结合生物力学模型重建左心室位移场

doi:10.11938/cjmr20212898

摘要/Abstract

摘要：

左心室心肌最为发达，心肌收缩产生的高压将动脉血泵入全身，集中体现了心脏的泵血能力.定量分析左心室收缩运动是诊断心血管疾病（如心肌梗死）的重要途径.本文采用描述左心室心肌材质的生物力学模型重建左心室位移场.该力学模型作为插值项，与心脏电影磁共振图像的观测位移场共同纳入贝叶斯估计框架，并采用有限元法求解位移场方程.实验比较了左心室射血无力组（46例）与正常组（55例）的左心室功能参数，发现两组在径向和圆周方向的位移、速度、应变和应变率都具有非常显著的差异（p < 0.001），这证明本文方法能够有效区别左心室运动正常与否.实验结果还与CVI软件测量的左心室功能参数具有较高的相关性，说明本文方法有望辅助心血管疾病的临床诊断.

关键词: 左心室, 位移场, 生物力学模型, 心脏电影磁共振图像

Abstract:

The left myocardium is the strongest cardiac muscle, representing the ability of heart to pump arterial blood throughout the body. Quantitative analysis of the contractive motion of left ventricle (LV) stands an important approach to diagnose cardiovascular diseases such as myocardial infarction. This paper utilized a biomechanical model describing the material of left myocardium to reconstruct the displacement field of LV. The mechanical model was used as an interpolation term, which was incorporated into the Bayesian estimation framework with the displacement field observed by cardiac cine magnetic resonance (CCMR) image, and the equation of displacement field was solved by finite element method. Functional parameters of LV were compared between the weak (46 cases) and normal (55 cases) ejection fraction groups of LV in the experiment. Significant differences in radial and circumferential displacement and velocity, strain and strain rate were observed between the two groups (p < 0.001), which proved that the method could effectively distinguish whether the motion of LV is normal or not. The experimental results are also highly correlated with the functional parameters of LV measured by CVI software, indicating that the method will facilitate the diagnosis of cardiovascular diseases.

Key words: left ventricle, displacement field, biomechanical model, cardiac cine magnetic resonance image

中图分类号:

TP391

林剑圣,王丽嘉. 结合生物力学模型重建左心室位移场[J]. 波谱学杂志, 2022, 39(1): 72-86.

Jian-sheng LIN,Li-jia WANG. Reconstruction of Displacement Field of Left Ventricle Combined with Biomechanical Model[J]. Chinese Journal of Magnetic Resonance, 2022, 39(1): 72-86.

