Automated Severity Grading of Arthritis from Digital X-Ray Images using Deep neural network

Osteoarthritis (OA) is a progressive joint disease is caused due to breakdown of cartilage tissue. OA is one of the most common forms of arthritis that primarily affects older adults. However, it can affect younger people also due to joint injuries or other factors. OA can affect knee joint, hips, hands and spine. There is no cure for OA, but treatments attempt to relieve symptoms, improve joint function, and slow disease progression and control it. The diagnostics procedures involve physical examination, laboratory tests and imaging tests. This paper aims to provide an assisting tool for the experts in predicting the severity of the disease. In this study, a comparative analysis on the VGG 16 model with modified VGG 16 models in combination of random forest and multiclass support vector machine (M-SVM) algorithms are made. The M-SVM model provided the classification accuracy of 73.83 % which is better than the other two algorithms.