Predictive Modeling For Early Detection Of Juvenile Rheumatoid Arthritis Using Machine Learning

Abstract

Rheumatoid Arthritis (RA) is a chronic autoimmune disorder characterized by joint inflammation, often leading to joint damage and functional impairment. Although predominantly observed in adults, RA can affect teenagers and adolescents, resulting in Juvenile Idiopathic Arthritis (JIA), which, if undiagnosed, can lead to significant morbidity. Early detection of RA in teenagers is challenging due to atypical symptom presentations, which may result in delayed diagnosis. This study explores the application of supervised machine learning models for early RA detection among teenagers. By analyzing a comprehensive dataset of clinical symptoms, laboratory results, and imaging data, several machine learning models, including Support Vector Machines (SVM), Decision Trees, and Random Forest, were trained and tested to predict RA. Results show that these models can predict RA with high accuracy, indicating potential for use as a supportive diagnostic tool. The study concludes with recommendations for model optimization and clinical integration to assist early RA diagnosis in teenagers.