图/表 11

图1

图2

图3

图4

图5

图6

表1

图7

图8

图9

表2

参考文献 32

1	DEMARINIS S . Cancer overtakes cardiovascular disease as leading cause of death in wealthy nations[J]. Explore, 2020, 16 (1): 6- 7. doi: 10.1016/j.explore.2019.11.003
2	PANG Y , LYU J , YU C , et al. Risk factors for cardiovascular disease in the Chinese population: recent progress and implications[J]. Glo Hea J, 2020, 4 (3): 65- 71. doi: 10.1016/j.glohj.2020.08.004
3	FRANGI A F , NIESSEN W J , VIERGEVER M A . Three-dimensional modeling for functional analysis of cardiac images: a review[J]. IEEE Trans Med Imaging, 2001, 20 (1): 2- 5. doi: 10.1109/42.906421
4	WANG H , AMINI A A . Cardiac motion and deformation recovery from MRI: A Review[J]. IEEE Trans Med Imaging, 2012, 31 (2): 487- 503. doi: 10.1109/TMI.2011.2171706
5	ZERHOUNI E A , PARISH D M , ROGERS W J , et al. Human heart: Tagging with MR imaging-A method for noninvasive assessment of myocardial motion[J]. Radiology, 1988, 169 (1): 59- 63. doi: 10.1148/radiology.169.1.3420283
6	CHEN T , WANG X , CHUNG S , et al. Automated 3D motion tracking using Gabor filter bank, robust point matching, and deformable models[J]. IEEE Trans Med Imaging, 2009, 29 (1): 1- 11.
7	LI M , GUPTA H , LLOYD S G , et al. A graph theoretic approach for computing 3D+ time biventricular cardiac strain from tagged MRI data[J]. Med Image Anal, 2017, 35, 46- 57. doi: 10.1016/j.media.2016.06.006
8	YU Y , ZHANG S T , LI K , et al. Deformable models with sparsity constraints for cardiac motion analysis[J]. Med Image Anal, 2014, 18 (6): 927- 937. doi: 10.1016/j.media.2014.03.002
9	TSADOK Y , FRIEDMAN Z , HALUSKA B A , et al. Myocardial strain assessment by cine cardiac magnetic resonance imaging using non-rigid registration[J]. Magn Reson Imaging, 2016, 34 (4): 381- 390. doi: 10.1016/j.mri.2015.12.035
10	SHI P , SINUSAS A J , CONSTABLE R T , et al. Volumetric deformation analysis using mechanics-based data fusion: applications in cardiac motion recovery[J]. Int J Comput Vision, 1999, 35 (1): 87- 107. doi: 10.1023/A:1008163112590
11	REMME E W , AUGENSTEIN K F , YOUNG A A , et al. Parameter distribution models for estimation of population based left ventricular deformation using sparse fiducial markers[J]. IEEE Trans Med Imaging, 2005, 24 (3): 381- 388. doi: 10.1109/TMI.2004.842458
12	VERESS A I , GULLBERG G T , WEISS J A . Measurement of strain in the left ventricle during diastole with cine-MRI and deformable image registration[J]. J Biomech Eng, 2005, 127 (7): 1195- 1207. doi: 10.1115/1.2073677
13	BISTOQUET A , SKRINJAR O M . Left ventricular deformation recovery from cine MRI using a 4d incompressible model[J]. IEEE Trans Med Imaging, 2007, 26 (9): 1136- 1153. doi: 10.1109/TMI.2007.903693
14	LIU H F , SHI P C . State-space analysis of cardiac motion with biomechanical constraints[J]. IEEE T Image Process, 2007, 16 (4): 901- 917. doi: 10.1109/TIP.2007.891773
15	WONG K C L , WANG L W , ZHANG H Y , et al. Physiological fusion of functional and structural images for cardiac deformation recovery[J]. IEEE Trans Med Imaging, 2011, 30 (4): 990- 1000. doi: 10.1109/TMI.2011.2105274
16	QIAO M Y , WANG Y Y , GUO Y , et al. Temporally coherent cardiac motion tracking from cine MRI: traditional registration method and modern CNN method[J]. Med Phys, 2020, 47 (9): 4189- 4198. doi: 10.1002/mp.14341
17	TANG J Y, GAN Z Y, YANG X. Cardiac motion tracking in short-axis MRI using siamese convolution network[C]//2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE, 2019: 865-870.
18	ZHANG N , YANG G , GAO Z F , et al. Deep learning for diagnosis of chronic myocardial infarction on nonenhanced cardiac cine MRI[J]. Radiology, 2019, 291 (3): 606- 617. doi: 10.1148/radiol.2019182304
19	ZHENG Q , DELINGETTE H , AYACHE N . Explainable cardiac pathology classification on cine MRI with motion characterization by semi-supervised learning of apparent flow[J]. Med Image Anal, 2019, 56, 80- 95. doi: 10.1016/j.media.2019.06.001
20	MANTILLA J J , PAREDES J L , BELLANGER J J , et al. Discriminative dictionary learning for local LV wall motion classification in cardiac MRI[J]. Expert Syst Appl, 2019, 129, 286- 295. doi: 10.1016/j.eswa.2019.04.010
21	SEDERBERG T W , PARRY S R . Free-form deformation of solid geometric models[J]. Siggraph, 1986, 20 (4): 151- 160. doi: 10.1145/15886.15903
22	COOTES T F , TAYLOR C J , COOPER D H , et al. Active shape models-their training and application[J]. Comput Vis Image Underst, 1995, 61 (1): 38- 59. doi: 10.1006/cviu.1995.1004
23	GONG J C , WANG Y , WANG Y J . A method for segmentation of glioma on multimodal magnetic resonance images based on wavelet fusion and deep learning[J]. Chinese J Magn Reson, 2020, 37 (2): 131- 143.
	宫进昌, 王宇, 王远军. 结合小波融合和深度学习的脑胶质瘤自动分割[J]. 波谱学杂志, 2020, 37 (2): 131- 143.
24	LIU P , ZHONG Y M , WANG L J . Automatic segmentation of right ventricle in cine cardiac magnetic resonance image based on a dense and multi-scale U-net method[J]. Chinese J Magn Reson, 2020, 37 (4): 456- 468.
	刘鹏, 钟玉敏, 王丽嘉. 基于密集多尺度U-net网络的电影心脏磁共振图像右心室自动分割[J]. 波谱学杂志, 2020, 37 (4): 456- 468.
25	GUCCIONE J M, MCCULLOCH A D. Finite element modeling of ventricular mechanics[M]. Springer New York, 1991.
26	PAPADEMETRIS X, SHI P C, DIONE D P, et al. Recovery of soft tissue object deformation from 3D image sequences using biomechanical models[C]//Proceedings of the 16th International Conference on Information Processing in Medical Imaging. Berlin, Heidelberg: Springer, 2000.
27	SHI P C , SINUSAS A J . Point-tracked quantitative analysis of left ventricular surface motion from 3-D image sequences[J]. IEEE Trans Med Imaging, 2000, 19 (1): 36- 50. doi: 10.1109/42.832958
28	PAPADEMETRIS , XENOPHON , SINUSAS , et al. Estimation of 3-D left ventricular deformation from medical images using biomechanical models[J]. IEEE Trans Med Imaging, 2002, 21 (7): 786- 800. doi: 10.1109/TMI.2002.801163
29	杨桂通. 弹性力学[M]. 第2版高等教育出版社, 2012.
30	GEMAN S , GERMAN D . Stochastic relaxation, gibbs distributions and the Bayesian restoration of images[J]. IEEE Trans Pattern Anal Mach Intell, 1984, 6, 721- 741.
31	ZHANG H Y , GAO Z , XU L , et al. A meshfree representation for cardiac medical image computing[J]. IEEE J Transl Eng He, 2018, 6, 1- 12.
32	李亚智. 有限元法基础与程序设计[M]. 科学出版社, 2004.

