Abstract
The heart murmur associated with atrial septal defects is often faint and can thus only be detected by chance. Although electrocardiogram examination can prompt diagnoses, identification of specific findings remains a major challenge. We demonstrate improved diagnostic accuracy realized by incorporating a proposed deep learning model, comprising a convolutional neural network (CNN) and long short-term memory (LSTM), with electrocardiograms. This retrospective observational study included 1192 electrocardiograms of 728 participants from January 1, 2000, to December 31, 2017, at Tokyo Women's Medical University Hospital. Using echocardiography, we confirmed the status of healthy subjects—no structural heart disease—and the diagnosis of atrial septal defects in patients. We used a deep learning model comprising a CNN and LTSMs. All pediatric cardiologists (n = 12) were blinded to patient groupings when analyzing them by electrocardiogram. Using electrocardiograms, the model’s diagnostic ability was compared with that of pediatric cardiologists. We assessed 1192 electrocardiograms (828 normally structured hearts and 364 atrial septal defects) pertaining to 792 participants. The deep learning model results revealed that the accuracy, sensitivity, specificity, positive predictive value, and F1 score were 0.89, 0.76, 0.96, 0.88, and 0.81, respectively. The pediatric cardiologists (n = 12) achieved means of accuracy, sensitivity, specificity, positive predictive value, and F1 score of 0.58 ± 0.06, 0.53 ± 0.04, 0.67 ± 0.10, 0.69 ± 0.18, and 0.58 ± 0.06, respectively. The proposed method is a superior alternative to accurately diagnose atrial septal defects.
Similar content being viewed by others
References
Reller MD, Strickland MJ, Riehle-Colarusso T, Mahle WT, Correa A (2008) Prevalence of congenital heart defects in metropolitan Atlanta, 1998–2005. J Pediatr 153:807–813
van der Linde D, Konings EEM, Slager MA, Witsenburg M, Helbing WA, Takkenberg JJ, Roos-Hesselink JW (2011) Birth prevalence of congenital heart disease worldwide. J Am Coll Cardiol 58:2241–2247
Martin SS, Shapiro EP, Mukherjee M (2015) Atrial septal defects – clinical manifestations, echo assessment, and intervention. Clinical Medicine Insights: Cardiology. London, UK: SAGE Publications) 8s1:CMC.S15715.
Attie F, Rosas M, Granados N, Zabal C, Buendía A, Calderón J (2001) Surgical treatment for secundum atrial septal defects in patients >40 years old. J Am Coll Cardiol Elsevier Masson SAS 38:2035–2042
Endorsed by the European Society of Gynecology (ESG), the Association for European Paediatric Cardiology (AEPC), and the German Society for Gender Medicine (DGesGM), Authors/Task Force Members, Regitz-Zagrosek V, Blomstrom Lundqvist C, Borghi C, Cifkova R, Ferreira R, Foidart JM, Gibbs JSR, Gohlke-Baerwolf C, Gorenek B, Iung B, Kirby M, Maas AHEM, Morais J, Nihoyannopoulos P, Pieper PG, Presbitero P, Roos-Hesselink JW, Schaufelberger M, Seeland U, Torracca L, ESC Committee for Practice Guidelines (CPG), Bax J, Auricchio A, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, et al. (2011) ESC Guidelines on the management of cardiovascular diseases during pregnancy: The Task Force on the Management of Cardiovascular Diseases during Pregnancy of the European Society of Cardiology (ESC). Eur Heart J 32:3147–3197.
Rigatelli G, Dell’avvocata F, Tarantini G, Giordan M, Cardaioli P, Nguyen T (2014) Clinical, hemodynamic, and intracardiac echocardiographic characteristics of secundum atrial septal defects-related paradoxical embolism in adulthood. J Interv Cardiol Wiley/Blackwell 27:542–547
Endorsed by the Association for European Paediatric Cardiology (AEPC), Authors/Task Force Members, Baumgartner H, Bonhoeffer P, De Groot NMS, de Haan F, Deanfield JE, Galie N, Gatzoulis MA, Gohlke-Baerwolf C, Kaemmerer H, Kilner P, Meijboom F, Mulder BJM, Oechslin E, Oliver JM, Serraf A, Szatmari A, Thaulow E, Vouhe PR, Walma E, ESC Committee for Practice Guidelines (CPG), Vahanian A, Auricchio A, Bax J, Ceconi C, Dean V, Filippatos G, Funck-Brentano C, Hobbs R et al (2010) ESC Guidelines for the management of grown-up congenital heart disease (new version 2010): The Task Force on the Management of Grown-up Congenital Heart Disease of the European Society of Cardiology (ESC). Eur Heart J 31:2915–2957
Silversides CK, Dore A, Poirier N, Taylor D, Harris L, Greutmann M, Benson L, Baumgartner H, Celermajer D, Therrien J (2010) Canadian Cardiovascular Society 2009 Consensus conference on the management of adults with congenital heart disease: Shunt lesions. Can J Cardiol 26:e70–e79
Warnes CA, Williams RG, Bashore TM et al (2008) ACC/AHA 2008 Guidelines for the management of adults with congenital heart disease. J Am Coll Cardiol Found 52:e143–e263
Murphy JG, Gersh BJ, McGoon MD, Mair DD, Porter CB, Ilstrup DM, McGoon DC, Puga FJ, Kirklin JW, Danielson GK (1990) Long-term outcome after surgical repair of isolated atrial septal defect. Follow-up at 27 to 32 years. N Engl J Med Massachusetts Med Soc 323:1645–1650
Roos-Hesselink JW, Meijboom FJ, Spitaels SEC, Van Domburg R, Van Rijen EH, Utens EM, Bogers AJ, Simoons ML (2003) Excellent survival and low incidence of arrhythmias, stroke and heart failure long-term after surgical ASD closure at young age. A prospective follow-up study of 21–33 years. Eur Heart J 24:190–197
Webb G, Gatzoulis MA (2006) Atrial septal defects in the adult. Circulation 114:1645–1653
Heller J, Hagège AA, Besse B, Desnos M, Marie FN, Guerot C (1996) ‘Crochetage’ (notch) on R wave in inferior limb leads: a new independent electrocardiographic sign of atrial septal defect. J Am Coll Cardiol 27:877–882
Cohen JS, Patton DJ, Giuffre RM (2000) The crochetage pattern in electrocardiograms of pediatric atrial septal defect patients. Can J Cardiol 16:1241–1247
Nakamura T, Saitou T, Nakayama Y, Hashida Y, Ota K (2013) Usefulness of the crochetage pattern in atrial septal defect patients. Pediatr Cardiol Card Surg 29:322–327
Goto S, Kimura M, Katsumata Y, Goto S, Kamatani T, Ichihara G, Ko S, Sasaki J, Fukuda K, Sano M (2019) Artificial intelligence to predict needs for urgent revascularization from 12-leads electrocardiography in emergency patients. PLoS ONE 14:e0210103
Ghongade R, Deshmukh M, Joshi D (2014) Arrhythmia classification using morphological features and probabilistic neural networks. IEEE 80–84.
