DenseNet for Breast Tumor Classification in Mammographic Images

Breast cancer screening is an efficient method to detect breast lesions early. The common techniques are tomosynthesis and mammography images. However, the traditional manual diagnosis requires intense workload for pathologists, hence prone diagnostic errors. Thus, aim of this study was build a deep convolutional neural network automatic detection, segmentation, classification in Based on learning Mask-CNN (RoIAlign) developed automate RoI segmentation. Then feature extraction, selection were carried out by DenseNet architecture. Finally, precision accuracy model evaluated AUC, metrics. To summarize, findings show that methodology may improve efficiency tumor localization through medical image classification.