ABSTRACT
Osteoarthritis (OA) is one of the diseases commonly occurring in the joints of the knee, hip, and hand. It results in a loss of cartilage. The patient suffering from the symptoms of OA will have severe pain, stiffness, and a grating sensation during the movement of the joints. OA is graded into five levels, namely 0 to 4 based on the severity. In the proposed work, patient-specific X-ray images are collected from Chidgupkar Hospital, Solapur, India. The images are preprocessed to remove noise and to enhance their quality. A Wiener filter for noise removal and a histogram modeling-based image enhancement technique are used. For the sake of comparison, features such as edge curvature and textures are extracted. Multiple classifiers such as support vector machine (SVM) and k-nearest neighbor (k-NN) are used to classify the images as normal or abnormal, and the results are compared. By confirming the results, the k-NN-based classification system gives 100% accuracy for both normal and OA-affected images.
