Deep Learning-Based Breast Cancer Analysis and Predicting the Severity

Abstract

Cancer disease is classified into various subtypes such as lung, breast, uterus, stomach, and throat cancers. Cancer is one of the dangerous, fast-spreading diseases and it directly leads to death. The early diagnosing and predicting cancer its severity becomes an essential task to save human life and it can facilitate the clinical management of patients. But early detection of cancer and classification of the severity level is a crucial task. It helps various research people working in biomedical, bioinformatics fields to understand the risk level of breast cancer. To provide a better approach for analyzing and predicting the severity level of breast cancer, model the progression and treatment of cancerous conditions. The proposed model utilizes the ability of Convolutional Neural Network to learn and detect a greater number of features which improves the classification accuracy. CNN is implemented and experimented in MATLAB software and the results are compared with other earlier methods such as Decision Trees (DTs), Bayesian Networks (BNs), Support Vector Machines (SVMs), and Artificial Neural Networks (ANNs) are highly used in breast cancer analysis. From the results, it has been found that CNN outperforms the other approaches.