Breast Cancer Detection with Multiparametric Magnetic Resonance Imaging using Deep Learning

Breast Cancer is a major cause of patient deaths all over the world. Timely treatment may help to reduce deaths to some extent. Screening techniques like mammography and Magnetic Resonance Imaging (MRI) are very popularly used for early detection of Breast Cancer. In this paper, an attempt is being made to assist radiologists to speed up the detection of lesions using deep learning. Dynamic contrast-enhanced MRI (DCE-MRI) and Diffusion Weighted Imaging (DWI) are applied as inputs to Convolutional neural network (CNN) models. The MRI images are collected and annotated by expert radiologists from a reputed hospital. Experimentation is done in MATLAB R2022a. The highest accuracy of 90.8% is achieved by Inception -V3, when compared with AlexNet and DenseNet-201.