Breast Cancer Classification as Malignant or Benign Based on Texture Features Using Multilayer Perceptron

Nassir Salman and Semaa Ali (Baghdad University, Iraq); Safa L. Kailan (Al-Nahrain, Iraq); Faisel Mohammed (Baghdad University, Iraq)

Breast cancer is one of the most common malignant diseases among women. Mammography is at present one of the available method for early detection of abnormalities, which is related to breast cancer there are different lesions that are breast cancer characteristic such as masses, which can be detected through this technique. , the images are divided according to the Mini-MIAS database. In order-to classify the cluster as-malignant and benign, the 2nd order co-occurrence statistical properties of the image such as contrast, correlation, homogeneity and energy were adopted. The gray-scale convergent matrices (GLCM) have been used with a suggested feature of gray level density matrices (GLDM) to identify abnormal tissue (malignant) and natural tissue (benign). The proposed method of analysis was tested on several images taken from an organization in the UK interested in understanding breast-x-ray images, generated a database of digital mammograms, and very good results were obtained for breast cancer detection. The proposed multilayered design performance significantly improved the diagnosis of breast cancer by more than 95.65% and 95.8% sensitivity, 95.5% specificity for all dataset, 91.1% with 100% sensitivity and 84% specificity for 70% training data and 30% testing data .

Journal: International Journal of Simulation- Systems, Science and Technology- IJSSST V20

Published: Feb 28, 2019

DOI: 10.5013/IJSSST.a.20.01.12