基于COLLATE融合多图谱的心脏电影MRI右心室分割

王丽嘉; 苏新宇; 李亚; 胡立伟; 聂生东

doi:10.11938/cjmr20182642

波谱学杂志 >

2018 , Vol. 35 >Issue 4: 407 - 416

DOI: https://doi.org/10.11938/cjmr20182642

研究论文

基于COLLATE融合多图谱的心脏电影MRI右心室分割

王丽嘉 ,
苏新宇 ,
李亚 ,
胡立伟 ,
聂生东

展开

1. 上海理工大学医疗器械与食品学院, 上海 200093;
2. 上海交通大学医学院附属上海儿童医学中心影像诊断中心, 上海 200127

收稿日期: 2018-04-27

网络出版日期: 2018-06-25

基金资助

上海市卫生和计划生育委员会科研课题（20164Y0150）.

收起

Segmentation of Right Ventricle in Cardiac Cine MRI Using COLLATE Fusion-Based Multi-Atlas

WANG Li-jia ,
SU Xin-yu ,
LI Ya ,
HU Li-wei ,
NIE Sheng-dong

Expand

1. School of Medical Instrument and Food Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
2. Department of Diagnostic Imaging Center, Shanghai Children's Medical Center(Shanghai Jiao Tong University School of Medicine), Shanghai 200127, China

Received date: 2018-04-27

Online published: 2018-06-25

Fold

摘要

右心室分割对肺动脉高压等疾病的心功能分析具有重要的临床意义.然而，右心室心肌薄、易变且不规则，其传统的医学图像分割方法仍然未能取得突破性进展.本文提出基于COLLATE（Consensus Level，Labeler Accuracy and Truth Estimation）的多图谱分割方法，首先以归一化互信息为相似测度对目标图像和图谱集进行B样条配准以获取粗分割结果；然后利用COLLATE对粗分割结果进行融合；最后采用基于形状约束的区域生长算法修正出现错误的数据.10例临床心脏磁共振短轴电影图像被用于算法验证.本文还将使用基于COLLATE的多图谱分割方法得到的结果与深度学习算法及手动分割进行了比较.结果显示与深度学习算法比较，使用本文算法得到的射血分数（Ejection Fraction，EF）与手动分割更加一致和相关，表明该算法的分割结果有望辅助临床心脏功能诊断.

关键词： 右心室; 心脏磁共振电影成像; 多图谱分割; COLLATE融合

本文引用格式

王丽嘉 , 苏新宇 , 李亚 , 胡立伟 , 聂生东 . 基于COLLATE融合多图谱的心脏电影MRI右心室分割[J]. 波谱学杂志, 2018 , 35(4) : 407 -416 . DOI: 10.11938/cjmr20182642

Abstract

Cardiac right ventricle (RV) segmentation plays an essential role in the functional analysis of heart diseases, such as pulmonary hypertension. The myocardium of RV is thin and irregular-shaped, making the traditional segmentation methods less effective. To improve RV segmentation, a COLLATE (Consensus Level, Labeler Accuracy and Truth Estimation) fusion-based multi-atlas method was developed. The preprocessed target image was first registered to atlas images with a B-spline algorithm optimizing normalized mutual information. The registration coefficients obtained were then used to get a rough RV segmentation for COLLATE fusion. Shape-constrained region growing algorithm was used to correct the segmentation errors. Ten cardiac magnetic resonance datasets were blindly selected to compare the performance of RV segmentation between the method developed and a method based on deep learning. The results of manual segmentation were used as the golden standard. Ejection fraction (EF) calculated with the proposed segmentation method showed better correlation and consistency with the golden standard, relative to the results calculated with the deep learning method.

Key words： cardiac right ventricle; cardiac cine magnetic resonance imaging; multi-atlas segmentation; COLLATE fusion

