Deep convolutional neural networks for computer-aided breast cancer diagnostic: a survey

Abstract

Advances in deep learning networks, especially deep convolutional neural networks (DCNNs), are causing remarkable breakthroughs in radiology and imaging sciences. These advances have influenced the development of computer-aided diagnosis (CAD). This study presents applications of DCNNs for computer-aided breast cancer diagnosis. We discuss the recent breakthrough, achievements, and notable advances in CAD for breast cancer. Various key and novel insights and challenges on the use of DCNNs for mammogram analysis have been presented in the paper. The latest deep learning toolkits and libraries that are available and insights for using them have been elaborated. We also point out the possible limitations in the use of DCNNs for breast cancer detection. Finally, give some ideas of future research which can address the existing limitations.