A Machine Learning Model for Predicting Depression in Moroccan Rheumatoid Arthritis Patients

Background and objective Rheumatoid arthritis (RA) is a chronic inflammatory condition that significantly impacts the quality of life. Depression in RA exacerbates pain and reduces the likelihood of remission. Predicting depression in RA is often neglected due to time and resource constraints. Hence, this study aimed to develop a machine learning (ML) model for predicting depression in RA patients. Methodology We included 112 RA patients from CHU Hassan II, Fez, Morocco. Depression was assessed using the Hospital Anxiety and Depression Scale (HADS) scale, and clinical data were extracted from medical records. Twelve features were used to develop five ML models: support vector machine (SVM), random forest (RF), decision tree (DT), logistic regression (LR), and gradient boosting classifier (GBC). Data preprocessing involved managing missing values, normalizing data, and encoding variables. Model performance was evaluated using accuracy, precision, recall, F1 score, and area under the curve (AUC). Results The ML-based feature selection method showed the optimal performance. The LR model performed best in predicting depression, with 76.5% accuracy, 72.2% precision, 81.2% recall, an F1 score of 0.765, and an area under the receiver operating characteristic curve (ROC AUC) of 0.767. Conclusions Our study highlights the significance of ML models in predicting depression in RA patients. The selected features and the LR model showed promising performance. Further research is required to validate these results and develop more advanced models. Utilizing such tools could significantly impact the management of RA patients by identifying those at risk of depression and providing appropriate psychological support.