Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Fast track – JSH2024 FUKUOKA
  • Published:

Deep learning assists early-detection of hypertension-mediated heart change on ECG signals

Hypertension Research volume 48, pages 681–692 (2025)Cite this article

A Correction to this article was published on 16 December 2024

This article has been updated

Abstract

Arterial hypertension is a major risk factor for cardiovascular diseases. While cardiac ultrasound is a typical way to diagnose hypertension-mediated heart change, it often fails to detect early subtle structural changes. Electrocardiogram(ECG) represents electrical activity of heart muscle, affected by the changes in heart’s structure. It is crucial to explore whether ECG can capture slight signals of hypertension-mediated heart change. However, reading ECG records is complex and some signals are too subtle to be captured by cardiologist’s visual inspection. In this study, we designed a deep learning model to predict hypertension on ECG signals and then to identify hypertension-associated ECG segments. From The First Affiliated Hospital of Xiamen University, we collected 210,120 10-s 12-lead ECGs using the FX-8322 manufactured by FUKUDA and 812 ECGs using the RAGE-12 manufactured by NALONG. We proposed a deep learning framework, including MML-Net, a multi-branch, multi-scale LSTM neural network to evaluate the potential of ECG signals to detect hypertension, and ECG-XAI, an ECG-oriented wave-alignment AI explanation pipeline to identify hypertension-associated ECG segments. MML-Net achieved an 82% recall and an 87% precision in the testing, and an 80% recall and an 82% precision in the independent testing. In contrast, experienced clinical cardiologists typically attain recall rates ranging from 30 to 50% by visual inspection. The experiments demonstrate that ECG signals are sensitive to slight changes in heart structure caused by hypertension. ECG-XAI detects that R-wave and P-wave are the hypertension-associated ECG segments. The proposed framework has the potential to facilitate early diagnosis of heart change.

Access through your institution
Buy or subscribe

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Change history

References

  1. Dzau VJ, Balatbat CA. Future of hypertension the need for transformation. Hypertension. 2019;74:450–7. https://doi.org/10.1161/Hypertensionaha.119.13437

    Article  CAS  PubMed  Google Scholar 

  2. Desai AN. High blood pressure. JAMA. 2020;324:1254–5. https://doi.org/10.1001/jama.2020.11289

    Article  PubMed  Google Scholar 

  3. Fuchs FD, Whelton PK. High blood pressure and cardiovascular disease. Hypertension. 2020;75:285–92. https://doi.org/10.1161/HYPERTENSIONAHA.119.14240

    Article  CAS  PubMed  Google Scholar 

  4. Dzeshka MS, Shantsila A, Shantsila E, Lip GYH. Atrial fibrillation and hypertension. Hypertension. 2017;70:854–61. https://doi.org/10.1161/HYPERTENSIONAHA.117.08934

    Article  CAS  PubMed  Google Scholar 

  5. Yildiz M, Oktay AA, Stewart MH, Milani RV, Ventura HO, Lavie CJ. Left ventricular hypertrophy and hypertension. Prog Cardiovascular Dis. 2020;63:10–21. https://doi.org/10.1016/j.pcad.2019.11.009

    Article  Google Scholar 

  6. Velagaleti RS, Gona P, Pencina MJ, Aragam J, Wang TJ, Levy D, et al. Left ventricular hypertrophy patterns and incidence of heart failure with preserved versus reduced ejection fraction. Am J Cardiol. 2014;113:117–22. https://doi.org/10.1016/j.amjcard.2013.09.028

    Article  PubMed  Google Scholar 

  7. Valente AM, Lakdawala NK, Powell AJ, Evans SP, Cirino AL, Orav EJ, et al. Comparison of echocardiographic and cardiac magnetic resonance imaging in hypertrophic cardiomyopathy sarcomere mutation carriers without left ventricular hypertrophy. Circ Cardiovasc Genet. 2013;6:230–7. https://doi.org/10.1161/CIRCGENETICS.113.000037

