Abstract: Breast cancer is the most common malignancy in women. It is often characterized by lack of early symptoms, which results in late detection of the disease. Breast cancer can occur in both men and women. It is projected that there will be rise in new cancer cases and cancer deaths in the United States. The early diagnosis of breast cancer can increase the survival rate, as it can promote timely clinical treatment to the affected patients. The mammography is the most common method for diagnosis through which we can have knowledge of abnormalities from the input image. This paper covers image processing based analysis of mammographic image. Firstly the image is converted from 3D to 2D array. Then the image is binaries twice, first to identify breast tissue and second to identify cancer affected region. Considering region of interest that indicates affected pixels and normal pixels the percentage of disease affected areas is calculated.
Keywords: Breast cancer, Mammogram, segmentation, thresholding, Binarization, ROI, python 3.
| DOI: 10.17148/IJARCCE.2021.105106