参考文献

[1] CHEN W W, GAO R L, LIU L S, et al. Report of 2014 Chinese cardiovascular disease[J]. Chin Circul J, 2015, 30(7):617-622. 陈伟伟, 高润霖, 刘力生, 等.《中国心血管病报告2014》概要[J]. 中国循环杂志, 2015, 30(7):617-622.
[2] HADDAD F, HUNT S A, ROSENTHAL D N, et al. Right ventricular function in cardiovascular disease, part I anatomy, physiology, aging, and functional assessment of the right ventricle[J]. Circulation, 2008, 117(11):1436-1448.
[3] CAUDRON J, FARES J, VIVIER P H, et al. Diagnostic accuracy and variability of three semi-quantitative methods for assessing right ventricular systolic function from cardiac MRI in patients with acquired heart disease[J]. Eur Radiol, 2011, 21(10):2111-2120.
[4] BAUR L H. Magnetic resonance imaging:the preferred imaging method for evaluation of the right ventricle[J]. Int J Cardiovas Imag, 2008, 24(7):699-700.
[5] ARMOUR J, PACE J, RANDALL W. Interrelationship of architecture and function of the right ventricle[J]. Am J Physiol, 2016, 218(1):174-179.
[6] COCOSCO C A, NIESSEN W J, NETSCH T, et al. Automatic image-driven segmentation of the ventricles in cardiac cine MRI[J]. J Magn Reson Imaging, 2008, 28(2):366-374.
[7] LÖTJÖNEN J M, JÄRVINEN V M, CHEONG B, et al. Evaluation of cardiac biventricular segmentation from multiaxis MRI data:A multicenter study[J]. J Magn Reson Imaging, 2008, 28:626-636.
[8] MAHAPATRA D. Cardiac image segmentation from cine cardiac MRI using graph cuts and shape priors[J]. J Digit Imaging, 2013, 26(4):721-730.
[9] ARRIETA C, URIBE S, SING-LONG C, et al. Simultaneous left and right ventricle segmentation using topology preserving level sets[J]. Biomed Signal Proces, 2017, 33:88-95.
[10] ELBAS M S, FAHMY A S. Active shape model with inter-profile modeling paradigm for cardiac right ventricle segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI, 2012, 691-698.
[11] ZHEN X T, WANG Z J, ISLAM A, et al. Multi-scale deep networks and regression forests for direct bi-ventricular volume estimation[J]. Med Image Anal, 2015, 30:120-129.
[12] HAN X, HOOGEMAN M S, LEVENDAG P C, et al. Atlas-based auto-segmentation of head and neck CT images[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2008, Springer, 2008, 434-441.
[13] WARFIELD S K, ZOU K H, WELLS W M. Simultaneous truth and performance level estimation (STAPLE):an algorithm for the validation of image segmentation[J]. IEEE T Med Imaging, 2004, 23(7):903-921.
[14] ASMAN A J, LANDMAN B A. Robust statistical label fusion through consensus level, labeler accuracy, and truth estimation (COLLATE)[J]. IEEE T Med Imaging, 2011, 30(10):1779-1794.
[15] WANG L J, PEI M C, CODELLA N C F, et al. Left ventricle:fully automated segmentation based on spatiotemporal continuity and myocardium information in cine cardiac magnetic resonance imaging (LV-FAST)[J]. Biomed Res Int, 2015, 2015:367583.
[16] ALJABAR P, HECKEMANN R A, Hammers A, et al. Multi-atlas based segmentation of brain images:atlas selection and its effect on accuracy[J]. Neuroimage, 2009, 46(3):726-738.
[17] AWATE S P, ZHU P, WHITAKER R T. How many templates does it take for a good segmentation:error analysis in multiatlas segmentation as a function of database size[C]//Multimodal Brain Image Analysis, Springer, 2012, 103-114.
[18] FOPPA M, ARORA G, GONA P, et al. Right ventricular volumes and systolic function by cardiac magnetic resonance and the impact of sex, age, and obesity in a longitudinally followed cohort free of pulmonary and cardiovascular disease the framingham heart study[J]. Cir Cardiovasc Imaging, 2016, 9(3):e003810.
[19] 孔德红. 实时三维超声心动图评价右心室整体和节段容积及收缩功能的临床与实验研究[D]. 上海:复旦大学, 2012.
[20] ANDERPOOL R R, RISCHARD F, NAEIJE R, et al. Simple functional imaging of the right ventricle in pulmonary hypertension:Can right ventricular ejection fraction be improved?[J]. Int J Cardiol, 2016, 223:93-94.
[21] KYHL K, AHTAROVSKI K A, IVERSEN K, et al. The decrease of cardiac chamber volumes and output during positive-pressure ventilation[J]. Am J Physiol Heart Circ Physiol, 2013, 305(7):H1004-9.
[22] CHILDS H, MA L C, MA M, et al. Comparison of long and short axis quantification of left ventricular volume parameters by cardiovascular magnetic resonance, with ex-vivo validation[J]. J Cardiovasc Magn Reson, 2011, 13:40
[23] LEI X L, LIU H, HAN Y C, et al. Reference values of cardiac ventricular structure and function by steady-state free-procession MRI at 3.0T in healthy adult chinese volunteers[J]. J Magn Reson Imaging, 2017, 45(6):1684-1692.
[24] OLOFSEN E, DAHAN A, BORSBOOM G, et al. Improvements in the application and reporting of advanced Bland-Altman methods of comparison[J]. J Clin Monit Comput, 2015, 29(1):127-139.
[25] TAHA A A, HANBURY A. An efficient algorithm for calculating the exact Hausdorff distance[J]. IEEE T Pattern Anal, 2015, 37(11):2153-2163.
[26] BAI W J, SHI W Z, WANG H Y, et al. Multiatlas based segmentation with local label fusion for right ventricle MR images[J]. Image, 2012, 6:9.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献