基于深度学习的心脏核磁共振图像自动分割算法

摘要/Abstract

摘要： 心脏核磁共振成像技术由于其无电离辐射的优点已成为医疗诊断中的主要手段。对左心室、右心室以及左心肌进行准确的分割与识别是心脏手术前的重要一步，手动分割心脏结构耗时且易出错，因此自动分割双心室与心肌至关重要。提出了一种能充分利用心脏图像信息的多尺度特征融合U型神经网络MFF U-Net。首先,选择以U-Net++作为网络基本框架。其次，为了提高特征复用率，解决网络深度增加导致的过拟合问题，在U-Net++的编码部分提出了密集残差模块，使得网络在下采样过程中学习到更多有用特征。此外，在解码部分，为了使网络的分割结果更加符合目标器官之间的物理特征，用多个卷积核来扩大感受野并利用长距离依赖模块共享全局上下文信息，使得网络在编码还原的过程中尽可能地获取到目标器官之间的关系信息，从而使得分割结果更为精准。最后，考虑到双心室与左心肌的连贯性与唯一性，还添加了获取最大连通域与填充细小孔洞的后处理操作。采用的实验数据为ACDC心脏分割挑战数据集，其包含150位志愿者收缩期末期与舒张期末期的短轴心脏磁共振图像。在该数据集的测试集上进行验证，并通过在线提交的方式获取实验结果。实验结果表明，相较于其他算法，所提出的算法能够有效地分割目标器官,特别是舒张期末期的Dice系数分别达到了左心室0.96、右心室0.94和左心肌0.89，收缩期末期的分割精度达到了0.87,0.86和0.89。

关键词: 图像处理, 医学图像, 双心室与心肌, 核磁共振图像, 深度学习

Abstract: Because of its advantages of no ionizing radiation, Cardiac Cine Magnetic Resonance Imag- ing (CMRI) has become the main method in medical diagnosis, and the accurate classification and re- cognition of left ventricular, right ventricular and left myocardium is an important step before the heart surgery. However, manual segmentation of cardiac structure time-consuming and error-prone, so automatic segmentation of both ventricular and myocardium is crucial. Firstly, U-Net++ is selected as the basic network framework. Secondly, in order to improve the feature reuse rate and solve the overfitting issue caused by the increase of network depth, a dense residual module is proposed in the encoding part of U-Net++, so that more features can be learned during the network down-sampling process. In addition, in the decoding part, in order to make the network segmentation results more in line with the target organ between the physical characteristics, multiple convolution kernels are used to expand the receptive field and a long distance dependent module is used share the global context information to make the network in the decoding process as much as possible to get the relationship information between the target organs and make the segmentation result more accurate. Finally, considering the consistency and uniqueness between the biventricle and the left myocardium, the post-treatment operation of obtaining the maximum connectivity domain and filling the small holes is added. The experimental data used in this paper is the ACDC Cardiac Segmentation Challenge data set, which includes SHORT axial MRI images of 150 volunteers at the end of systolic and diastolic phases. In this paper, the test set of this data set is verified, and the experimental results are obtained by online submission. Experimental results show that, compared with other methods, the proposed method can effectively segment the target organ. Specially, the Dice coefficient at the end of diastolic period reaches 0.96, 0.94 and 0.89 respectively in the left ventricle, right ventricle and left myocardium, and the segmentation precision at the end of systolic period reaches 0.87, 0.86 and 0.89 respectively.

Key words: image processing, medical image；bi-ventricles and myocardium, magnetic resonance image, deep learning

刘从军, 徐佳陈, 肖志勇, 柴志雷. 基于深度学习的心脏核磁共振图像自动分割算法[J]. 计算机工程与科学, 2022, 44(09): 1646-1654.

LIU Cong-jun, XU Jia-chen, XIAO Zhi-yong, CHAI Zhi-lei. An automatic cardiac magnetic resonance image segmentation algorithm based on deep learning[J]. Computer Engineering & Science, 2022, 44(09): 1646-1654.

