Adversarial Networks for the Detection of Aggressive Prostate Cancer
This work addresses the challenge of data-hungry deep learning models in medical imaging, specifically for detecting aggressive prostate cancer, offering an incremental improvement over existing methods.
The authors tackled the problem of semantic segmentation for aggressive prostate cancer detection in MRI images with limited data by using an adversarial network to train fully convolutional networks, achieving superior detection sensitivity and dice-score compared to the standard method, with relative gains especially pronounced in the small dataset limit.
Semantic segmentation constitutes an integral part of medical image analyses for which breakthroughs in the field of deep learning were of high relevance. The large number of trainable parameters of deep neural networks however renders them inherently data hungry, a characteristic that heavily challenges the medical imaging community. Though interestingly, with the de facto standard training of fully convolutional networks (FCNs) for semantic segmentation being agnostic towards the `structure' of the predicted label maps, valuable complementary information about the global quality of the segmentation lies idle. In order to tap into this potential, we propose utilizing an adversarial network which discriminates between expert and generated annotations in order to train FCNs for semantic segmentation. Because the adversary constitutes a learned parametrization of what makes a good segmentation at a global level, we hypothesize that the method holds particular advantages for segmentation tasks on complex structured, small datasets. This holds true in our experiments: We learn to segment aggressive prostate cancer utilizing MRI images of 152 patients and show that the proposed scheme is superior over the de facto standard in terms of the detection sensitivity and the dice-score for aggressive prostate cancer. The achieved relative gains are shown to be particularly pronounced in the small dataset limit.