The Use of Artificial Intelligence in Improving Diagnostic Modalities in Rheumatoid Arthritis: A Narrative Review

Abstract

Rheumatoid arthritis (RA) is an inflammatory autoimmune condition affecting the joints and other organs such as the heart, eyes, and lungs. For decades, it has been diagnosed through assessing a combination of clinical picture, serologic biomarkers, and radiographic studies. However, the possibility of false negative test results and inability to detect early arthritic changes make RA diagnosis challenging. The diagnostic accuracy of the RA diagnostic modalities has substantially improved since the emergence of artificial intelligence (AI)-based medical algorithms, resulting in timely disease prediction and prevention of irreversible joint damage. AI computational models employ machine learning (ML), natural language processing (NLP), and rule-based expert systems to enhance the diagnostic accuracy of rheumatological diseases, particularly rheumatoid arthritis. AI-based algorithms not only identify specific disease patterns to predict the early course of disease but also use visual scoring systems, enhancing imaging characteristics. Radiological studies such as X-ray, MRI, CT, and PET scan can quantify joint space narrowing, cartilage loss, synovitis, bone erosions, and bone marrow edema. In addition, ML-integrated microRNA gene profiling reshaped the microenvironment of joint space by modulating gene expression and reducing joint deterioration in rheumatoid arthritis patients, surpassing the rheumatoid factor (RF) and cyclic citrullinated peptide (CCP) risk scoring models.

Ähnliche Arbeiten