基于方向约束强化学习的左心室内膜分割

doi:10.12052/gdutxb.230156

广东工业大学学报

• •

基于方向约束强化学习的左心室内膜分割

曾安¹, 庞耀幸¹, 潘丹², 赵靖亮¹

1. 广东工业大学计算机学院, 广东广州 510006;
2. 广东技术师范大学电子与信息学院, 广东广州 510665

收稿日期:2023-10-07 出版日期:2024-09-27 发布日期:2024-09-27
通信作者: 庞耀幸(2000–)，男，硕士研究生，主要研究方向为人工智能与疾病辅助诊断，E-mail：2603509767@qq.com
作者简介:曾安(1978–)，女，教授，博士生导师，主要研究方向为图像处理、模式识别、人工智能，E-mail：zengan@gdut.edu.cn
基金资助:
国家自然科学基金资助项目(61976058，92267107) ；广东省重点领域研发计划项目(2021B0101220006) ；广东省科技计划项目(2019A050510041) ；广东省自然科学基金资助项目(2021A1515012300) ；广州市科技计划项目(202103000034)

Segmentation of Left Ventricular Endocardium Using Direction-constrained Reinforcement Learning

Zeng An¹, Pang Yao-xing¹, Pan Dan², Zhao Jing-liang¹

1. School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, China;
2. School of Electronics and Information Engineering, Guangdong University of Technology and Education, Guangzhou 510665, China

Received:2023-10-07 Online:2024-09-27 Published:2024-09-27

摘要/Abstract

摘要： 从心脏核磁共振成像中准确分割左心室内膜，进而得到左心室区域，是对心脏功能进行分析的重要步骤。针对强化学习通过定位左心室内膜边缘来进行分割时容易出现定位偏差，导致分割效果下降的问题，本文提出一种基于方向约束强化学习的左心室内膜分割方法。方法将分割任务划分为两个阶段：在第一阶段提取内膜的全局边缘特征；在第二阶段利用强化学习迭代定位内膜边缘点得到边缘，从而分割出左心室内膜。方法对智能体的定位方向进行约束，能减少定位偏差和重叠的情况从而提高分割精度。最后，在两个公共数据集自动心脏诊断挑战(Automatic Cardiac Diagnosis Challenge, ACDC)和Sunnybrook心脏MR左心室分割挑战(Sunnybrook Cardiac MR Left Ventricle Segmentation Challenge, Sunnybrook)上的实验结果显示：与其他方法相比，本文方法精度较高，其F1-score指标分别为0.9482、0.9387，平均垂直距离(Average Perpendicular Distance, APD)分别为3.5863、4.9447，能有效地分割左心室内膜。

关键词: 图像分割, Transformer, 深度强化学习, 边缘定位

Abstract: Accurate segmentation of the left ventricular endocardium from cardiac magnetic resonance imaging to obtain the left ventricular region is an important step in the analysis of cardiac function. It is noted that reinforcement learning is prone to localization deviation in left ventricular endocardium segmentation by locating the left ventricular endocardial edges, leading to a performance decrease in the segmentation. To address this, this paper proposes a direction-constrained reinforcement learning method for left ventricular endocardium segmentation, which divides the segmentation task into two stages. In the first stage, the proposed method extracts global edge features of the endocardium, and in the second stage, reinforcement learning is used to iteratively locate the endocardial edge points to obtain the edge, obtaining the segmented left ventricular endocardium. The proposed method constrains the direction of agent positioning, which can reduce the localization deviation and overlap, such that the segmentation accuracy can be improved. Finally, the experimental results on two public datasets, including the Automated Cardiac Diagnosis Challenge (ACDC) and Sunnybrook Cardiac MR Left Ventricle Segmentation Challenge (Sunnybrook) , show that the proposed method has higher accuracy than the compared methods. Specifically, the F1-score of the proposed method are 0.9482 and 0.9387, and the Average perpendicular distance (APD) are 3.5863 and 4.9447, which can effectively segment the left ventricular endocardium.

