Automatic 3D liver segmentation based on deep learning and globally optimized surface evolution
Automatic 3D liver segmentation based on deep learning and globally optimized surface evolution
复制标题
基于深度学习和全局优化表面演化的自动 3D 肝脏分割
DOI：
10.1088/1361-6560/61/24/8676
复制
发表时间：
2016-12-21
影响因子：
3.5
通讯作者：
Kong, Dexing
中科院分区：
文献类型：
--
作者：
Hu, Peijun;Wu, Fa;Kong, Dexing
从腹部3D计算机断层扫描（CT）图像中检测和描绘肝脏是计算机辅助肝脏手术规划中的基本任务。然而，自动和准确的分割，特别是肝脏检测，仍然具有挑战性，由于复杂的背景，模糊的边界，异质外观和高度变化的形状的肝脏。为了解决这些困难，我们提出了一个基于3D卷积神经网络（CNN）和全局优化表面进化的自动分割框架。首先，训练深度3D CNN来学习肝脏的特定于对象的概率图，该概率图给出初始表面并在接下来的分割步骤中充当先验形状。然后，全局和局部的外观信息从先前的分割自适应地纳入到一个分割模型，这是全局优化的表面进化的方式。所提出的方法已经验证了42 CT图像从公共Sliver 07数据库和当地医院。在Sliver 07在线测试集上，该方法的总得分为80.3±4.5，平均Dice相似系数为97.25± 0.65%，平均对称面距离为0.84±0.25 mm，定量验证和比较表明，该方法准确有效，可用于临床应用。
The detection and delineation of the liver from abdominal 3D computed tomography (CT) images are fundamental tasks in computer-assisted liver surgery planning. However, automatic and accurate segmentation, especially liver detection, remains challenging due to complex backgrounds, ambiguous boundaries, heterogeneous appearances and highly varied shapes of the liver. To address these difficulties, we propose an automatic segmentation framework based on 3D convolutional neural network (CNN) and globally optimized surface evolution. First, a deep 3D CNN is trained to learn a subject-specific probability map of the liver, which gives the initial surface and acts as a shape prior in the following segmentation step. Then, both global and local appearance information from the prior segmentation are adaptively incorporated into a segmentation model, which is globally optimized in a surface evolution way. The proposed method has been validated on 42 CT images from the public Sliver07 database and local hospitals. On the Sliver07 online testing set, the proposed method can achieve an overall score of 80.3±4.5, yielding a mean Dice similarity coefficient of 97.25±0.65%, and an average symmetric surface distance of 0.84±0.25 mm. The quantitative validations and comparisons show that the proposed method is accurate and effective for clinical application.