A new combined system using ANN and complex wavelet transform for tissue density classification in mammography images

Abstract

Breast cancer is the most common type of cancer that occurs in one of every eight women in the world and is the most common in women. Early diagnosis of the disease is of great importance in order to reduce tissue loss and disease-related deaths. For this reason, in the literature, many studies have been done such as automatic breast tissue density classification, automatic normal-abnormal tissue classification and automatic benign-malignant tissue classification. In this study, a new combined system based on artificial neural networks (ANN) and complex wavelet transform is proposed to classify tissue density from mammography images. The study using 322 images of the MIAS database have resulted in classification success rates ranging from 80% to 94.79% for different breast tissue density classes (fatty, fatty-glandular, dense-glandular).