[1]常艳奎,彭昭,周解平,等.基于U-net的心脏自动勾画模型的临床应用及改进[J].中国医学物理学杂志,2020,37(10):1218-1223.[doi:DOI:10.3969/j.issn.1005-202X.2020.10.002]
　CHANG Yankui,PENG Zhao,ZHOU Jieping,et al.Clinical application and improvement of U-net-based model for automatic segmentation of the heart[J].Chinese Journal of Medical Physics,2020,37(10):1218-1223.[doi:DOI:10.3969/j.issn.1005-202X.2020.10.002]
点击复制

基于U-net的心脏自动勾画模型的临床应用及改进()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
37
期数:
2020年第10期
页码:
1218-1223
栏目:
医学放射物理
出版日期:
2020-10-29

文章信息/Info

Title:
Clinical application and improvement of U-net-based model for automatic segmentation of the heart
文章编号:
1005-202X（2020）10-1218-06
作者:
常艳奎1; 彭昭1; 周解平1; 2; 皮一飞3; 吴昊天4; 吴爱东2; 徐榭1; 4; 裴曦1; 4
1.中国科学技术大学放射医学物理中心， 安徽 合肥 230025； 2.中国科学技术大学附属第一医院放疗科， 安徽 合肥 230001； 3.郑州大学第一附属医院放疗科， 河南 郑州 450052； 4.安徽慧软科技有限公司， 安徽 合肥 230088
Author(s):
CHANG Yankui1;  PENG Zhao1;  ZHOU Jieping1;  2;  PI Yifei3;  WU Haotian4;  WU Aidong2;  XU Xie1;  4;  PEI Xi1;  4
1. Center of Radiological Medical Physics, University of Science and Technology of China, Hefei 230025, China 2. Department of Radiation Oncology, the First Affiliated Hospital of University of Science and Technology of China, Hefei 230001, China 3. Department of Radiation Oncology, the First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China 4. Anhui Wisdom Technology Co. Ltd., Hefei 230088, China
关键词:
U-net; 心脏; 自动勾画; 形状相似性系数
Keywords:
U-net heart automatic segmentation Dice similarity coefficient
分类号:
R811.1；R318
DOI:
DOI:10.3969/j.issn.1005-202X.2020.10.002
文献标志码:
A
摘要:
目的：拟分析基于不同医院数据的心脏自动勾画模型在临床应用中的适用性及其改进方法。方法：首先，建立基于U-net和Inception模块的心脏自动勾画网络。其次，收集不同治疗中心的患者数据：中国科学技术大学附属第一医院65例（数据1）、MICCAI2019比赛数据50例（数据2）、数据1和2的混合数据（数据3）、郑州大学第一附属医院50例（数据4）和郑州大学第一附属医院100例（数据5），分别训练得到模型1～5。然后，以郑州大学第一附属医院59例患者作为测试集，使用形状相似性系数（DSC）评估该测试集在不同模型上的分割精度，比较模型之间的差别。最后，将模型3作为心脏预训练模型，采用数据5进行模型再训练，分别测试3组实验（20例/次×5次、10例/次×10次、5例/次×20次）对心脏预训练模型的改进情况。结果：测试集在模型1~5中的平均DSC为0.926、0.932、0.939、0.941和0.950。在再训练过程中，模型在20例/次×5次的实验中表现更稳定。结论：基于不同医院的数据训练模型在心脏自动勾画的临床应用上表现存在差异，使用本地医院数据进行训练的模型预测精度更高。对于非本地数据训练的模型，基于本地数据再训练可以有效提高模型预测的精度，其中以20例/次的再训练方式效果较好。
Abstract:
Abstract: Objective To investigate the clinical applicability of the model established based on the data from different hospitals for the automatic segmentation of the heart and to discuss the methods to improve the model. Methods A network based on U-net and Inception module was firstly constructed for the automatic segmentation of the heart, and the clinical data from different hospitals were collected, including 65 cases from the First Affiliated Hospital of University of Science and Technology of China (data 1), 50 cases from MICCAI2019 match data (data 2), the mixed data of data 1 and data 2 (data 3), 50 cases from the First Affiliated Hospital of Zhengzhou University (data 4) and 100 cases from the First Affiliated Hospital of Zhengzhou University (data 5). The collected data were trained for obtaining 5 different models. Then, with the clinical data of another 59 patients from the First Affiliated Hospital of Zhengzhou University as test set, the segmentation accuracies of test set on different models were evaluated using Dice similarity coefficient (DSC), and the differences in segmentation accuracies among different models were also compared. Finally, model 3 was used as a pre-trained model of the heart, and the model was retrained with data 5. Three groups of experiments (20 cases each time × 5 times, 10 cases each time× 10 times, 5 cases each time× 20 times) were carried out to observe the improvement of the pre-trained model. Results The average DSC of the test set based on models 1 to 5 was 0.926, 0.932, 0.939, 0.941 and 0.950, respectively. During the retraining of the pre-trained model of the heart, the model was more stable in the experiment of 20 cases each time × 5 times. Conclusion The trained model established based on the data from different hospitals has different performances in the automatic segmentation of the heart, and the model trained with local hospital data has a higher prediction accuracy. For the model based on non-local data training, the retraining with local data can effectively improve the accuracy of model prediction, in which the retraining with 20 cases each time has the optimal performance.