    Article  PubMed  PubMed Central  Google Scholar 

  8. Oseni AO, Qureshi WT, Almahmoud MF, Bertoni AG, Bluemke DA, Hundley WG, et al. Left ventricular hypertrophy by ECG versus cardiac MRI as a predictor for heart failure. Heart. 2017;103:49–54. https://doi.org/10.1136/heartjnl-2016-309516

    Article  PubMed  Google Scholar 

  9. Derumeaux G, Mulder P, Richard V, Chagraoui A, Nafeh C, Bauer F, et al. Tissue Doppler imaging differentiates physiological from pathological pressure-overload left ventricular hypertrophy in rats. Circulation. 2002;105:1602–8. https://doi.org/10.1161/01.cir.0000012943.91101.d7

    Article  PubMed  Google Scholar 

  10. Bacharova L. ECG in left ventricular hypertrophy: A change in paradigm from assessing left ventricular mass to its electrophysiological properties. J Electrocardiol. 2022;73:153–6. https://doi.org/10.1016/j.jelectrocard.2022.06.002

    Article  PubMed  Google Scholar 

  11. Boles U, Enriquez A, Ghabra WA, Abdollah H, Michael KA. Early changes on the electrocardiogram in hypertension. e-journal of the ESC Council for Cardiology Practice. 2015;13:30.

  12. Miceli F, Presta V, Citoni B, Canichella F, Figliuzzi I, Ferrucci A, et al. Conventional and new electrocardiographic criteria for hypertension-mediated cardiac organ damage: A narrative review. J Clin Hypertens (Greenwich). 2019;21:1863–71. https://doi.org/10.1111/jch.13726

    Article  PubMed  Google Scholar 

  13. De la Garza Salazar F, Romero Ibarguengoitia ME, Azpiri Lopez JR, Gonzalez Cantu A. Optimizing ECG to detect echocardiographic left ventricular hypertrophy with computer-based ECG data and machine learning. PLoS One. 2021;16:e0260661 https://doi.org/10.1371/journal.pone.0260661

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Su F-Y, Li Y-H, Lin Y-P, Lee C-J, Wang C-H, Meng F-C, et al. A comparison of Cornell and Sokolow-Lyon electrocardiographic criteria for left ventricular hypertrophy in a military male population in Taiwan: the Cardiorespiratory fitness and HospItalization Events in armed Forces study. Cardiovascular Diagn Ther. 2017;7:244–51.

    Article  Google Scholar 

  15. Levy D, Labib SB, Anderson KM, Christiansen JC, Kannel WB, Castelli WP. Determinants of sensitivity and specificity of electrocardiographic criteria for left ventricular hypertrophy. Circulation. 1990;81:815–20. https://doi.org/10.1161/01.cir.81.3.815

    Article  CAS  PubMed  Google Scholar 

  16. Molloy TJ, Okin PM, Devereux RB, Kligfield P. Electrocardiographic detection of left ventricular hypertrophy by the simple QRS voltage-duration product. J Am Coll Cardiol. 1992;20:1180–6. https://doi.org/10.1016/0735-1097(92)90376-x

    Article  CAS  PubMed  Google Scholar 

  17. Somani S, Russak AJ, Richter F, Zhao S, Vaid A, Chaudhry F, et al. Deep learning and the electrocardiogram: review of the current state-of-the-art. Europace. 2021;23:1179–91. https://doi.org/10.1093/europace/euaa377

    Article  PubMed  PubMed Central  Google Scholar 

  18. Kwon JM, Lee SY, Jeon KH, Lee Y, Kim KH, Park J, et al. Deep learning-based algorithm for detecting aortic stenosis using electrocardiography. J Am Heart Assoc. 2020;9:e014717 https://doi.org/10.1161/JAHA.119.014717

