Implementation and Study of Performance Analyses of Various Classifiers on Mammograms
Tripty Singh, V. V. S. N. BharatKumar · 2014
In present research, designing and analysis of the performance of classifiers for breast cancer is done. In digital mammography, data mining techniques are used to detect and characterize abnormalities in images and clinical reports. In the existing approaches, the mammogram image classification is done in either clinical data or statistical features of an image using neural networks and Support Vector Machine (SVM) classifier. Present work address the issue by assigning different labels to automatically separating mass tissue from normal breast tissue given a region of interest in a digitized mammogram is investigated. It is the critical stage in developing a robust automated classification system because the classification depends on the accurate assessment of the tumour-normal tissue border as well as information gathered from the tumour area. Since the ultimate goal is robust classification, the qualities of the tissue segmentation are assessed by its impact on the overall classification performance. Computer Aided Diagnosis (CAD) technology helps in identifying lesions and assists the radiologist makes his final decision. A CAD system had been previously developed to perform the following tasks: (a) pre-processing, (b) segmentation and (c) feature extraction of mammogram images. The main focus of this work was two-fold: (a) to analyze these features, select the most important features among them and study their impact on classification accuracy and (b) to implement and compare Neural Networks (NNs),Support Vector Machines (SVMs) and KNN(K-Nearest Neighbour) Classifier's. And also evaluate their performances with these features. From obtained results it is shows that KNN classifier gives the better accuracy compared to NN and SVM classifiers. From the obtained results, the superiority of the proposed approach in terms of accuracy is justified.