左心室功能参数	径向方向（峰值）				圆周方向（峰值）
左心室功能参数	RD/mm	RS/%	RV/(mm/s)	RSR/s^?1	CD/mm	CS/%	CV/(mm/s)	CSR/s^?1
射血正常	4.86±1.10	20.30±6.50	28.15±8.28	1.40±0.68	?3.09±0.71	?19.43±5.61	?18.19±5.34	?1.22±0.52
射血无力	3.29±0.92	11.62±5.65	19.69±5.60	0.78±0.32	?2.05±0.61	?10.86±4.63	?12.62±3.58	?0.63±0.24
p值	1.46e-11	2.04e-10	5.24e-08	1.45e-07	7.67e-12	6.40e-13	2.86e-08	2.06e-10

约束条件	径向方向（峰值）				圆周方向（峰值）
约束条件	RD/mm	RS/%	RV/(mm/s)	RSR/s^?1	CD/mm	CS/%	CV/(mm/s)	CSR/s^?1
无约束	4.07±0.97	21.03±6.89	24.83±6.45	1.35±0.49	?2.31±0.56	?13.06±6.56	?14.07±3.63	?1.19±0.88
平滑约束	4.23±0.96	18.76±5.05	25.72±6.48	1.10±0.35	?2.28±0.58	?10.51±3.68	?14.25±3.58	?0.76±0.33
模型约束	4.08±1.01	15.96±6.08	23.92±6.94	1.09±0.50	?2.57±0.66	?15.15±5.12	?15.40±4.46	?0.93±0.38

[1]	霍璐,胡晓欣,肖勤,顾雅佳,褚旭,姜娈. 基于nnU-Net的乳腺DCE-MR图像中乳房和腺体自动分割[J]. 波谱学杂志, 2021, 38(3): 367-380.
[2]	石沁祎,闫方,杨阳,陈玥甫,林晓浪,王远军. 基于水平集的牙齿牙槽骨图像分割[J]. 波谱学杂志, 2021, 38(2): 182-193.
[3]	岳晴, 王远军. 基于非局部约束球面反卷积模型的纤维追踪算法[J]. 波谱学杂志, 2020, 37(4): 422-433.
[4]	娄云重, 刘颖, 江华, 章浩伟. 基于MRI和深度学习的桥小脑角区脑膜瘤与听神经瘤分类算法研究[J]. 波谱学杂志, 2020, 37(3): 300-310.
[5]	赵尚义, 王远军. 基于磁共振图像和改进的UNet++模型区分阿尔茨海默症患者和健康人群[J]. 波谱学杂志, 2020, 37(3): 321-331.
[6]	蔡文琴, 王远军. 基于磁共振成像的人脑图谱构建方法研究进展[J]. 波谱学杂志, 2020, 37(2): 241-253.
[7]	宫进昌, 王宇, 王远军. 结合小波融合和深度学习的脑胶质瘤自动分割[J]. 波谱学杂志, 2020, 37(2): 131-143.
[8]	蒋帆, 王远军. 扩散张量成像的人脑模板构建[J]. 波谱学杂志, 2018, 35(4): 520-530.
[9]	颜志煜, 陈志伟. 基于Eclipse图形建模框架的图形化脉冲序列设计软件的实现[J]. 波谱学杂志, 2017, 34(2): 175-182.
[10]	周小龙1，聂生东1*，王远军1，张英力2，杨培强2. 基于迭代TSVD的NMR二维谱反演算法[J]. 波谱学杂志, 2013, 30(4): 541-551.
[11]	李新军1，聂生东1*，王远军1，杨培强2. 低场核磁共振扩散-横向弛豫二维反演算法[J]. 波谱学杂志, 2013, 30(3): 322-335.
[12]	蒋帆, 王远军. 扩散张量图像的插值方法综述[J]. 波谱学杂志, 2019, 36(3): 392-407.