Automatic diagnosis of the 12-lead ECG using a deep neural network

Antônio H Ribeiro, Manoel Horta Ribeiro, Gabriela M M Paixão, Derick M Oliveira, Paulo R Gomes, Jéssica A Canazart, Milton P S Ferreira, Carl R Andersson, Peter W Macfarlane, Wagner Meira Jr, Thomas B Schön, Antonio Luiz P Ribeiro, Antônio H Ribeiro, Manoel Horta Ribeiro, Gabriela M M Paixão, Derick M Oliveira, Paulo R Gomes, Jéssica A Canazart, Milton P S Ferreira, Carl R Andersson, Peter W Macfarlane, Wagner Meira Jr, Thomas B Schön, Antonio Luiz P Ribeiro

Abstract

The role of automatic electrocardiogram (ECG) analysis in clinical practice is limited by the accuracy of existing models. Deep Neural Networks (DNNs) are models composed of stacked transformations that learn tasks by examples. This technology has recently achieved striking success in a variety of task and there are great expectations on how it might improve clinical practice. Here we present a DNN model trained in a dataset with more than 2 million labeled exams analyzed by the Telehealth Network of Minas Gerais and collected under the scope of the CODE (Clinical Outcomes in Digital Electrocardiology) study. The DNN outperform cardiology resident medical doctors in recognizing 6 types of abnormalities in 12-lead ECG recordings, with F1 scores above 80% and specificity over 99%. These results indicate ECG analysis based on DNNs, previously studied in a single-lead setup, generalizes well to 12-lead exams, taking the technology closer to the standard clinical practice.

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1. Abnormalities examples.
Fig. 1. Abnormalities examples.
A list of all the abnormalities the model classifies. We show only three representative leads (DII, V1 and V6).
Fig. 2. Precision-recall curve.
Fig. 2. Precision-recall curve.
Show precision-recall curve for our nominal prediction model on the test set (strong line) with regard to each ECG abnormalities. The shaded region shows the range between maximum and minimum precision for neural networks trained with the same configuration and different initialization. Points corresponding to the performance of resident medical doctors and students are also displayed, together with the point corresponding to the DNN performance for the same threshold used for generating Table 2. Gray dashed curves in the background correspond to iso-F1 curves (i.e. curves in the precision-recall plane with constant F1 score).
Fig. 3. (DNN architecture).
Fig. 3. (DNN architecture).
The unidimensional residual neural network architecture used for ECG classification.

