Automatic Whole-Heart Segmentation in Congenital Heart Disease Using Deeply-Supervised 3D FCN

被引:8
作者
Li, Jinpeng [1 ]
Zhang, Rongzhao [2 ]
Shi, Lin [2 ,3 ]
Wang, Defeng [1 ,4 ]
机构
[1] Chinese Univ Hong Kong, Dept Imaging & Intervent Radiol, Shenzhen, Peoples R China
[2] Chinese Univ Hong Kong, Dept Med & Therapeut, Shenzhen, Peoples R China
[3] Chinese Univ Hong Kong, Chow Yuk Ho Technol Ctr Innovat Med, Shenzhen, Peoples R China
[4] Chinese Univ Hong Kong, Shenzhen Res Inst, Shenzhen, Peoples R China
来源
RECONSTRUCTION, SEGMENTATION, AND ANALYSIS OF MEDICAL IMAGES | 2017年 / 10129卷
基金
中国国家自然科学基金;
关键词
D O I
10.1007/978-3-319-52280-7_11
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Accurate whole-heart segmentation plays an important role in the surgical planning for heart defects such as congenital heart disease (CHD). In this work, we propose a deep learning method for automatic whole-heart segmentation in cardiac magnetic resonance (CMR) images with CHD. First, we start with a 3D fully convolutional network (3D FCN) in order to ensure an efficient voxel-wise labeling. Then we introduce dilated convolutional layers (3D-HOL layers) into the base-line model to expand its receptive field, so as to make better use of the spatial information. Last, we employ deeply-supervised pathways to accelerate training and exploit multi-scale information. We evaluate the proposed method on 3D CMR images from the dataset of the HVSMR 2016 Challenge. The results of controlled experiments demonstrate the efficacy of the proposed 3D-HOL layers and deeply-supervised pathways. We achieve an average Dice score of 80.1% in training (5-fold cross-validation) and 69.5% in testing.
引用
收藏
页码:111 / 118
页数:8
相关论文
共 10 条
[1]  
[Anonymous], 2015, IEEE I CONF COMP VIS, DOI DOI 10.1109/ICCV.2015.123
[2]  
Chen L.-C., 2014, ARXIV
[3]   Automatic Detection of Cerebral Microbleeds From MR Images via 3D Convolutional Neural Networks [J].
Dou, Qi ;
Chen, Hao ;
Yu, Lequan ;
Zhao, Lei ;
Qin, Jing ;
Wang, Defeng ;
Mok, Vincent C. T. ;
Shi, Lin ;
Heng, Pheng-Ann .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2016, 35 (05) :1182-1195
[4]  
Ioffe Sergey, 2015, PROC INT C MACH LEAR, V37, P448, DOI DOI 10.48550/ARXIV.1502.03167
[5]  
Kamnitsas K., ARXIV160305959
[6]  
Lee CY, 2015, JMLR WORKSH CONF PRO, V38, P562
[7]  
Long J, 2015, PROC CVPR IEEE, P3431, DOI 10.1109/CVPR.2015.7298965
[8]   Interactive Whole-Heart Segmentation in Congenital Heart Disease [J].
Pace, Danielle F. ;
Dalca, Adrian V. ;
Geva, Tal ;
Powell, Andrew J. ;
Moghari, Mehdi H. ;
Golland, Polina .
MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION, PT III, 2015, 9351 :80-88
[9]   Multi-scale patch and multi-modality atlases for whole heart segmentation of MRI [J].
Zhuang, Xiahai ;
Shen, Juan .
MEDICAL IMAGE ANALYSIS, 2016, 31 :77-87
[10]   Challenges and Methodologies of Fully Automatic Whole Heart Segmentation: A Review [J].
Zhuang, Xiahai .
JOURNAL OF HEALTHCARE ENGINEERING, 2013, 4 (03) :371-407