The data mining applications of shoulder pain patients treatment: physical therapy equipment usage approaches

Abstract

The purpose of this paper is to apply the data mining techniques to discover and predict the recovery duration from physical therapy equipment usage patterns based on a classification system and establish selection rules of physical therapy techniques based on the association rule discovery method to support the decision making for physical therapists in the treatment of shoulder pain patients. The prediction system is driven by the usage patterns of physical therapy equipment and the association rule discovering method is applied for studying of the association in the amount of physical therapy equipment. The classification system is experimented and compared among the Naive Bayes, Neural Network, and Decision Tree. The best result is 91.35% accurate. In addition, we present the association rule discovering method for study the association within equipment usage amount of physical therapy equipment. The best top five interesting rules are demonstrated. Both data mining applications of this research could support the decision making in the treatment of shoulder pain patients.