, the most stable trial was selected for analysis. The video recording of this specific trial was used to extract gait parameters with OpenPose, ensuring direct comparability between the two methods. The extracted parameters included step time, stance time, swing time, double support time, step length, and gait speed. A paired-sample t-test was conducted to compare the gait parameters obtained from both methods, while correlation analysis was performed to evaluate the level of agreement between the two measurement techniques. The paired-sample t-test showed no significant differences (p > 0.05) between gait parameters extracted using OpenPose and those measured by the Zeno walkway. Additionally, correlation analysis revealed a significant relationship between the parameters obtained from both methods, indicating strong agreement. These findings suggest that OpenPose can serve as a reliable alternative to electronic walkways for gait analysis in children with intellectual disabilities. The findings support the feasibility of OpenPose as a cost-effective tool for gait analysis, potentially replacing electronic walkways in certain contexts. However, its accuracy may be influenced by factors such as camera angle, individual posture variations, and lighting conditions. Future research should focus on optimizing tracking algorithms and validating the method under different movement conditions. This study contributes to the advancement of gait analysis technology, improving the accessibility and efficiency of motor function assessments in children with intellectual disabilities.
Ling et al. (Thu,) studied this question.