Korean Circulation Journal

  • 제53권11호
  • /
  • Pages.758-771
  • /
  • 2023
  • /
  • 1738-5520(pISSN)
  • /
  • 1738-5555(eISSN)
대한심장학회 (The Korean Society of Cardiology)
권호 표지
DOI QR코드

DOI QR Code

Identifying Atrial Fibrillation With Sinus Rhythm Electrocardiogram in Embolic Stroke of Undetermined Source: A Validation Study With Insertable Cardiac Monitors

  • Ki-Hyun Jeon (Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital) ;
  • Jong-Hwan Jang (Medical Research Team, Medical AI Inc.) ;
  • Sora Kang (Medical Research Team, Medical AI Inc.) ;
  • Hak Seung Lee (Medical Research Team, Medical AI Inc.) ;
  • Min Sung Lee (Medical Research Team, Medical AI Inc.) ;
  • Jeong Min Son (Medical Research Team, Medical AI Inc.) ;
  • Yong-Yeon Jo (Medical Research Team, Medical AI Inc.) ;
  • Tae Jun Park (Medical Research Team, Medical AI Inc.) ;
  • Il-Young Oh (Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital) ;
  • Joon-myoung Kwon (Medical Research Team, Medical AI Inc.) ;
  • Ji Hyun Lee (Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital)
  • 투고 : 2023.01.04
  • 심사 : 2023.06.28
  • 발행 : 2023.11.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Background and Objectives: Paroxysmal atrial fibrillation (AF) is a major potential cause of embolic stroke of undetermined source (ESUS). However, identifying AF remains challenging because it occurs sporadically. Deep learning could be used to identify hidden AF based on the sinus rhythm (SR) electrocardiogram (ECG). We combined known AF risk factors and developed a deep learning algorithm (DLA) for predicting AF to optimize diagnostic performance in ESUS patients. Methods: A DLA was developed to identify AF using SR 12-lead ECG with the database consisting of AF patients and non-AF patients. The accuracy of the DLA was validated in 221 ESUS patients who underwent insertable cardiac monitor (ICM) insertion to identify AF. Results: A total of 44,085 ECGs from 12,666 patient were used for developing the DLA. The internal validation of the DLA revealed 0.862 (95% confidence interval, 0.850-0.873) area under the curve (AUC) in the receiver operating curve analysis. In external validation data from 221 ESUS patients, the diagnostic accuracy of DLA and AUC were 0.811 and 0.827, respectively, and DLA outperformed conventional predictive models, including CHARGE-AF, C2HEST, and HATCH. The combined model, comprising atrial ectopic burden, left atrial diameter and the DLA, showed excellent performance in AF prediction with AUC of 0.906. Conclusions: The DLA accurately identified paroxysmal AF using 12-lead SR ECG in patients with ESUS and outperformed the conventional models. The DLA model along with the traditional AF risk factors could be a useful tool to identify paroxysmal AF in ESUS patients.

