DSU-net: Dense SegU-net for automatic head-and-neck tumor segmentation in MR images

TL;DR

This paper introduces DSU-net, a novel deep learning framework that improves head-and-neck tumor segmentation in MRI by using unpooling, dense blocks, and combined loss functions, leading to superior accuracy.

Contribution

The paper proposes a new segmentation architecture, DSU-net, with unpooling, dense blocks, and combined loss functions, enhancing boundary accuracy and feature reuse.

Findings

Outperforms existing segmentation networks in accuracy.

Effective boundary localization with unpooling.

Improved feature propagation with dense blocks.

Abstract

Precise and accurate segmentation of the most common head-and-neck tumor, nasopharyngeal carcinoma (NPC), in MRI sheds light on treatment and regulatory decisions making. However, the large variations in the lesion size and shape of NPC, boundary ambiguity, as well as the limited available annotated samples conspire NPC segmentation in MRI towards a challenging task. In this paper, we propose a Dense SegU-net (DSU-net) framework for automatic NPC segmentation in MRI. Our contribution is threefold. First, different from the traditional decoder in U-net using upconvolution for upsamling, we argue that the restoration from low resolution features to high resolution output should be capable of preserving information significant for precise boundary localization. Hence, we use unpooling to unsample and propose SegU-net. Second, to combat the potential vanishing-gradient problem, we introduce dense blocks which can facilitate feature propagation and reuse. Third, using only cross entropy (CE) as loss function may bring about troubles such as miss-prediction, therefore we propose to use a loss function comprised of both CE loss and Dice loss to train the network. Quantitative and qualitative comparisons are carried out extensively on in-house datasets, the experimental results show that our proposed architecture outperforms the existing state-of-the-art segmentation networks.