IMPROVEMENT IN DETECTION ACCURACY OF DIGITAL MAMMOGRAM USING POINT TRANSFORM AND DATA MINING TECHNIQUE

Cancer is one of the dangerous diseases faced by humans. Every one out of 100 women is facing breast cancer. So, to overcome this huge ratio many researches are being carried out. Prevention is better than cure; this paper presents one such attempt of detecting breast cancer in the early stages. In proposed method exponential point transform is carried out for image enhancement and in preprocessing stage pectoral mass is removed from the mammogram image. As the next step we apply K-means algorithm and morphological processing to identify the infected region and removal of unwanted region. Finally, Decision Tree Data mining technique is used for classifying features to detect presence of tumor. Hence by this approach we get more accurate results. The experimental results gave an accuracy of 97.03 %. Decision tree support vector machine classifier (SVM), linear quadratic discriminant analysis classifier