Application of the Random Forest Algorithm for Identifying Abnormal Patterns of Knee Joint Movement

Abstract

Gait analysis through advanced technologies such as motion capture sensors and machine learning techniques enables a detailed study of body movement patterns, significantly contributing to a better understanding of health conditions and sports performance. The application of the Random Forest algorithm for identifying abnormal gait patterns, particularly in cases of anterior cruciate ligament injuries and knee osteoarthritis, has enabled precise classification of gait parameters. Using the OptiTrack system for data collection on both healthy individuals and those with injuries or diseases, a high accuracy model has been achieved in classification, expressed through metrics such as accuracy, precision, recall, and F1 Score. This approach provides efficient and objective classification of potential injuries or diseases, significantly enhancing the diagnosis, therapy, and prevention of knee joint injuries and diseases.