Abstract
Accurately quantifying multidimensional indices of the left ventricle (LV) in 2D echocardiography (echo) is clinically significant for cardiac disease diagnosis. However, the challenges of the frequently missing information, the high geometric variability and the uncertain multidimensional indices relation hinder its automated analysis development. Here, we propose an EchoQuan-Net to directly quantify LV in echo sequence from the multidimension, covering length and width for 1D, area for 2D, and volume for 3D. The net consist of three components: (1) Global-Local Learning to capture contextual information in the cardiac cycle for each frame, with global information from whole sequence and local information from the individual frame; (2) Geometric Adjustment to promote a canonical region of interest for LV, with translation, rotation and scale invariant; (3) Multi-target Relation learning to promote joint quantification for LV multidimensional indices, with sparse latent regression. The experiments reveal that EchoQuan-Net gains high accuracy, with mean accuracy error of 3.14 mm, 3.10 mm, 276 mm\(^2\) and 13.5 ml for length, width, area and volume. The results show great potential of our method in clinical cardiac function assessment.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Schiller, N.B., Shah, P.M., Crawford, M., DeMaria, A., Devereux, R., Feigenbaum, H., et al.: Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. J. Am. Soc. Echocardiogr. 2(5), 358–367 (1989). https://doi.org/10.1016/S0894-7317(89)80014-8
Kjaergaard, J., Petersen, C.L., Kjaer, A., Schaadt, B.K., Oh, J.K., Hassager, C.: Evaluation of right ventricular volume and function by 2D and 3D echocardiography compared to MRI. Eur. J. Echocardiogr. 7(6), 430–438 (2006). https://doi.org/10.1016/j.euje.2005.10.009
Margossian, R., et al.: The reproducibility and absolute values of echocardiographic measurements of left ventricular size and function in children are algorithm dependent. J. Am. Soc. Echocardiogr. 28(5), 549–558 (2015). https://doi.org/10.1016/j.echo.2015.01.014
Abdi, A.H., Luong, C., Tsang, T., Allan, G., Nouranian, S., Jue, J., et al.: Automatic quality assessment of echocardiograms using convolutional neural networks: feasibility on the apical four-chamber view. IEEE Trans. Med. Imaging 36(6), 1221–1230 (2017). https://doi.org/10.1109/TMI.2017.2690836
Pascual, M., et al.: Effects of isolated obesity on systolic and diastolic left ventricular function. Heart 89(10), 1152–1156 (2003). https://doi.org/10.1136/heart.89.10.1152
Xue, W., Islam, A., Bhaduri, M., Li, S.: Direct multitype cardiac indices estimation via joint representation and regression learning. IEEE Trans. Med. Imaging 36(10), 2057–2067 (2017). https://doi.org/10.1109/TMI.2017.2709251
Carneiro, G., Nascimento, J.C., Freitas, A.: The segmentation of the left ventricle of the heart from ultrasound data using deep learning architectures and derivative-based search methods. IEEE Trans. Image Process. 21(3), 968–982 (2012). https://doi.org/10.1109/TIP.2011.2169273
Zhen, X., Wang, Z., Islam, A., Bhaduri, M., Chan, I., Li, S.: Direct estimation of cardiac bi-ventricular volumes with regression forests. In: Golland, P., Hata, N., Barillot, C., Hornegger, J., Howe, R. (eds.) MICCAI 2014. LNCS, vol. 8674, pp. 586–593. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10470-6_73
Zhen, X., Islam, A., Bhaduri, M., Chan, I., Li, S.: Direct and simultaneous four-chamber volume estimation by multi-output regression. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9349, pp. 669–676. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24553-9_82
Zhen, X., Wang, Z., Islam, A., Bhaduri, M., Chan, I., Li, S.: Multi-scale deep networks and regression forests for direct bi-ventricular volume estimation. Med. Image Anal. 30, 120–129 (2016). https://doi.org/10.1016/j.media.2015.07.003
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016). https://doi.org/10.1109/CVPR.2016.90
Jaderberg, M., Simonyan, K., Zisserman, A.: Spatial transformer networks. In: Advances in Neural Information Processing Systems, pp. 2017–2025 (2015). arXiv:1506.02025
Dai, J., He, K., Sun, J.: Instance-aware semantic segmentation via multi-task network cascades. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3150–3158 (2016). https://doi.org/10.1109/CVPR.2016.343
Vigneault, D.M., Xie, W., Ho, C.Y., Bluemke, D.A., Noble, J.A.: \(\varOmega \)-net (omega-net): fully automatic, multi-view cardiac MR detection, orientation, and segmentation with deep neural networks. Med. Image Anal. 48, 95–106 (2018). https://doi.org/10.1016/j.media.2018.05.008
Kimeldorf, G.S., Wahba, G.: A correspondence between Bayesian estimation on stochastic processes and smoothing by splines. Ann. Math. Stat. 41(2), 495–502 (1970). https://doi.org/10.1214/aoms/1177697089
Ding, C., Zhou, D., He, X., Zha, H.: R 1-PCA: rotational invariant L\(_1\)-norm principal component analysis for robust subspace factorization. In: Proceedings of the 23rd International Conference on Machine Learning, pp. 281–288. ACM (2006). https://doi.org/10.1145/1143844.1143880
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Acknowledgments
This work was supported by the Postgraduate Research & Practice Innovation Program of Jiangsu Province (No. KYCX17_0104); the China Scholarship Council (No. 201706090248); the National Natural Science Foundation (No. 61871117, 61828101); the States Key Project of Research and Development Plan (No. 2017YFA0104302, 2017YFC0109202 and 2017YFC0107900); and the Science and Technology Program of Guangdong (No. 2018B030333001).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Ge, R. et al. (2019). EchoQuan-Net: Direct Quantification of Echo Sequence for Left Ventricle Multidimensional Indices via Global-Local Learning, Geometric Adjustment and Multi-target Relation Learning. In: Tetko, I., Kůrková, V., Karpov, P., Theis, F. (eds) Artificial Neural Networks and Machine Learning – ICANN 2019: Workshop and Special Sessions. ICANN 2019. Lecture Notes in Computer Science(), vol 11731. Springer, Cham. https://doi.org/10.1007/978-3-030-30493-5_24
Download citation
DOI: https://doi.org/10.1007/978-3-030-30493-5_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-30492-8
Online ISBN: 978-3-030-30493-5
eBook Packages: Computer ScienceComputer Science (R0)