Chapter 7 - Explainable deep learning approach to predict chemotherapy effect on breast tumor’s MRI

Abstract

To reduce breast tumor size before surgery, neoadjuvant chemotherapy is applied systematically to patients with local breast cancer. However, with the current protocols, it is not yet workable to have an early prediction on the effect of chemotherapy on a patient. Predicting response to chemotherapy could reduce toxicity and delays to effective treatment. Computational analysis of dynamic contrast-enhanced magnetic resonance images (DCE-MRI) through deep convolution neural network (CNN) has proved a significant performance to classify responsive and nonresponsive patients. This presents a new explainable deep learning (DL) model for predicting breast cancer response to chemotherapy based on multiple MRI inputs.