Edward Zhu¹ and Yu Sun², ¹Hong Kong, ²California State Polytechnic University, USA

Cerebral palsy is a common motor disability that causes gait abnormalities. Clinical gait analysis is expensive and inaccessible. We investigated the research question: how can we create an affordable and effective method using AI to provide gait analysis data to cerebral palsy patients? Machine learning and computer vision were used to develop a gait analysis mobile application. The MediaPipe library extracted pose vectors from both patient and expert gait videos. K-Means algorithm was utilized to match frames and determine joint angle differences. Flutter was used to create a complete app for real-time tracking and feedback. The AI model was deployed in the cloud. This research presents an application of machine learning and computer vision in an accessible and accurate solution. The K-means algorithm showed high accuracy with an average silhouette score 0.514 for expert videos. MediaPipe output had a high 29.2FPS on average.

Cerebral palsy, Gait abnormalities, Artificial intelligence, Machine learning