A Deep Neuro-Fuzzy Rule-Based System for Mammography Image Classification

Breast cancer has become one of the most significant diseases all over the world as the mortality rate is increasing day by day. Therefore, early diagnosis of breast cancer is very important to reduce the mortality rate. This is because the earlier breast cancer is detected; the more successful treatments are applied to get rid of the disease. The aim of this work is to calculate the breast cancer risk score using deep learning methods and a fuzzy rule-based system to enable people to take precautions against breast cancer. Therefore, a hybrid system consisting of two independent parts is described in this paper: In the first part, a deep neural network is trained to provide inputs to the fuzzy rule-based system. In the second part, the fuzzy rule-based system calculates the risk scores for breast cancer by using the output of the deep neural network structure. We also propose a simple rule-based classifier that uses the calculated risk value to diagnose breast cancer. We compare the results of our model with the existing neural network-based models and show that breast cancer can be diagnosed with high confidence using the proposed neuro-fuzzy rule based system.