Large-scale mammography CAD with Deformable Conv-Nets
This work addresses breast cancer screening for radiologists by improving CAD accuracy, though it is incremental as it adapts existing methods to a new domain.
The authors tackled breast cancer detection in mammograms by adapting a deformable convolutional network (DCN) architecture to handle large image sizes without resolution loss, achieving an AUC of 0.879 on a withheld dataset of 130,000 images, which outperformed other submissions in the DREAMS challenge.
State-of-the-art deep learning methods for image processing are evolving into increasingly complex meta-architectures with a growing number of modules. Among them, region-based fully convolutional networks (R-FCN) and deformable convolutional nets (DCN) can improve CAD for mammography: R-FCN optimizes for speed and low consumption of memory, which is crucial for processing the high resolutions of to 50 micrometers used by radiologists. Deformable convolution and pooling can model a wide range of mammographic findings of different morphology and scales, thanks to their versatility. In this study, we present a neural net architecture based on R-FCN / DCN, that we have adapted from the natural image domain to suit mammograms -- particularly their larger image size -- without compromising resolution. We trained the network on a large, recently released dataset (Optimam) including 6,500 cancerous mammograms. By combining our modern architecture with such a rich dataset, we achieved an area under the ROC curve of 0.879 for breast-wise detection in the DREAMS challenge (130,000 withheld images), which surpassed all other submissions in the competitive phase.