Key words: image segmentation, Transformer, deep reinforcement learning, edge localization

中图分类号:

TP389.1

曾安, 庞耀幸, 潘丹, 赵靖亮. 基于方向约束强化学习的左心室内膜分割[J]. 广东工业大学学报,doi: 10.12052/gdutxb.230156

Zeng An, Pang Yao-xing, Pan Dan, Zhao Jing-liang. Segmentation of Left Ventricular Endocardium Using Direction-constrained Reinforcement Learning[J]. Journal of Guangdong University of Technology, 2024, 41(0): 0-.doi: 10.12052/gdutxb.230156

参考文献

[1] 中国心血管健康与疾病报告2022概要[J]. 中国循环杂志, 2023, 38(6) : 583-612.
[2] CHEN C, QIN C, QIU H Q, et al. Deep learning for cardiac image segmentation: a review [J]. Frontiers in Cardiovascular Medicine, 2020, 7: 25-43.
[3] XIONG J J, Po L M, CHEUNG K W, et al. Edge-sensitive left ventricle segmentation using deep reinforcement learning [J]. Sensors, 2021, 21(7): 1-19.
[4] ZHOU Z W, RAHMAN SIDDIQUEE M M, TAJBAKHSH N, et al. Unet++: a nested U-net architecture for medical image segmentation[C]//Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. [S. l. ]: Springer Cham, 2018: 3-11.
[5] RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation[C]//Proceedings of International Conference on Medical Image Computing and Computer Assisted Intervention. New York: Springer, 2015: 234–241.
[6] SCHLEMPER J, OKTAY O, SCHAAP M, et al. Attention gated networks: learning to leverage salient regions in medical images [J]. Medical Image Analysis, 2019, 53: 197-207.
[7] ABDELTAWAB H, KHALIFA F, TAHER F, et al. A deep learning-based approach for automatic segmentation and quantification of the left ventricle from cardiac cine MR images [J]. Computerized Medical Imaging and Graphics, 2020, 81: 1-11.
[8] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of Neural Information Processing Systems. Long Beach: MIT Press, 2017: 5998-6008.
[9] CAO H, WANG Y Y, CHEN J, et al. Swin-unet: Unet-like pure transformer for medical image segmentation[C]//European Conference on Computer Vision. Cham: Springer Nature Switzerland, 2022: 205-218.
[10] ZHOU H Y, GUO J, ZHANG Y H, et al. Nnformer: interleaved transformer for volumetric segmentation[EB/OL]. arXiv: 2109.03201(2022-02-04) [2023-09-30]. https://arxiv.org/abs/2109.03201.
[11] TRAGAKIS A, KAUL C, MURRAY-SMITH R, et al. The fully convolutional transformer for medical image segmentation[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. Waikoloa: IEEE, 2023: 3649-3658.
[12] 梁礼明, 何安军, 朱晨锟, 等. 融合Transformer和跨级相位感知的结肠息肉分割方法[J]. 生物医学工程学杂志, 2023, 40(2): 234-243.
LIANG L M, HE A J, ZHU C K, et al. Colorectal polyp segmentation method based on fusion of transformer and cross-level phase awareness [J]. Journal of Biomedical Engineering, 2023, 40(2): 234-243.
[13] STEMBER J N, SHALU H. Unsupervised deep clustering and reinforcement learning can accurately segment MRI brain tumors with very small training sets[C]//International Symposium on Intelligent Informatics. Singapore: Springer Nature Singapore, 2022: 255-270.
[14] STEMBER J N, SHALU H. Deep reinforcement learning-based image classification achieves perfect testing set accuracy for MRI brain tumors with a training set of only 30 images [EB/OL]. arXiv: 2102.02895(2022-05-19) [2023-09-30]. https://arxiv.org/abs/2102.02895.
[15] ZHOU S K, LE H N, LUU K, et al. Deep reinforcement learning in medical imaging: a literature review [J]. Medical Image Analysis, 2021, 73: 1-20.
[16] MORTAZI A, BAGCI U. Automatically designing CNN architectures for medical image segmentation[C]//Machine Learning in Medical Imaging: 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018. Granada: Springer International Publishing, 2018: 98-106.
[17] QIN T X, WANG Z Y, HE K L, et al. Automatic data augmentation via deep reinforcement learning for effective kidney tumor segmentation[C]// ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . Barcelona: IEEE, 2020: 1419-1423.
[18] LIAO X, LI W H, XU Q, et al. Iteratively-refined interactive 3D medical image segmentation with multi-agent reinforcement learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: CVF, 2020: 9394-9402.
[19] BERNARD O, LALANDE A, ZOTTI C, et al. Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? [J]. IEEE Transactions on Medical Imaging, 2018, 37(11): 2514-2525.
[20] RADAU P, LU Y, CONNELLY K, et al. Evaluation framework for algorithms segmenting short axis cardiac MRI [J]. The MIDAS Journal, 2009, 7: 100-107.
[21] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 770-778.
[22] CHEN J, LU Y, YU Q, et al. Transunet: transformers make strong encoders for medical image segmentation[EB/OL]. arXiv: 2102.04306(2021-02-08) [2023-09-30]. https://arxiv.org/abs/2102.04306.
[23] HUANG Y, WEI G L, WANG Y X. V-D D3QN: the variant of double deep q-learning network with dueling architecture[C]//2018 37th Chinese Control Conference (CCC) . Wuhan: IEEE, 2018: 9130-9135.
[24] VAN HASSELT H, GUEZ A, SILVER D. Deep reinforcement learning with double q-learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Phoenix: AAAI Press, 2016: 2094–2100.
[25] WANG Z Y, SCHAUL T, HESSEL M, et al. Dueling network architectures for deep reinforcement learning[C]//International Conference on Machine Learning. New York: JMLR Organization, 2016: 1995-2003.
[26] ZHANG Z J, FU H Z, DAI H, et al. ET-Net: a generic edge-attention guidance network for medical image segmentation[C]//Medical Image Computing and Computer Assisted Intervention-MICCAI 2019: 22nd International Conference. Shenzhen: Springer International Publishing, 2019: 442-450.
[27] SHEELA C J J, SUGANTHI G. Morphological edge detection and brain tumor segmentation in Magnetic Resonance (MR) images based on region growing and performance evaluation of modified Fuzzy C-Means (FCM) algorithm [J]. Multimedia Tools and Applications, 2020, 79: 17483-17496.
[28] SHEELA C J J, SUGANTHI G. Brain tumor segmentation with radius contraction and expansion based initial contour detection for active contour model [J]. Multimedia Tools and Applications, 2020, 79: 23793-23819.
[29] KINGMA D, Ba J. Adam: a method for stochastic optimization[EB/OL]. arXiv: 1412.6980(2017-01-30) [2023-09-30]. https://arxiv.org/abs/1412.6980.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于方向约束强化学习的左心室内膜分割

