[1M4-CC] Late Breaking Research Session - A

[1M4-CC-01] Prediction of mTSS in Rheumatoid Arthritis of Hand: A Comparison of Approaches Using Machine Learning and Deep Learning

Kohei Nakatsu¹, Kento Morita², Daisuke Fujita¹, Syoji Kobashi¹ (1. University of Hyogo, 2. Mie University)

It is estimated that there are 600,000 to 1,000,000 patients with rheumatoid arthritis (RA) in Japan. The modified total sharp score (mTSS) calculated from hand X-ray images is a standard diagnosis method of RA progression but the determination of the score depends on the experience of the physicians. It desires computer-aided diagnosis (CAD) systems to improve the quality of diagnosis for RA patients. In this study, we compare a method using ridge regression and methods using 3 models (VGG16, DenseNet and Xception) of deep learning with respect to the mTSS prediction of hand joints. To compare the 4 method, we conducted an experiment on 90 RA patients using X-ray images of their hands. The experimental results showed that the method using ridge regression gave the best accuracy.