Identifying Bone Tumor Using X Ray Images

More number of deaths occurring every year due to a common health issue, cancer. It is one of the diseases distributed over large areas. Among all, bone cancer is most predominant, having a high death rate. Hence this is most important and a lot of research is going on and to identify bone cancer X-RAY scanned images are used. They provide a picture detailed with the growth of the tumor and track its growth in the body so it is better by providing good results compared to other techniques. Therefore in medical fields most of the image processing techniques are used to detect cancer in its early stages .In this project we present an approach to detect bone cancer in scanned images. The steps involved in it are proposed using methods of image preprocessing where median filter is used followed by segmentation where mathematical morphological operations are used. Geometrical features like area, perimeter, and eccentricity are calculated to the tumor detected part. At last CNN classifier is used to classify whether it is cancerous (Malignant) or normal (Benign). In our project we will be using the following algorithms for classification like Support Vector Machine (SVM) as existing and Convolution Neural Network (CNN) as proposed and its accuracy has been calculated and well compared. From the results obtained it is proved that the proposed Convolution Neural Network (CNN) performs better than existing Support Vector Machine (SVM).