Background and objective

Liver segmentation is an essential prerequisite for liver cancer diagnosis and surgical planning. Traditionally, liver contour is delineated manually by radiologist in a slice-by-slice fashion. However, this process is time-consuming and prone to errors depending on radiologist's experience. In this paper, a modified U-Net based framework is presented, which leverages techniques from Squeeze-and-Excitation (SE) block, Atrous Spatial Pyramid Pooling (ASPP) and residual learning for accurate and robust liver Computed Tomography (CT) segmentation, and the effectiveness of the proposed method was tested on two public datasets LiTS17 and SLiver07.

Methods

A new network architecture, called SAR-U-Net was designed, which is grounded in the classical U-Net. Firstly, the SE block is introduced to adaptively extract image features after each convolution in the U-Net encoder, while suppressing irrelevant regions, and highlighting features of specific segmentation task; Secondly, the ASPP is employed to replace the transition layer and the output layer, and acquire multi-scale image information via different receptive fields. Thirdly, to alleviate the gradient vanishment problem, the traditional convolution block is replaced with the residual structures, and thus prompt the network to gain accuracy from considerably increased depth.

Results

In the LiTS17 database experiment, five popular metrics were used for evaluation, including Dice coefficient, VOE, RVD, ASD and MSD. Compared with other closely related models, the proposed method achieved the highest accuracy. In addition, in the experiment of the SLiver07 dataset, compared with other closely related models, the proposed method achieved the highest segmentation accuracy except for the RVD.

Conclusion

An improved U-Net network combining SE, ASPP, and residual structures is developed for automatic liver segmentation from CT images. This new model shows a great improvement on the accuracy compared to other closely related models, and its robustness to challenging problems, including small liver regions, discontinuous liver regions, and fuzzy liver boundaries, is also well demonstrated and validated.

中文翻译：

---

SAR-U-Net：基于挤压和激励块和多孔空间金字塔池化的残差 U-Net，用于计算机断层扫描中的自动肝脏分割

背景和目的

肝脏分割是肝癌诊断和手术计划的必要先决条件。传统上，肝脏轮廓由放射科医生以逐片方式手动描绘。然而，根据放射科医生的经验，此过程耗时且容易出错。在本文中，提出了一种改进的基于 U-Net 的框架，该框架利用来自 Squeeze-and-Excitation (SE) 块、Atrous Spatial Pyramid Pooling (ASPP) 和残差学习的技术，用于准确和稳健的肝脏计算机断层扫描 (CT) 分割，并且在两个公共数据集 LiTS17 和 SLiver07 上测试了所提出方法的有效性。

方法

设计了一种新的网络架构，称为 SAR-U-Net，它以经典的 U-Net 为基础。首先，在U-Net编码器每次卷积后，引入SE块自适应提取图像特征，同时抑制不相关区域，突出特定分割任务的特征；其次，采用ASPP替代过渡层和输出层，通过不同的感受野获取多尺度图像信息。第三，为了缓解梯度消失问题，将传统的卷积块替换为残差结构，从而促使网络从显着增加的深度中获得精度。

结果

在 LiTS17 数据库实验中，使用了五个流行的指标进行评估，包括Dice 系数、VOE、RVD、ASD和MSD。与其他密切相关的模型相比，所提出的方法取得了最高的准确率。此外，在 SLiver07 数据集的实验中，与其他密切相关的模型相比，所提出的方法实现了除RVD之外的最高分割精度。

结论

开发了一种结合 SE、ASPP 和残差结构的改进 U-Net 网络，用于从 CT 图像中自动进行肝脏分割。与其他密切相关的模型相比，这种新模型在准确性上有了很大的提高，并且它对具有挑战性的问题（包括小肝脏区域、不连续肝脏区域和模糊肝脏边界）的鲁棒性也得到了很好的证明和验证。