相似文献/References:

[1]李丹明,孙新臣,张 胜,等.食道癌患者心脏图谱指导下心脏勾画的一致性研究[J].中国医学物理学杂志,2014,31(03):4857.[doi:10.3969/j.issn.1005-202X.2014.03.003]
[2]苏清秀,张 矛,金海国,等.比较调强放射治疗与三维适形放疗治疗中下段食管癌时心脏和冠状动脉的受量[J].中国医学物理学杂志,2014,31(05):5139.[doi:10.3969/j.issn.1005-202X.2014.05.010]
[3]柏晗,等.基于四维CT下胸部肿瘤调强计划中心脏和食道的剂量评估[J].中国医学物理学杂志,2016,33(3):293.[doi:DOI:10.3969/j.issn.1005-202X.2016.03.015]
　[J].Chinese Journal of Medical Physics,2016,33(10):293.[doi:DOI:10.3969/j.issn.1005-202X.2016.03.015]
[4]黄迎松,吴剑.心腔动态重建的迭代替换算法[J].中国医学物理学杂志,2017,34(1):70.[doi:10.3969/j.issn.1005-202X.2017.01.014]
　[J].Chinese Journal of Medical Physics,2017,34(10):70.[doi:10.3969/j.issn.1005-202X.2017.01.014]
[5]佟颖,尹勇,程品晶,等. 形变配准技术在心脏剂量评估中的应用[J].中国医学物理学杂志,2018,35(5):503.[doi:DOI:10.3969/j.issn.1005-202X.2018.05.002]
　TONG Ying,YIN Yong,CHENG Pinjing,et al. Application of deformable image registration in the evaluation of radiation dose to the heart[J].Chinese Journal of Medical Physics,2018,35(10):503.[doi:DOI:10.3969/j.issn.1005-202X.2018.05.002]
[6]安义均,赵彪,余立丹,等. 胸中段食管癌VMAT与IMRT的心脏及其亚结构剂量学比较[J].中国医学物理学杂志,2018,35(6):648.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.006]
　AN Yijun,ZHAO Biao,YU Lidan,et al. Comparison of doses of heart and substructures in VMAT versus IMRT for middle thoracic esophageal cancer[J].Chinese Journal of Medical Physics,2018,35(10):648.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.006]
[7]周博雅,金丹,严琼,等. 超声在心肌致密化不全患者诊断中的应用及其临床价值[J].中国医学物理学杂志,2019,36(9):1045.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.010]
　ZHOU Boya,JIN Dan,YAN Qiong,et al.Application and clinical value of ultrasound in the diagnosis of noncompaction of the ventricular myocardium[J].Chinese Journal of Medical Physics,2019,36(10):1045.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.010]
[8]时飞跃,王敏,秦伟,等.智能放疗云平台自动勾画食管癌患者心脏结构的应用[J].中国医学物理学杂志,2019,36(12):1377.[doi:DOI:10.3969/j.issn.1005-202X.2019.12.003]
　SHI Feiyue,WANG Min,et al.Application of RAIC.OIS in automatic segmentation of the heart in patients with esophageal cancer[J].Chinese Journal of Medical Physics,2019,36(10):1377.[doi:DOI:10.3969/j.issn.1005-202X.2019.12.003]
[9]秦楠楠,薛旭东,吴爱林,等.基于U-net卷积神经网络的宫颈癌临床靶区和危及器官自动勾画的研究[J].中国医学物理学杂志,2020,37(4):524.[doi:DOI:10.3969/j.issn.1005-202X.2020.04.023]
　QIN Nannan,XUE Xudong,WU Ailin,et al.Automatic segmentation of clinical target volumes and organs-at-risk in radiotherapy for cervical cancer using U-net convolutional neural network[J].Chinese Journal of Medical Physics,2020,37(10):524.[doi:DOI:10.3969/j.issn.1005-202X.2020.04.023]
[10]董国亚,宋立明,李雅芬,等.基于深度学习的跨模态医学图像转换[J].中国医学物理学杂志,2020,37(10):1335.[doi:DOI:10.3969/j.issn.1005-202X.2020.10.021]
　DONG Guoya,SONG Liming,et al.Cross-modality medical image synthesis based on deep learning[J].Chinese Journal of Medical Physics,2020,37(10):1335.[doi:DOI:10.3969/j.issn.1005-202X.2020.10.021]

备注/Memo

备注/Memo:
【收稿日期】2020-04-05 【基金项目】国家自然科学基金（11575180）；安徽省自然科学基金（1908085MA27）；安徽省重点研究与开发计划（1804a09020039） 【作者简介】常艳奎，硕士，主要从事医学图像分割、深度学习等研究，E-mail: cykanhui@mail.ustc.edu.cn 【通信作者】裴曦，博士，副教授，主要从事医学物理、人工智能和医学影像等研究，E-mail: xpei@ustc.edu.cn
更新日期/Last Update: 2020-10-29