    Article  PubMed  PubMed Central  Google Scholar 

  19. Kwon JM, Jeon KH, Kim HM, Kim MJ, Lim SM, Kim KH, et al. Comparing the performance of artificial intelligence and conventional diagnosis criteria for detecting left ventricular hypertrophy using electrocardiography. Europace. 2020;22:412–9. https://doi.org/10.1093/europace/euz324

    Article  PubMed  Google Scholar 

  20. Cai W, Chen Y, Guo J, Han B, Shi Y, Ji L, et al. Accurate detection of atrial fibrillation from 12-lead ECG using deep neural network. Comput Biol Med. 2020;116:103378 https://doi.org/10.1016/j.compbiomed.2019.103378

    Article  PubMed  Google Scholar 

  21. Ko WY, Siontis KC, Attia ZI, Carter RE, Kapa S, Ommen SR, et al. Detection of hypertrophic cardiomyopathy using a convolutional neural network-enabled electrocardiogram. J Am Coll Cardiol. 2020;75:722–33. https://doi.org/10.1016/j.jacc.2019.12.030

    Article  PubMed  Google Scholar 

  22. Goto S, Solanki D, John JE, Yagi R, Homilius M, Ichihara G, et al. Multinational federated learning approach to train ECG and echocardiogram models for hypertrophic cardiomyopathy detection. Circulation. 2022;146:755–69. https://doi.org/10.1161/CIRCULATIONAHA.121.058696

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Hu J, Shen L, Sun G. Squeeze-and-excitation networks. Paper/Poster presented at: Proceedings of the IEEE conference on computer vision and pattern recognition; 2018;

  24. Makowski D, Pham T, Lau ZJ, Brammer JC, Lespinasse F, Pham H, et al. NeuroKit2: a python toolbox for neurophysiological signal processing. Behav Res Methods. 2021;53:1689–96. https://doi.org/10.3758/s13428-020-01516-y

    Article  PubMed  Google Scholar 

  25. Jiang P-T, Zhang C-B, Hou Q, Cheng M-M, Wei Y. LayerCAM: exploring hierarchical class activation maps for localization. IEEE Transations Image Process. 2021;30:5875–88. https://doi.org/10.1109/TIP.2021.3089943.

    Article  Google Scholar 

  26. Sakli N, Ghabri H, Soufiene BO, Almalki FA, Sakli H, Ali O, et al. ResNet-50 for 12-lead electrocardiogram automated diagnosis. Comput Intell Neurosci. 2022;2022:7617551 https://doi.org/10.1155/2022/7617551

    Article  PubMed  PubMed Central  Google Scholar 

  27. Murugesan B, Ravichandran V, Ram K, P SP, Joseph J, Shankaranarayana SM, et al. ECGNet: Deep Network for Arrhythmia Classification. Paper/Poster presented at: 2018 IEEE International Symposium on Medical Measurements and Applications (MeMeA); 11-13 2018, 2018;

  28. Kropf M, Hayn D, Schreier G. ECG classification based on time and frequency ___domain features using random forests. Paper/Poster presented at: 2017 Computing in Cardiology (CinC); 24-27 Sept. 2017, 2017;

  29. Ryu SY, Lee SH, Isenberg G, Ho WK, Earm YE. Monitoring of ANP secretion from single atrial myocytes using densitometry. Pflug Arch. 2002;444:568–77. https://doi.org/10.1007/s00424-002-0852-7

    Article  CAS  Google Scholar 

  30. Mayyas F, Niebauer M, Zurick A, Barnard J, Gillinov AM, Chung MK, et al. Association of left atrial endothelin-1 with atrial rhythm, size, and fibrosis in patients with structural heart disease. Circ Arrhythm Electrophysiol. 2010;3:369–79. https://doi.org/10.1161/CIRCEP.109.924985

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Triposkiadis F, Pieske B, Butler J, Parissis J, Giamouzis G, Skoularigis J, et al. Global left atrial failure in heart failure. Eur J Heart Fail. 2016;18:1307–20. https://doi.org/10.1002/ejhf.645

