10.22376/ijpbs.2019.10.1.p1-12 Volume 5 Issue 3 2014 (July- September) Image processing is a widely used domain and its applied in various fields. Biomedical imaging is one of the research areas of image processing which is used to detect diseases. Biomedical imaging consists of various subresearch areas such as bone imaging, cancer imaging, brain imaging and blood cells imaging. The proposed work comes under bone imaging and the objective is to detect the occurrence of Rheumatoid Arthritis (RA). Rheumatoid Arthritis is an autoimmune disease that mainly affects the joints in the human body. Bone organ plays a vital role in identifying these types of diseases. Manual analysis of bone images is a long term process which needs an expert orthopedist for continuous assessment of bone scans which is costly and time consuming. The proposed work involves steps such as denoising, histogram smoothing, segmentation and edge detection in order to enhance the given image and separate the region of interest. The strength of a bone depends on Bone Mineral Density (BMD) which is a major factor in identifying bone diseases and fracture risk. The mathematical relationship between Bone Mineral Content and Volume by Region of Interest will helps in calculating Bone Mineral Density and various features of bone images such as area, mean, standard deviation and variance. The Gray Level Co-occurrence Matrix (GLCM) is one of the image analysis techniques that can be used for extracting features (Energy, Entropy, Contrast, Homogeneity and Correlation) of a bone image. A dataset is created from both the normal and eroded bone images by applying BMD and GLCM features. Whenever a new bone image is given as an input, the features of the image are extracted and compared against the dataset. The input image is classified as whether infected or not infected using neural network which achieved classification accuracy of about 96.66%. M.VINOTH AND B.JAYALAKSHMI Rheumatoid Arthritis, Bone Mineral Density, Gray Level Co-occurrence Matrix and Neural Network. 104-121