Browsing by Author "Alshana, Ghassan"

Now showing 1 - 1 of 1

Analysis of mammography images for cancer detection
(2016) Alshana, Ghassan
Mammography is the best available technique for early detection of breast cancer. The most common breast abnormalities that may indicate breast cancer are masses. Also, there are some signs that can lead to breast cancer diagnosis, such as architectural distortion and bilateral asymmetry. In this study, an algorithm is used to detect breast cancer in mammography images. Four stages are presented: (1) preprocessing, (2) segmentations of regions of interest (ROI), (3) feature selection and extraction, and (4) classification. In the preprocessing stage, the digital mammogram is pruned, 2D-median filter is used to filter the image and unnecessary labels are removed from the breast. In the segmentation stage, global thresholding is used for segmenting the breast. Morphological operations like erosion, dilation, opening and closing are used to enhance the breast. Seeded region growing is used for removing the pectoral muscle and for segmenting the mass in the breast. In the feature selection and extraction stage, intensity features are selected and extracted from the ROI. In the classification stage, the extracted features are fed into artificial neural network (ANN) classifier to classify the mass as malignant or benign. The output of the proposed method would assist radiologists to examine images containing unusual masses more closely and to help them minimize misinterpretation. The method achieved 91.30% sensitivity, 91.30% specificity and 91.30% accuracy resulting from the confusion matrix which is a performance evaluation metric.