Mathematical Morphological Approach for Mammogram Image Segmentation and Classification
S. Pitchumani Angayarkanni, Nadira Banu Kamal · International journal of advanced computer science · 2015
This paper presents the mathematical morphological and rough set based approach in detection and classification of cancerous masses in MRI mammogram images. Breast cancer increases the mortality rate in India especially in women since it is considered to be the second largest form of disease which leads to death. Mammography is the best method of diagnosing early cancer. The Computer Aided Diagnosis lacks accuracy and it is time consuming. So we propose a hybrid mathematical approach for detection of cancerous masses in MRI mammogram. MRI Mammogram images are enhanced and the artifacts are removed using the Fuzzification technique. The ROI(Region of Interest) is extracted using Graph Cut method and the Four mathematical morphological features are calculated for the segmented contour . The features which play a vital role in classification of masses in mammogram into Normal, Benign and Malignant are extracted using ID3 algorithm. The sensitivity, the specificity, positive prediction value and negative prediction value of the proposed algorithm were determined and compared with the existing algorithms. Automatic classification of the mammogram MRI images is done through three layered Multilayered Perceptron .The weights are adjusted based the Artificial Bee Colony Optimization technique .Both qualitative and quantitative methods are used to detect the accuracy of the proposed system. The sensitivity, the specificity, positive prediction value and negative prediction value of the proposed algorithm accounts to 98.78%, 98.9%, 92% and 96.5% which rates very high when compared to the existing algorithms. The area under the ROC curve is 0.89. A GUI based tool was developed for the proposed methodology. An android application using simulator was developed to make the doctor and patient to view the image with appropriate information like Patient Name, age ,Size of tumor, Nature of tumor and type of treatment .