参考文献［6］

［1］	Romaguera L V,Costa M G F,Romero F P,et al.Left ventricle segmentation in cardiac MRI images using fully convolutional neural networks［C］∥Proc of the 40th Annual International Conference on the IEEE Engineering in Medicine and Biology Society,2018:1275-1278.
［2］	Ronneberger O,Fischer P,Brox T.U-Net:Convolutional networks for biomedical image segmentation［C］∥Proc of International Conference on Medical Image Computing and Computer-Assisted Intervention,2015:234-241.
［3］	Zhang J,Jin Y,Xu J,et al.MDU-NET:Multi-scale densely connected U-Net for biomedical image segmentation［J］.arXiv:1812.00352,2018.
［4］	Chen W,Zhang Y,He J,et al.Prostate segmentation using 2D bridged U-Net［C］∥Proc of 2019 International Joint Conference on Neural Networks,2019:1-7.
［5］	Oktay O, Schlemper J,Folgoc L L,et al.Attention U-Net:Learning where to look for the pancreas［J］.arXiv:1804.03999,2018.
［6］	Sun Jin-feng, Ding Hui, Wang Guang-zhi.Automatic segmentation of hepatic vein and portal vein based on W-Net［J］.Chinese Journal of Biomedical Engineering,2019,38(5):513-521.(in Chinese)
［7］	Milletari F,Navab N,Ahmadi S A.V-Net:Fully convolutional neural networks for volumetric medical image segmentation［C］∥Proc of 2016 4th International Conference on 3D Vision,2016:565-571.
［8］	Isensee F,Petersen J,Klein A,et al.nnU-Net:Self-adapting framework for U-Net-based medical image segmentation［J］.arXiv:1809.10486,2018.
［9］	Huang C,Han H,Yao Q,et al.3D U2-Net:A 3D universal U-Net for multi-domain medical image segmentation［C］∥Proc of the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention,2019:291-299.
［10］	Baumgartner C F,Koch L M,Pollefeys M,et al.An exploration of 2D and 3D deep learning techniques for cardiac MR image segmentation［C］∥Proc of International Workshop on Statistical Atlases and Computational Models of the Heart,2017:111-119.
［11］	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.
［12］	Zhou Z,Siddiquee M M R,Tajbakhsh N,et al.UNet++:A nested U-Net architecture for medical image segmentation［C］∥Proc of International Workshop on Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support,2018:3-11.
［13］	He K,Zhang X,Ren S,et al.Deep residual learning for image recognition［C］∥Proc of the IEEE Conference on Computer Vision and Pattern Recognition,2016:770-778.
［14］	Huang G,Liu Z,Van Der Maaten L,et al.Densely connected convolutional networks［C］∥Proc of the IEEE Confe- rence on Computer Vision and Pattern Recognition,2017:4700-4708.
［15］	Cao Y,Xu J,Lin S,et al.GCNet:Non-local networks meet squeeze-excitation networks and beyond［C］∥Proc of 2019 IEEE/CVF International Conference on Computer Vision Workshops,2019:1971-1980.
［16］	Simonyan K, Zisserman A.Very deep convolutional networks for large-scale image recognition［J］.arXiv:1409.1556,2014.
［17］	Zhang Y,Tian Y,Kong Y,et al.Residual dense network for image super-resolution［C］∥Proc of the IEEE Conference on Computer Vision and Pattern Recognition,2018:2472-2481.
［18］	Ba J L,Kiros J R,Hinton G E.Layer normalization［J］.arXiv:1607.06450,2016.
［19］	Kingma D P,Ba J.Adam:A method for stochastic optimization［J］.arXiv:1412.6980,2014.
［20］	Grinias E,Tziritas G.Fast fully-automatic cardiac segmentation in MRI using MRF model optimization,substructures tracking and B-spline smoothing［C］∥Proc of International Workshop on Statistical Atlases and Computational Models of the Heart,2017:91-100.
［21］	Wolterink J M,Leiner T,Viergever M A,et al.Automatic segmentation and disease classification using cardiac cine MR images［C］∥Proc of International Workshop on Statistical Atlases and Computational Models of the Heart,2017:101-110.
［22］	Khened M, Alex V,Krishnamurthi G.Densely connected fully convolutional network for short-axis cardiac cine MR image segmentation and heart diagnosis using random forest［C］∥Proc of International Workshop on Statistical Atlases and Computational Models of the Heart,2017:140-151.
［23］	Chang Y,Song B,Jung C,et al.Automatic segmentation and cardiopathy classification in cardiac MRI images based on deep neural networks［C］∥Proc of 2018 IEEE International Conference on Acoustics,Speech and Signal Processing,2018:1020-1024.
［24］	Redmon J,Divvala S,Girshick R,et al.You only look once:Unified,real-time object detection［C］∥Proc of the IEEE Conference on Computer Vision and Pattern Recognition,2016:779-788.
［25］	Liu T,Tian Y,Zhao S,et al.Residual convolutional neural network for cardiac image segmentation and heart disease diagnosis［J］.IEEE Access,2020,8:82153-82161.
	附中文参考文献：
［6］	孙锦峰,丁辉,王广志.基于W-Net的肝静脉和肝门静脉全自动分割［J］.中国生物医学工程学报,2019,38(5):513-521.

