Adversarial Deep Structured Nets for Mass Segmentation from Mammograms
This work addresses mass segmentation for breast cancer diagnosis in medical imaging, representing an incremental improvement with specific gains in accuracy.
The paper tackles mass segmentation in mammograms by proposing an end-to-end network combining a fully convolutional network with a CRF, position priors, adversarial training, and multi-scale features to address dataset limitations and improve accuracy. It achieves better performance than state-of-the-art methods on INbreast and DDSM-BCRP datasets.
Mass segmentation provides effective morphological features which are important for mass diagnosis. In this work, we propose a novel end-to-end network for mammographic mass segmentation which employs a fully convolutional network (FCN) to model a potential function, followed by a CRF to perform structured learning. Because the mass distribution varies greatly with pixel position, the FCN is combined with a position priori. Further, we employ adversarial training to eliminate over-fitting due to the small sizes of mammogram datasets. Multi-scale FCN is employed to improve the segmentation performance. Experimental results on two public datasets, INbreast and DDSM-BCRP, demonstrate that our end-to-end network achieves better performance than state-of-the-art approaches. \footnote{https://github.com/wentaozhu/adversarial-deep-structural-networks.git}