Identifying Important Factors of Arthroplasty in Patients with Degenerative Knee Osteoarthritis Based on Association Rule Mining Approach

Background and Aim: Total Knee Arthroplasty (TKA) aims to reduce the pain and improve the quality of life of patients with progressive osteoarthritis. When the indication of patients' disease is established, this type of surgery should be performed as soon as possible because patients' late attendance increases surgical complications. Therefore, identification of factors influencing the choice of this type of treatment approach is of great importance. The purpose of this study is to identify the factors that influence the choice of this treatment approach in patients using the Apriori algorithm in the form of Association Rules. Materials and Methods: This study is performed on 233 patients referring to Imam Khomeini Hospital in Tehran for a knee replacement surgery; the needed data have been registered at Bone and Joint Reconstruction Research Center. In this study, after the preprocessing stage, the important factors in decision making of knee replacement surgery have been identified by using the Apriori algorithm and by its implementation in the software environment of RStudio. After being extracted, these factors and the relationship among them are given to orthopedic practitioners for confirmation. Results: In this study, flexion contracture above 20 degrees, deformity (varous-valgus) above 15 degrees, final flexion between 51-75 degrees, and medial cartilage destruction were, respectively, the most important factors in selecting patients for knee replacement therapy. Conclusion: The results showed that data-mining Algorithms could be used to identify effective factors to select patients for this treatment approach.