Share：
Share this content in WeChat
X
[Chinese][PDF]1734103
• Clinical Article •
Quantitative analysis and differentiation of MR images between hypertrophic cardiomyopathy and hypertensive left ventricular hypertrophy with U-Net neural network
JIAO Ziling  WEI Hanyu  LI Jifan  CHEN Shuo  CHAI Yezi  LIU Qiming  LI Rui  JIANG Meng 

Cite this article as: Jiao ZL, Wei HY, Li JF, et al. Quantitative analysis and differentiation of MR images between hypertrophic cardiomyopathy and hypertensive left ventricular hypertrophy with U-Net neural network. Chin J Magn Reson Imaging, 2020, 11(9): 741-746. DOI:10.12015/issn.1674-8034.2020.09.005.


[Abstract] Objective: To investigate the value of quantitative information of MRI got from U-Net neural network in the differentiation of hypertrophic cardiomyopathy and hypertensive left ventricular hypertrophy.Materials and Methods: We retrospectively analyzed 100 heart disease subjects collected from Medical Image Computing and Computer Assisted Intervention Society (MICCAI) 2017 automated cardiac diagnosis challenge and 45 hypertrophic cardiomyopathy patients and 48 hypertensive left ventricular hypertrophy patients collected from July 2013 to March 2019 in the department of cardiology, Renji Hospital of Shanghai Jiaotong University. All patients underwent the steady state free precession cine sequence MRI scan in short axis. MICCAI dataset, separated into 1710 images and 190 images, were used as training dataset and validating dataset. Five hypertrophic cardiomyopathy patients and 5 hypertensive left ventricular hypertrophy patients, including 190 images, were selected as test dataset. The U-Net model was utilized in the segmentation of heart in cine MR images. The image segmentation was performed on all the hypertrophic cardiomyopathy and hypertensive left ventricular hypertrophy patients and the quantitative parameters were calculated based on the segmentation results. Independent t test was applied to compare the differences of all the parameters between the two diseases groups. Multivariate logistic regression and a 4-fold cross-validation method were applied to fit a diagnosis model and to validate the robust and diagnostic accuracy of the model.Results: Thirteen of all the 55 quantitative parameters had significant differences between the hypertrophic cardiomyopathy group and hypertensive left ventricular hypertrophy group, and 3 of them had significant influences on the classification between the two groups. The training set and the test set were 70 and 23 cases, and the areas under curves of ROC in test set produced from 4-fold cross-validation were 0.939, 0.984, 0.972 and 0.963. The accuracy of the test set corresponding to the best model was 86.96% (20/23).Conclusions: Automatic segmentation of heart in cine MR images based on U-Net neural network can provide more quantification information, which can help to diagnose the hypertrophic cardiomyopathy and hypertensive left ventricular hypertrophy.
[Keywords] convolutional neural network；hypertrophic cardiomyopathy；hypertensive left ventricular hypertrophy；quantitative analysis；magnetic resonance imaging

JIAO Ziling Center for Biomedical Imaging Reseach, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing 100084, China

WEI Hanyu Center for Biomedical Imaging Reseach, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing 100084, China

LI Jifan Center for Biomedical Imaging Reseach, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing 100084, China

CHEN Shuo Center for Biomedical Imaging Reseach, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing 100084, China

CHAI Yezi Department of Cardiology, Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200127, China

LIU Qiming Department of Cardiology, Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200127, China

LI Rui* Center for Biomedical Imaging Reseach, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing 100084, China

JIANG Meng* Department of Cardiology, Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200127, China

*Correspondence to: Li R, Email: leerui@tsinghua.edu.cn. Jiang M, Email: jiangmeng0919@163.com

Conflicts of interest   None.

ACKNOWLEDGMENTS  This work was part of National Key Research and Development Projects No. 2016YFC1301601, 2017YFC0109002 the National Natural Science Foundation of China No. 81971604
Received  2020-01-31
Accepted  2020-07-12
DOI: 10.12015/issn.1674-8034.2020.09.005
Cite this article as: Jiao ZL, Wei HY, Li JF, et al. Quantitative analysis and differentiation of MR images between hypertrophic cardiomyopathy and hypertensive left ventricular hypertrophy with U-Net neural network. Chin J Magn Reson Imaging, 2020, 11(9): 741-746. DOI:10.12015/issn.1674-8034.2020.09.005.

