Fadhil, Abdullah FreidoonOrnek, Humar Kahramanli2023-01-292023-01-292021Fadhil, A. F., Ornek, H. K., (2021). A Computer-Aided Detection System for Breast Cancer Detection and Classification. Selcuk University Journal of Engineering Sciences, 20 (01), 23-31.2757-8828https://hdl.handle.net/20.500.12395/45059Breast cancer is a dangerous disease and considered the second cause of death for women globally. Reading breast cancer images requires experienced radiologists. Radiologists may have a problem with their visual decision about breast cancer. Therefore, a computer-aided detection (CAD) system is needed to help radiologists in their decisions. The early detection of breast cancer using computer vision systems, such as image processing, increases the success of treatment. Developing a well-designed CAD system is still a challenging problem because of the low rate of accuracy performance. In this paper, an improved CAD system is introduced for classifying breast cancer tumors into normal and abnormal classes. In this CAD system, a region-based segmentation approach, namely region growing, is used. Discrete wavelet transform is used for the histogram and texture-based feature extraction. The 120 candidate features were ranked and selected according to two criteria which are the interclass separation and classification accuracy criterion. Four different classifiers, Linear Discriminant Analysis, Artificial Neural Network, Decision Tree, and Support Vector Machine, were used for classification. The results are obtained using a 10-Fold cross-validation technique on the MIAS data set. The highest accuracy achieved was 93.6% by Support Vector Machine classifier using the best 69 features from the interclass separation method. The sensitivity and specificity achieved were 89.2% and 99.0%, respectively. The results show improved accuracy compared to previous works selected from the literature review.eninfo:eu-repo/semantics/openAccessArtificial Neural NetworkBreast Cancer DetectionDecision TreeDiscrete Wavelet TransformLinear Discriminant AnalysisSupport Vector MachineArticle2012331