Rajpurkar P, Hannun AY, Haghpanahi M, Bourn C, Ng AY (2017) Cardiologist-level arrhythmia detection with convolutional neural networks. ArXiv 1–9.
Oh SL, Ng EYK, Tan RS, Acharya UR (2018) Automated diagnosis of arrhythmia using combination of CNN and LSTM techniques with variable length heart beats. Comput Biol Med 102:278–287
Yildirim Ö (2018) A novel wavelet sequence based on deep bidirectional LSTM network model for ECG signal classification. Comput Biol Med 96:189–202
Alom MZ, Taha TM, Yakopcic C, Westberg S, Sidike P, Nasrin MS, Hasan M, Van Essen BC, Awwal AA, Asari VK (2019) A state-of-the-art survey on deep learning theory and architectures. Electronics 8:292–367
Ciresan DC, Meier U, Gambardella LM, Schmidhuber J (2011) Flexible, high performance convolutional neural networks for image classification. IJCAI Proceed Twenty-Seci Int Joint Conf Artific Intellig 2:1–6
Wang W, Yang Y (2019) Development of convolutional neural network and its application in image classification: a survey. Opt Eng 58:1–20
Hu C, Wu Q, Li H, Jian S, Li N, Lou Z (2018) Deep learning with a long short-term memory networks approach for rainfall-runoff simulation. Water 10:1543–1616
Bellec G, Subramoney A, Legenstein R, Maass W (2019) Long short-term memory and learning-to-learn in networks of spiking neurons. In Conference on Neural Information Processing Systems NeurIPS, Montreal, Canada 1–11.
Bao W, Yue J, Rao Y (2017) A deep learning framework for financial time series using stacked autoencoders and long-short term memory. Podobnik B, ed. PLoS ONE 12:e0180944-e181024
Kohavi R (1995) A study of cross validation and bootstrap for accuracy estimation and model selection. International Joint Conference on Artificial Intelligence 1–7.
Schuster M, Paliwal KK (1997) Bidirectional recurrent neural networks. IEEE Trans Signal Proc 45.
Hurst JW (1998) Current perspective naming of the waves in the ECG, with a brief account of their genesis. Circulation 98:1937–1942
Guo Z, Li X, Huang H, Guo N, Li Q (2017) Medical image segmentation based on multi-modal convolutional neural network: Study on image fusion schemes. ArXiv 1711.00049v2.
Hoffman JIE, Kaplan S (2002) The incidence of congenital heart disease. J Am Coll Cardiol 39:1890–1900
Roushdy AM, Attia H, Nossir H (2018) Immediate and short term effects of percutaneous atrial septal defect device closure on cardiac electrical remodeling in children. Egypt Heart J 70:243–247
Davies DH, Pryor R, Blount SG (1960) Electrocardiographic changes in atrial septal defect following surgical correction. Br Heart J 22:274–280
Kristjánsson Á, Ásgeirsson ÁG (2018) Attentional priming: recent insights and current controversies. Curr Opin Psychol 29:71–75
McCambridge J, Witton J, Elbourne DR (2014) Systematic review of the Hawthorne effect: new concepts are needed to study research participation effects. J Clin Epidemiol 67:267–277
Acknowledgments
We would like to express much appreciation to Dr. Nagata, Dr. Hattori, and Dr. Hagiwara for their valuable and constructive suggestions during the planning and development of this work. We would also like to thank Editage for English language editing.
Funding
This work was supported in part by the 4th Miyata Foundation Award, for which we thank Mr. Miyata.
Author information
Authors and Affiliations
Contributions
HM conceived of the presented idea. HM developed the theory and performed the computations. HS and KI verified the analytical methods. YM encouraged HM to investigate and supervised the findings of this work.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no conflict of interest to declare.
Ethical Approval
All authors discussed the results and contributed to the final manuscript.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
All the authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Mori, H., Inai, K., Sugiyama, H. et al. Diagnosing Atrial Septal Defect from Electrocardiogram with Deep Learning. Pediatr Cardiol 42, 1379–1387 (2021). https://doi.org/10.1007/s00246-021-02622-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00246-021-02622-0