    Article  PubMed  Google Scholar 

  32. Lau YF, Yiu KH, Siu CW, Tse HF. Hypertension and atrial fibrillation: epidemiology, pathophysiology and therapeutic implications. J Hum Hypertens. 2012;26:563–9. https://doi.org/10.1038/jhh.2011.105

    Article  CAS  PubMed  Google Scholar 

  33. Noresson E, Ricksten SE, Thoren P. Left atrial pressure in normotensive and spontaneously hypertensive rats. Acta Physiol Scand. 1979;107:9–12. https://doi.org/10.1111/j.1748-1716.1979.tb06436.x

    Article  CAS  PubMed  Google Scholar 

  34. Choisy SC, Arberry LA, Hancox JC, James AF. Increased susceptibility to atrial tachyarrhythmia in spontaneously hypertensive rat hearts. Hypertension. 2007;49:498–505. https://doi.org/10.1161/01.HYP.0000257123.95372.ab

    Article  CAS  PubMed  Google Scholar 

  35. Kistler PM, Sanders P, Dodic M, Spence SJ, Samuel CS, Zhao C, et al. Atrial electrical and structural abnormalities in an ovine model of chronic blood pressure elevation after prenatal corticosteroid exposure: implications for development of atrial fibrillation. Eur Heart J. 2006;27:3045–56. https://doi.org/10.1093/eurheartj/ehl360

    Article  PubMed  Google Scholar 

  36. Lau DH, Mackenzie L, Kelly DJ, Psaltis PJ, Worthington M, Rajendram A, et al. Short-term hypertension is associated with the development of atrial fibrillation substrate: a study in an ovine hypertensive model. Heart Rhythm. 2010;7:396–404. https://doi.org/10.1016/j.hrthm.2009.11.031

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

Shaorong Fang and Tianfu Wu from Information and Network Center of Xiamen University are acknowledged for the help with high performance computing (HPC).

Funding

This work was supported by National Natural Science Foundation of China (62173282).

Author information

Authors and Affiliations

  1. Department of Automation, Xiamen University, Xiamen, Fujian, China

    Chengwei Liang, Fan Yang & Ying Wang

  2. National Institute for Data Science in Health and Medicine, Xiamen University, Xiamen, Fujian, China

    Chengwei Liang, Fan Yang & Ying Wang

  3. Xiamen Key Laboratory of Big Data Intelligent Analysis and Decision, Xiamen, Fujian, China

    Fan Yang & Ying Wang

  4. Fuzhou First General Hospital, Fujian Medical University, Fujian, China

    Xiaobing Huang

  5. The First Affiliation Hospital of Xiamen University, Xiamen University, Xiamen, Fujian, China

    Lijuan Zhang

Authors
  1. Chengwei Liang
    View author publications

    Search author on:PubMed Google Scholar

  2. Fan Yang
    View author publications

    Search author on:PubMed Google Scholar

  3. Xiaobing Huang
    View author publications

    Search author on:PubMed Google Scholar

  4. Lijuan Zhang
    View author publications

    Search author on:PubMed Google Scholar

  5. Ying Wang
    View author publications

    Search author on:PubMed Google Scholar

Corresponding authors

Correspondence to Lijuan Zhang or Ying Wang.

Ethics declarations

Conflict of interest

The authors declare no competing interests.

Ethics approval and consent to participate

The collected ECG records was approved by First Affiliated Hospital of Xiamen University Ethics Committee (reference 2023131).

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Fig. 1 in the original version of this article has been replaced. The original article has been corrected.

Supplementary information

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liang, C., Yang, F., Huang, X. et al. Deep learning assists early-detection of hypertension-mediated heart change on ECG signals. Hypertens Res 48, 681–692 (2025). https://doi.org/10.1038/s41440-024-01938-7

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41440-024-01938-7

Keywords

Search

Advanced search

Quick links