The success of neural networks on medical image segmentation tasks typically relies on large labeled datasets for model training. However, acquiring and manually labeling a large medical image set is resource-intensive, expensive, and sometimes impractical due to data sharing and privacy issues. To address this challenge, we propose AdvChain, a generic adversarial data augmentation framework, aiming at improving both the diversity and effectiveness of training data for medical image segmentation tasks. AdvChain augments data with dynamic data augmentation, generating randomly chained photo-metric and geometric transformations to resemble realistic yet challenging imaging variations to expand training data. By jointly optimizing the data augmentation model and a segmentation network during training, challenging examples are generated to enhance network generalizability for the downstream task. The proposed adversarial data augmentation does not rely on generative networks and can be used as a plug-in module in general segmentation networks. It is computationally efficient and applicable for both low-shot supervised and semi-supervised learning. We analyze and evaluate the method on two MR image segmentation tasks: cardiac segmentation and prostate segmentation with limited labeled data. Results show that the proposed approach can alleviate the need for labeled data while improving model generalization ability, indicating its practical value in medical imaging applications.

Treatment decisions for an incidental renal mass are mostly made with pathologic uncertainty. Improving the diagnosis of benign renal masses and distinguishing aggressive cancers from indolent ones is key to better treatment selection. We analyze 13261 pre-operative computed computed tomography (CT) volumes of 4557 patients. Two multi-phase convolutional neural networks are developed to predict the malignancy and aggressiveness of renal masses. The first diagnostic model designed to predict the malignancy of renal masses achieves area under the curve (AUC) of 0.871 in the prospective test set. This model surpasses the average performance of seven seasoned radiologists. The second diagnostic model differentiating aggressive from indolent tumors has AUC of 0.783 in the prospective test set. Both models outperform corresponding radiomics models and the nephrometry score nomogram. Here we show that the deep learning models can non-invasively predict the likelihood of malignant and aggressive pathology of a renal mass based on preoperative multi-phase CT images.