MEDICAL DECISION SUPPORT FOR FOOTBALL PLAYERS BASED ON MACHINE LEARNING HISTORICAL INJURY DATA

Abstract

Utilizing massive clinical data in the field of football players injuries for assisted medical decision support is the core technology and inevitable development trend of smart healthcare. However, due to the characteristics of medical data such as feature redundancy and imbalance of data sample categories, it has been difficult for traditional data mining algorithms to be directly applied in medical data research. In this paper, we propose a data-driven football players injury prediction method based on the experimental study of football players injuries occurring during the learning process of youth professional soccer training, which is based on the machine learning method of decision tree classifier. Through a semester of data statistics and experiments, the model has a high accuracy of injury prediction, which can provide early warning of youth football players injuries and support medical decision-making.