키워드

참고문헌

  1. Ntaios G. Embolic stroke of undetermined source: JACC review topic of the week. J Am Coll Cardiol 2020;75:333-40. 
  2. Kirchhof P, Benussi S, Kotecha D, et al. 2016 ESC Guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Europace 2016;18:1609-78. 
  3. Ziegler PD, Rogers JD, Ferreira SW, et al. Long-term detection of atrial fibrillation with insertable cardiac monitors in a real-world cryptogenic stroke population. Int J Cardiol 2017;244:175-9. 
  4. Li YG, Pastori D, Farcomeni A, et al. A simple clinical risk score (C2HEST) for predicting incident atrial fibrillation in Asian subjects: derivation in 471,446 Chinese subjects, with internal validation and external application in 451,199 Korean subjects. Chest 2019;155:510-8. 
  5. Suenari K, Chao TF, Liu CJ, Kihara Y, Chen TJ, Chen SA. Usefulness of HATCH score in the prediction of new-onset atrial fibrillation for Asians. Medicine (Baltimore) 2017;96:e5597. 
  6. Alonso A, Krijthe BP, Aspelund T, et al. Simple risk model predicts incidence of atrial fibrillation in a racially and geographically diverse population: the CHARGE-AF consortium. J Am Heart Assoc 2013;2:e000102. 
  7. Kwon JM, Jeon KH, Kim HM, et al. Comparing the performance of artificial intelligence and conventional diagnosis criteria for detecting left ventricular hypertrophy using electrocardiography. Europace 2020;22:412-9. 
  8. Kwon JM, Cho Y, Jeon KH, et al. A deep learning algorithm to detect anaemia with ECGs: a retrospective, multicentre study. Lancet Digit Health 2020;2:e358-67. 
  9. Siontis KC, Noseworthy PA, Attia ZI, Friedman PA. Artificial intelligence-enhanced electrocardiography in cardiovascular disease management. Nat Rev Cardiol 2021;18:465-78. 
  10. Attia ZI, Noseworthy PA, Lopez-Jimenez F, et al. An artificial intelligence-enabled ECG algorithm for the identification of patients with atrial fibrillation during sinus rhythm: a retrospective analysis of outcome prediction. Lancet 2019;394:861-7. 
  11. Hannun AY, Rajpurkar P, Haghpanahi M, et al. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nat Med 2019;25:65-9. 
  12. Hart RG, Diener HC, Coutts SB, et al. Embolic strokes of undetermined source: the case for a new clinical construct. Lancet Neurol 2014;13:429-38. 
  13. Lee JH, Moon IT, Cho Y, et al. Left atrial diameter and atrial ectopic burden in patients with embolic stroke of undetermined source: risk stratification of atrial fibrillation with insertable cardiac monitor analysis. J Clin Neurol 2021;17:213-9. 
  14. Li YG, Bisson A, Bodin A, et al. C2 HEST score and prediction of incident atrial fibrillation in poststroke patients: a French nationwide study. J Am Heart Assoc 2019;8:e012546. 
  15. Barrett TW, Self WH, Wasserman BS, McNaughton CD, Darbar D. Evaluating the HATCH score for predicting progression to sustained atrial fibrillation in ED patients with new atrial fibrillation. Am J Emerg Med 2013;31:792-7.
  16. Shih SM, Tien PJ, Karnin Z. GANMEX: One-vs-one attributions guided by GAN-based counterfactual explanation baselines. arXiv. 2020 [Epub ahead of print]. 
  17. Lang O, Gandelsman Y, Yarom M, et al. Explaining in style: training a GAN to explain a classifier in StyleSpace. arXiv. 2021 [Epub ahead of print]. 
  18. Hindricks G, Potpara T, Dagres N, et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): The Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) Developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J 2021;42:373-498. 
  19. Purerfellner H, Sanders P, Sarkar S, et al. Adapting detection sensitivity based on evidence of irregular sinus arrhythmia to improve atrial fibrillation detection in insertable cardiac monitors. Europace 2018;20:f321-8. 
  20. Tao Y, Xu J, Gong X, Sun J, Yang D. Premature atrial complexes can predict atrial fibrillation in ischemic stroke patients: a systematic review and meta-analysis. Pacing Clin Electrophysiol 2021;44:1599-606. 
  21. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-CAM: visual explanations from deep networks via gradient-based localization. Int J Comput Vis 2020;128:336-59. 
  22. Smith JW, O'Neal WT, Shoemaker MB, et al. PR-interval components and atrial fibrillation risk (from the Atherosclerosis Risk in Communities Study). Am J Cardiol 2017;119:466-72. 
  23. Mandyam MC, Soliman EZ, Alonso A, et al. The QT interval and risk of incident atrial fibrillation. Heart Rhythm 2013;10:1562-8. 
  24. Xiang H, Xue Y, Chen Z, et al. The association between left ventricular hypertrophy and the occurrence and prognosis of atrial fibrillation: a meta-analysis. Front Cardiovasc Med 2021;8:639993. 
  25. Hart RG, Catanese L, Perera KS, Ntaios G, Connolly SJ. Embolic stroke of undetermined source: a systematic review and clinical update. Stroke 2017;48:867-72. 
  26. Sanna T, Diener HC, Passman RS, et al. Cryptogenic stroke and underlying atrial fibrillation. N Engl J Med 2014;370:2478-86. 
  27. Himmelreich JC, Lucassen WA, Harskamp RE, Aussems C, van Weert HC, Nielen MM. CHARGE-AF in a national routine primary care electronic health records database in the Netherlands: validation for 5-year risk of atrial fibrillation and implications for patient selection in atrial fibrillation screening. Open Heart 2021;8:e001459. 
  28. Lip GY, Skjoth F, Nielsen PB, Larsen TB. Evaluation of the C2HEST risk score as a possible opportunistic screening tool for incident atrial fibrillation in a healthy population (from a Nationwide Danish Cohort Study). Am J Cardiol 2020;125:48-54. 
  29. Hu WS, Lin CL. Prediction of new-onset atrial fibrillation for general population in Asia: a comparison of C2HEST and HATCH scores. Int J Cardiol 2020;313:60-3. 
  30. Baek YS, Lee SC, Choi W, Kim DH. A new deep learning algorithm of 12-lead electrocardiogram for identifying atrial fibrillation during sinus rhythm. Sci Rep 2021;11:12818.