Segmentation of Left Ventricular Endocardium Using Direction-constrained Reinforcement Learning

HTML

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	冯广, 鲍龙. 基于红外可见光融合的复杂环境下人脸识别方法[J]. 广东工业大学学报, 2024, 41(03): 62-70,109.
[2]	郭傲, 许柏炎, 蔡瑞初, 郝志峰. 基于时序对齐的风格控制语音合成算法[J]. 广东工业大学学报, 2024, 41(02): 84-92.
[3]	曾安, 赖峻浩, 杨宝瑶, 潘丹. 基于多尺度卷积和注意力机制的病理图像分割网络[J]. 广东工业大学学报, 2024, 41(0): 0-.
[4]	刘诚辉, 李光平. 基于可分离Transformer的点云分类方法[J]. 广东工业大学学报, 2024, 41(0): 0-.
[5]	赖志茂, 章云, 李东. 基于Transformer的人脸深度伪造检测技术综述[J]. 广东工业大学学报, 2023, 40(06): 155-167.
[6]	甘孟坤, 曾安, 张小波. 基于Swin-Unet的主动脉再缩窄预测研究[J]. 广东工业大学学报, 2023, 40(05): 34-40.
[7]	苏天赐, 何梓楠, 崔苗, 张广驰. 多无人机辅助数据收集系统的智能路径规划算法[J]. 广东工业大学学报, 2023, 40(04): 77-84.
[8]	何一汕, 王永华, 万频, 王磊, 伍文韬. 面向多用户动态频谱接入的改进双深度Q网络方法研究[J]. 广东工业大学学报, 2023, 40(04): 85-93.
[9]	郭心德, 丁宏强. 离散制造智能工厂场景的AGV路径规划方法[J]. 广东工业大学学报, 2021, 38(06): 70-76.
[10]	叶伟杰, 高军礼, 蒋丰, 郭靖. 一种提升机器人强化学习开发效率的训练模式研究[J]. 广东工业大学学报, 2020, 37(05): 46-50.
[11]	吴运雄, 曾碧. 基于深度强化学习的移动机器人轨迹跟踪和动态避障[J]. 广东工业大学学报, 2019, 36(01): 42-50.
[12]	陈志惠, 汪仁煌, 汪志敏. 一种改进的CV水平集图像分割方法[J]. 广东工业大学学报, 2015, 32(2): 104-108.
[13]	王振友, 郑少杰,何远才,林洁芝,苏娜,龙洁丽. 一种PET图像感兴趣区域体积特征的计算方法[J]. 广东工业大学学报, 2013, 30(3): 65-69.
[14]	刘丹；汪仁煌；高延增；刘洪江； . 基于高斯函数的LED彩色图像分割方法[J]. 广东工业大学学报, 2007, 24(4): 73-75.
[15]	张海舰；成思源；骆少明；丁炜； . 基于动态规划法的B样条主动轮廓模型[J]. 广东工业大学学报, 2005, 22(4): 26-30.