AUTOMATED DIAGNOSIS AND QUANTIFICATION OF RHEUMATOID ARTHRITIS USING MRI

摘要

Rheumatoid arthritis (RA) is a common chronic inflammatory joint disease which is characterized by persistent synovitis and frequent development of cartilage and bone deteriorations A major problem in the management of RA is the lack of reliable methods for monitoring joint inflammation, disease progression, and therapeutic response In this work we propose an automated method for the diagnosis and quantification of rheumatoid arthritis using magnetic resonance images of the hand The technique relies on the use of fuzzy C-mean algorithm for image segmentation. The number of clusters is determined by minimizing the appropriate validity index. Each vector, used as an input, has four features related to the intensity and position of pixels A defuzzification process follows the segmentation Small objects and objects corresponding to vessels are eliminated The method is evaluated using data from twenty three patients The obtained sensitivity and positive predictive rate of the proposed method are quite high.