[1]
Goff ZD, Calkins H. Sudden death related cardiomyopathies: hypertrophic cardiomyopathy. Progr Cardiovascular Dis, 2019, 62(3): 212-216. DOI: 10.1016/j.pcad.2019.04.001
[2]
中华医学会心血管病学分会中国成人肥厚型心肌病诊断与治疗指南编写组，中华心血管病杂志编辑委员会．中国成人肥厚型心肌病诊断与治疗指南．中华心血管病杂志, 2017, 45(12): 1015-1032. DOI: 10.3760/cma.j.issn.0253-3758.2017.12.005
[3]
McNally EM, Barefield DY, Puckelwartz MJ. The Genetic landscape of cardiomyopathy and its role in heart failure. Cell Metab, 2015, 21(2): 174-182. DOI: 10.1016/j.cmet.2015.01.013
[4]
Maron BJ, Maron MS. Hypertrophic cardiomyopathy. Lancet, 2013, 381(9862): 242-255. DOI: 10.1016/S0140-6736(12)60397-3
[5]
Daimee UA, Lande MB, Tang W, et al. Blood pressure and left ventricular mass index in healthy adolescents. Blood pressure monitoring, 2017, 22(1): 48-50. DOI: 10.1097/MBP.0000000000000219
[6]
踪念玉．使用心脏彩超诊断高血压左室肥厚伴左心力衰竭的价值研究．当代医药论丛, 2015, 13(12): 238-239.
[7]
Stabouli S, Kotsis V, Rizos Z, et al. Left ventricular mass in normotensive, prehypertensive and hypertensive children and adolescents. Pediatr Nephrol, 2009, 24(8): 1545-1551. DOI: 10.1007/s00467-009-1165-2
[8]
Urbina EM, Khoury PR, McCoy C, et al. Cardiac and vascular consequences of pre-hypertension in youth. J Clin Hypertens, 2011, 13(5): 332-342. DOI: 10.1111/j.1751-7176.2011.00471.x
[9]
Drukteinis JS, Roman MJ, Fabsitz RR, et al. Cardiac and systemic hemodynamic characteristics of hypertension and prehypertension in adolescents and young adults: the strong heart study. Circulation, 2007, 115(2): 221-227. DOI: 10.1161/CIRCULATIONAHA.106.668921
[10]
程赛楠，赵世华．心脏磁共振成像新技术进展与展望．磁共振成像, 2016, 7(7): 541-545. DOI: 10.12015/issn.1674-8034.2016.07.012
[11]
Ripley DP, Musa TA, Dobson LE, et al. Cardiovascular magnetic resonance imaging: what the general cardiologist should know. Heart, 2016, 102(19): 589-603. DOI: 10.1136/heartjnl-2015-307896
[12]
Cui Y, Chen Y, Cao Y, et al. Myocardial extracellular volume fraction measurements with MOLLI 5(3)3 by cardiovascular MRI for the discrimination of healthy volunteers from dilated and hypertrophic cardiomyopathy patients. Clin Radiol, 2019, 74(9): 732.e9-732, e16. DOI: 10.1016/j.crad.2019.04.019
[13]
Bernard O, Lalande A, Zotti C, et al. Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE Transact Med Imaging, 2018, 37(11): 2514-2525. DOI: 10.1109/TMI.2018.2837502
[14]
Hu HF, Pan N, Wang JY, et al. Automatic segmentation of left ventricle from cardiac MRI via deep learning and region constrained dynamic programming. Neurocomputing, 2019, 347(1): 139-148. DOI: 10.1016/j.neucom.2019.02.008
[15]
Zotti C, Luo Z, Lalande A, et al. Convolutional neural network with shape prior applied to cardiac MRI segmentation. IEEE J Biomed Health Inform, 2019, 23(3): 1119-1128. DOI: 10.1109/JBHI.2018.2865450
[16]
Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Analy, 2017, 42(1): 60-88. DOI: 10.1016/j.media.2017.07.005
[17]
Lee EJ, Kim YH, Kim N, et al. Deep into the brain: artificial intelligence in stroke imaging. J Stroke, 2017, 19(3): 277-285. DOI: 10.5853/jos.2017.02054
[18]
Roy AG, Navab N, Wachinger C. Recalibrating fully convolutional networks with spatial and channel "squeeze and excitation" blocks. IEEE Transact Med Imaging, 2019, 38(2): 540-549. DOI: 10.1109/TMI.2018.2867261
[19]
Greenspan H, van Ginneken B, Summers RM. Guest editorial deep learning in medical imaging: overview and future promise of an exciting new technique. IEEE Transact Med Imaging, 2016, 35(5): 1153-1159. DOI: 10.1109/TMI.2016.2553401
[20]
Jin Liu, Yi Pan, Min Li, et al. Applications of deep learning to MRI images: a survey. Big Data Mining Analy, 2018, 1(1): 1-18. DOI: 10.26599/BDMA.2018.9020001

PREV Evaluating blood flow status after revascularization in patients with moyamoya disease by using TOF-MRA, 3D-ASL and territory ASL
NEXT Preliminary study of multimodal magnetic resonance imaging in the evaluation of cervical spondylotic myelopathy
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn