Automatic liver segmentation from abdominal CT volumes using improved convolution neural networks
Automatic liver segmentation from abdominal CT volumes using improved convolution neural networks
复制标题
使用改进的卷积神经网络从腹部 CT 体积自动分割肝脏
DOI：
10.1007/s00530-020-00709-x
复制
发表时间：
2020-11-09
影响因子：
3.9
通讯作者：
Sheng, Victor S.
中科院分区：
文献类型：
--
作者：
Liu, Zhe;Han, Kai;Sheng, Victor S.
从腹部CT图像中分割肝脏是计算机辅助诊断和手术计划的重要步骤。U-Net架构是最著名的CNN架构之一，在医学和生物图像分割领域都取得了显著的成功。然而，当目标区域很小或被分割时，它的性能不佳。在本文中，我们提出了一种新的架构，称为密集特征选择U-Net (DFS U-Net)，它解决了这一具有挑战性的问题。具体来说，Hounsfield单位值在一定范围内加窗以排除无关器官，然后使用预处理数据来训练我们提出的DFS U-Net模型。为了进一步提高在训练数据集有限的情况下对兴趣小区域和不连通区域的分割精度，我们通过在公式中加入一个参数来改进损失函数。相对于地面真相，肝脏的骰子得分比可以达到94.9%以上。实验结果表明，该方法具有较高的有效性、鲁棒性和高效性，具有临床应用潜力。
Segmentation of the liver from abdominal CT images is an essential step for computer-aided diagnosis and surgery planning. The U-Net architecture is one of the most well-known CNN architectures which achieved remarkable successes in both medical and biological image segmentation domain. However, it does not perform well when the target area is small or partitioned. In this paper, we propose a novel architecture, called dense feature selection U-Net (DFS U-Net), which addresses this challenging problem. Specifically, The Hounsfield unit values were windowed in a range to exclude irrelevant organs, and then use the pre-processed data to train our proposed DFS U-Net model. To further improve the segmentation accuracy of the small region and disconnected regions of interests with limited training datasets, we improve the loss function by adding a parameter to the formula. With respect to the ground truth, the Dice score ratio can reach over 94.9% for the liver. Our experimental results demonstrate its potential in clinical usage with high effectiveness, robustness and efficiency.