Ensemble Approach for Predicting the Diagnosis of Osteoarthritis Using Soft Voting Classifier

Background Osteoarthritis (OA) is a common degenerative disease of the joints. Risk factors for OA include non-modifiable factors such as age and gender and modifiable factors such as physical activity. Purpose This study aimed to construct a soft voting ensemble model to predict OA diagnosis using variables related to individual characteristics and physical activity and to identify important variables in constructing the model through permutation importance. Method Using the RFECV technique, the variables with the best predictive performance were selected among variables, and an ensemble model combining the RandomForest, XGBoost, and LightGBM algorithms was constructed, and the predictive performance and permutation importance of each variable were evaluated. Result The variables selected to construct the model were age, gender, grip strength, and quality of life, and the accuracy of the ensemble model was 0.828. The most important variable in constructing the model was age (0.199), followed by grip strength (0.053), quality of life (0.043), and gender (0.034). Conclusion The performance of the model for predicting OA was relatively good, and if this model is continuously used and updated, this model could readily be used to predict OA diagnosis and the predictive performance of OA may be further improved.