Densely Dilated Spatial Pooling Convolutional Network using benign loss functions for imbalanced volumetric prostate segmentation

TL;DR

This paper introduces a novel convolutional network architecture, DDSP ConNet, for prostate segmentation in MRI, utilizing dense dilated pooling and benign loss functions, achieving state-of-the-art results on a benchmark dataset.

Contribution

The paper proposes a new encoder-decoder network with dense dilated pooling and residual connections, combined with benign loss functions, for improved prostate segmentation in MRI images.

Findings

Achieved a segmentation score of 85.78 on MICCAI PROMISE12 dataset.

Demonstrated the effectiveness of DSC and Jaccard loss functions over re-weighted cross entropy.

Outperformed most existing methods in prostate MRI segmentation.

Abstract

The high incidence rate of prostate disease poses a requirement in early detection for diagnosis. As one of the main imaging methods used for prostate cancer detection, Magnetic Resonance Imaging (MRI) has wide range of appearance and imbalance problems, making automated prostate segmentation fundamental but challenging. Here we propose a novel Densely Dilated Spatial Pooling Convolutional Network (DDSP ConNet) in encoder-decoder structure. It employs dense structure to combine dilated convolution and global pooling, thus supplies coarse segmentation results from encoder and decoder subnet and preserves more contextual information. To obtain richer hierarchical feature maps, residual long connection is furtherly adopted to fuse contexture features. Meanwhile, we adopt DSC loss and Jaccard loss functions to train our DDSP ConNet. We surprisingly found and proved that, in contrast to re-weighted cross entropy, DSC loss and Jaccard loss have a lot of benign properties in theory, including symmetry, continuity and differentiability about the parameters of network. Extensive experiments on the MICCAI PROMISE12 challenge dataset have been done to corroborate the effectiveness of our DDSP ConNet with DSC loss and Jaccard loss. Totally, our method achieves a score of 85.78 in the test dataset, outperforming most of other competitors.