Automatic Detection of Breast Cancer in Mammographic Image Using the Histogram Oriented Gradient (HOG) Descriptor and Deep Rule Based (DRB) Classifier Method
Dian Candra Rini Novitasari, Dewi Wahyuni, Misbakhul Munir, Irul Hidayati, Faris Mushlihul Amin, Kresna Oktafianto · 2019
Breast cancer is a cell disease in the breast glands that evolve abnormally and become benign or malignant tumors. Breast cancer is a common disease in developing countries and refers to malignancies that can be life-threatening. Mammographic imaging examination is performed to determine the mass of the breast lump to be analyzed for its malignancy. In this research, digital image processing is performed to diagnose breast cancer using mammographic images and feature extraction data with Histogram of Oriented Gradient (HOG), which is further classified using Deep Rule Based (DRB) Classifier. This research was conducted by dividing the data 90% for training data and 10% for testing data. Data are classified into three classes, consisting of normal, benignant, and malignant. The best accuracy obtained is 92.00%.