CNN-Based CAD for Breast Cancer Classification in Digital Breast Tomosynthesis

Digital breast tomosynthesis (DBT) is a promising new technique for breast cancer diagnosis. DBT has the potential to overcome the tissue superimposition problems that occur on traditional mammograms for tumor detection. However, DBT generates numerous images, thereby creating a heavy workload for radiologists. Therefore, constructing an automatic computer-aided diagnosis (CAD) system for DBT image analysis is necessary. This study compared feature-based CAD and convolutional neural network (CNN)-based CAD for breast cancer classification from DBT images. The research methods included image preprocessing, candidate tumor identification, three-dimensional feature generation, classification, image cropping, augmentation, CNN model design, and deep learning. The accuracy rates (standard deviation) of the CNN- and feature-based CAD for breast cancer classification were 74.85% (0.122) and 87.12% (0.035), respectively. The T value was -6.229, and the P value was 0.00 < 0.05, which indicated that the CNN-based CAD significantly outperformed feature-based CAD. The results can be applied to clinical medicine and assist radiologists in breast cancer identification.