[1]	马思远, 焦佳辉, 任晟岐, 宋伟. 基于注意力机制的城市多元空气质量数据缺失值填充[J]. 计算机工程与科学, 2023, 45(08): 1354-1364.
[2]	邓姗姗, 黄慧, 马燕. 基于改进Faster R-CNN的小目标检测算法[J]. 计算机工程与科学, 2023, 45(05): 869-877.
[3]	马冬梅, 黄欣悦, 李煜. 基于特征融合和注意力机制的图像语义分割[J]. 计算机工程与科学, 2023, 45(03): 495-503.
[4]	董佩杰, 牛新, 魏自勉, 陈学晖. 单次神经网络结构搜索研究综述[J]. 计算机工程与科学, 2023, 45(02): 191-203.
[5]	何涛, 施慧莉, 李大亮. 基于深度学习的SAR目标识别DSP设计[J]. 计算机工程与科学, 2022, 44(08): 1357-1363.
[6]	苟淞, 赵绪言, 侯松, 李威. 基于多尺度优化感知网络的口罩检测方法[J]. 计算机工程与科学, 2022, 44(08): 1440-1448.
[7]	刘李漫, 谭龙雨, 彭源, 刘佳. 基于全融合网络的三维点云语义分割[J]. 计算机工程与科学, 2022, 44(05): 862-869.
[8]	刘榕, 伍欣, 敖斌, 文青, 李宽. 用于CD56图像分割的细胞标注精细化与自适应加权损失[J]. 计算机工程与科学, 2022, 44(05): 870-878.
[9]	卢凯良. 基于可见光视觉图像的路面裂缝识别深度学习方法述评[J]. 计算机工程与科学, 2022, 44(04): 674-685.
[10]	刘云, 郑文凤, 张轶. 代价约束算法对入侵检测特征提取的优化研究[J]. 计算机工程与科学, 2022, 44(03): 447-453.
[11]	朱天宇，全惠敏，刘国才. 联合弯曲能量和标志点对应约束的非刚性医学图像配准[J]. 计算机工程与科学, 2019, 41(05): 851-857.
[12]	张娟,蒋芸,胡学伟,肖吉泽. 基于卷积受限玻尔兹曼机的医学图像分类新方法[J]. 计算机工程与科学, 2017, 39(02): 323-329.
[13]	胡学伟，蒋芸，邹丽，李志磊，沈健. 基于邻域关系模糊粗糙集的医学图像分类研究[J]. J4, 2016, 38(04): 739-746.
[14]	翦鹏1，陈志刚1，邓小鸿1,2，梁涤青1，黄伟琦1. 基于预测误差修改的医学图像数据隐藏方法[J]. J4, 2012, 34(12): 105-109.
[15]	李自胜1,2,肖晓萍3,龚〓伟1,2,李光明1,2. 医学影像学诊断报告书智能化书写方法研究[J]. J4, 2011, 33(4): 186-191.