An Efficient CNN for Hand X-Ray Classification of Rheumatoid Arthritis

Abstract

Hand radiography is the initial imaging test for the monitoring of rheumatoid arthritis (RA) progress. Identifying the exact stage of RA is a complex task, as human capabilities often limit the methods of diagnosis involved.Convolutional Neural Network(CNN) is the core for hand detection for identifying intricate patterns. The human brain functions in an advanced manner, so CNN has been designed based on biological neural networks in the human brain for mimicking its complex functions.This paper therefore introduces the Convolutional Neural Network (CNN) that can automatically learn the characteristics and predict the class of hand radiographs from a broad data set.The simulation of the CNN intermediate layers, which describes the dynamics of the current network, is also shown. For model preparation, the 290 hand x-ray dataset is used. The result indicates that hand x-rays are rated with an accuracy of 94.46% by the proposed methodology. The network sensitivity is 0.95 and 0.82 is the specificity.