Semi-automatic sigmoid colon segmentation in CT for radiation therapy treatment planning via an iterative 2.5-D deep learning approach.
Semi-automatic sigmoid colon segmentation in CT for radiation therapy treatment planning via an iterative 2.5-D deep learning approach.
复制标题
通过迭代2.5-D深度学习方法在CT中进行半自动乙状结肠分割，用于放射治疗治疗计划。
DOI：
10.1016/j.media.2020.101896
复制
发表时间：
2021-03
影响因子：
10.9
通讯作者：
Jia X
中科院分区：
文献类型：
--
作者：
Gonzalez Y;Shen C;Jung H;Nguyen D;Jiang SB;Albuquerque K;Jia X
由于器官形状复杂、与其他器官的距离近、大小、形状和充盈状态变化大，CT中乙状结肠的自动分割在放射治疗计划中是具有挑战性的。患者的肠道经常不排空，也不使用CT增强扫描，这进一步增加了问题的难度。深度学习在许多分割问题中都显示出了其强大的能力。然而，由于几何信息不完全，标准的2-D方法不能处理Sigmoid分割问题，而3-D方法经常遇到训练数据量有限的挑战。受人类在考虑相邻切片之间的连通性的情况下逐个切片分割乙状图的行为的启发，我们提出了一种迭代2.5维动态链接法来解决这个问题。我们构建了一个网络，将轴位CT切片、该切片中的乙状结肠掩模和相邻的CT切片作为输入进行分割，并在相邻的切片上输出预测的掩模。我们还考虑了其他器官面罩作为先验信息。我们使用五次交叉验证对迭代网络进行了训练，其中有50例患者。训练好的网络被重复地应用于逐片生成掩模。该方法在不使用和使用先验信息的10个测试用例上分别获得了0.82 0.06和0.88 0.02的平均Dice相似系数。
Automatic sigmoid colon segmentation in CT for radiotherapy treatment planning is challenging due to complex organ shape, close distances to other organs, and large variations in size, shape, and filling status. The patient bowel is often not evacuated, and CT contrast enhancement is not used, which further increase problem difficulty. Deep learning (DL) has demonstrated its power in many segmentation problems. However, standard 2-D approaches cannot handle the sigmoid segmentation problem due to incomplete geometry information and 3-D approaches often encounters the challenge of a limited training data size. Motivated by human’s behavior that segments the sigmoid slice by slice while considering connectivity between adjacent slices, we proposed an iterative 2.5-D DL approach to solve this problem. We constructed a network that took an axial CT slice, the sigmoid mask in this slice, and an adjacent CT slice to segment as input and output the predicted mask on the adjacent slice. We also considered other organ masks as prior information. We trained the iterative network with 50 patient cases using five-fold cross validation. The trained network was repeatedly applied to generate masks slice by slice. The method achieved average Dice similarity coefficients of 0.82 0.06 and 0.88 0.02 in 10 test cases without and with using prior information.
登录
查看更多内容
影响因子：
4.6
作者：
Lu，Lin;Zhang，Danfeng;Zhao，Jun
通讯作者：
Zhao，Jun
影响因子：
3.5
作者：
Balagopal, Anjali;Kazemifar, Samaneh;Jiang, Steve
通讯作者：
Jiang, Steve
影响因子：
0.9
作者：
Sharma N;Aggarwal LM
通讯作者：
Aggarwal LM
影响因子：
1.2
作者：
Gomez, Laura;Andres, Carlos;Ruiz, Antonio
通讯作者：
Ruiz, Antonio
使用独特曲线引导的全卷积网络进行盆腔器官分割
影响因子：
10.6
作者：
He K;Cao X;Shi Y;Nie D;Gao Y;Shen D
通讯作者：
Shen D