A Study on Analyzing Symptoms of Korean Rheumatoid Arthritis Patients Using Machine Learning

Abstract

Rheumatoid arthritis, one of the musculoskeletal disorders, is an autoimmune disease that the immune system attacks itself and causes inflammation of the joint tissue. Rheumatoid arthritis is found in a variety of ages, mainly in 30s and 50s, and is more common in women than in men. Golden time of treatment would be within 6 months after the symptoms of rheumatoid arthritis appear, and the sooner it starts, the better it is for patients. If early diagnosis and treatment are performed, the symptoms may improve sufficiently, but the awareness or diagnosis rate is not high. If significant symptoms associated with rheumatoid arthritis are known, self-diagnosis is possible, and the disease may be detected early. In this study, the decision forest algorithm, which is one of artificial intelligence algorithms, was applied to the EMR data of Eulji University Hospital to analyze the symptoms of rheumatoid arthritis patients. The results of the experiment were compared with the opinion of the doctor and 90% agreement was obtained. If the results of this study are used for self - diagnosis, it will be able to easily diagnose the disease and the quality of life can be improved by early detection and prevention of deterioration.