A novel Computer Aided Diagnosis (CADx) component is proposed for breast cancer classifications. Four major phases were conducted in this research. The first phase is pre-processing, this is followed by features extraction phase by using the Speed Up Robust

No abstract

Abstract. Classification of breast cancer is essential in determining the type of treatment that should be applied. Thus, a Computer Aided Diagnosis (CADx) may assist radiologists in making appropriate decision based on the classification results. In this paper, the classification is divided into two categories; to classify the cancer into benign and malignant (two classes) and to classify the character of the background tissue either fatty, glandular or dense (multi class). The Haralick texture features and Hu Invariants moments were proposed as the features extraction. There are three phases conducted in this study. The first phase is the pre-processing phase. This is followed by the features extraction phase where combination of moment based features with addition of four features was proposed. The final phase is the classification phase by using SVM classifiers. Results obtained shows that the accuracy of the proposed features are 90.5% and 77.5% for two classes and multi class respectively.

No abstract

Edge detection has been widely used especially in medical image processing field. In this paper we are comparing Sobel, Prewitt and Laplacian of Gaussian (LoG) edge detection techniques in segmenting the boundary of microcalcifications. The edge detection must satisfy the breast phantom scoring criteria before the segmentation phase is carried out. Then, all of the edge detection techniques are implemented in the Enhanced Distance Active Contour (EDAC) model for the segmentation process. Results obtained from Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve shows that the Prewitt edge detection has the highest value of AUC, followed by the Sobel and LoG which are 0.79, 0.72 and 0.71 respectively.