References

    1. Roth GA, et al. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2018;392:1736–1788. doi: 10.1016/S0140-6736(18)32203-7.
    1. Willems JL, et al. Testing the performance of ECG computer programs: the CSE diagnostic pilot study. J. Electrocardiol. 1987;20(Suppl):73–77.
    1. Schläpfer J, Wellens HJ. Computer-interpreted electrocardiograms: benefits and limitations. J. Am. Coll. Cardiol. 2017;70:1183. doi: 10.1016/j.jacc.2017.07.723.
    1. Willems JL, et al. The diagnostic performance of computer programs for the interpretation of electrocardiograms. N. Engl. J. Med. 1991;325:1767–1773. doi: 10.1056/NEJM199112193252503.
    1. Shah AP, Rubin SA. Errors in the computerized electrocardiogram interpretation of cardiac rhythm. J. Electrocardiol. 2007;40:385–390. doi: 10.1016/j.jelectrocard.2007.03.008.
    1. Estes NAM. Computerized interpretation of ECGs: supplement not a substitute. Circulation. Arrhythmia Electrophysiol. 2013;6:2–4. doi: 10.1161/CIRCEP.111.000097.
    1. Krizhevsky, A., Sutskever, I. & Hinton, G. E. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems 1097–1105 (Curran Associates, Inc,. 2012).
    1. Hinton G, et al. Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Signal Process. Mag. 2012;29:82–97. doi: 10.1109/MSP.2012.2205597.
    1. Stead WW. Clinical implications and challenges of artificial intelligence and deep learning. JAMA. 2018;320:1107–1108. doi: 10.1001/jama.2018.11029.
    1. Naylor C. On the prospects for a (deep) learning health care system. JAMA. 2018;320:1099–1100. doi: 10.1001/jama.2018.11103.
    1. Hinton G. Deep learning—a technology with the potential to transform health care. JAMA. 2018;320:1101–1102. doi: 10.1001/jama.2018.11100.
    1. Bejnordi BE, et al. Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer. JAMA. 2017;318:2199. doi: 10.1001/jama.2017.14585.
    1. De Fauw J, et al. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat. Med. 2018;24:1342–1350. doi: 10.1038/s41591-018-0107-6.
    1. Beck EJ, Gill W, De Lay PR. Protecting the confidentiality and security of personal health information in low- and middle-income countries in the era of SDGs and Big Data. Glob. Health Action. 2016;9:32089. doi: 10.3402/gha.v9.32089.
    1. Hannun AY, et al. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nat. Med. 2019;25:65–69. doi: 10.1038/s41591-018-0268-3.
    1. Clifford GD, et al. AF classification from a short single lead ECG recording: the PhysioNet/Computing in Cardiology Challenge 2017. Comput. Cardiol. 2017;44:1–4.
    1. Mant J, et al. Accuracy of diagnosing atrial fibrillation on electrocardiogram by primary care practitioners and interpretative diagnostic software: analysis of data from screening for atrial fibrillation in the elderly (SAFE) trial. BMJ (Clin. Res. ed.) 2007;335:380. doi: 10.1136/.
    1. Veronese G, et al. Emergency physician accuracy in interpreting electrocardiograms with potential ST-segment elevation myocardial infarction: is it enough? Acute Card. Care. 2016;18:7–10. doi: 10.1080/17482941.2016.1234058.
    1. World Health Organization. Global Status Report on Noncommunicable Diseases 2014: Attaining the Nine Global Noncommunicable Diseases Targets; A Shared Responsibility OCLC: 907517003 (World Health Organization, Geneva, 2014).
    1. Sassi R, et al. PDF-ECG in clinical practice: a model for long-term preservation of digital 12-lead ECG data. J. Electrocardiol. 2017;50:776–780. doi: 10.1016/j.jelectrocard.2017.08.001.
    1. Lyon A, Mincholé A, Martínez JP, Laguna P, Rodriguez B. Computational techniques for ECG analysis and interpretation in light of their contribution to medical advances. J. R. Soc. Interface. 2018;15:20170821. doi: 10.1098/rsif.2017.0821.
    1. Alkmim MB, et al. Improving patient access to specialized health care: the Telehealth Network of Minas Gerais, Brazil. Bull. World Health Organ. 2012;90:373–378. doi: 10.2471/BLT.11.099408.
    1. He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In Proc. IEEE Conferenceon Computer Vision and Pattern Recognition (CVPR) 770–778 (IEEE, 2016).
    1. Hong, S. et al. ENCASE: an ENsemble ClASsifiEr for ECG classification using expert features and deep neural networks. In 2017 Computingin Cardiology Conference (IEEE, 2017).
    1. Kamaleswaran R, Mahajan R, Akbilgic O. A robust deep convolutional neural network for the classification of abnormal cardiac rhythm using single lead electrocardiograms of variable length. Physiological Meas. 2018;39:035006. doi: 10.1088/1361-6579/aaaa9d.
    1. Efron, B. & Tibshirani, R. J. An Introduction to the Bootstrap (CRC Press, 1994).
    1. McNemar Q. Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika. 1947;12:153–157. doi: 10.1007/BF02295996.
    1. Jambukia, S. H., Dabhi, V. K. & Prajapati, H. B. Classification of ECG signals using machine learning techniques: a survey. In Proc. International Conferenceon Advances in Computer Engineering and Applications (ICACEA) 714–721 (IEEE, 2015).
    1. Macfarlane PW, Devine B, Clark E. The University of Glasgow (Uni-G) ECG analysis program. Comput. Cardiol. 2005;32:451–454.
    1. Smith SW, et al. A deep neural network learning algorithm outperforms a conventional algorithm for emergency department electrocardiogram interpretation. J. Electrocardiol. 2019;52:88–95. doi: 10.1016/j.jelectrocard.2018.11.013.
    1. Cubanski D, Cyganski D, Antman EM, Feldman CL. A neural network system for detection of atrial fibrillation in ambulatory electrocardiograms. J. Cardiovasc. Electrophysiol. 1994;5:602–608. doi: 10.1111/j.1540-8167.1994.tb01301.x.
    1. Tripathy RK, Bhattacharyya A, Pachori RB. A novel approach for detection of myocardial infarction from ECG signals of multiple electrodes. IEEE Sens. J. 2019;19:4509–4517. doi: 10.1109/JSEN.2019.2896308.
    1. Rubin, J., Parvaneh, S., Rahman, A., Conroy, B. & Babaeizadeh, S. Densely connected convolutional networks and signal quality analysis to detect atrial fibrillation using short single-lead ECG recordings. pp. 1–4 (Computing in Cardiology (CinC), Rennes, 2017). .
    1. Acharya UR, et al. Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals. Inf. Sci. 2017;415-416:190–198. doi: 10.1016/j.ins.2017.06.027.
    1. Shashikumar, S. P., Shah, A. J., Clifford, G. D. & Nemati, S. Detection of paroxysmal atrial fibrillation using attention-based bidirectional recurrent neural networks. In Proc. 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD ’18, New York, NY, USA, (eds Guo, Y. & Farooq, F.) 715–723 (ACM, 2018).
    1. Rahhal MA, et al. Deep learning approach for active classification of electrocardiogram signals. Inf. Sci. 2016;345:340–354. doi: 10.1016/j.ins.2016.01.082.
    1. Goldberger AL, et al. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. Circulation. 2000;101:E215–E220.
    1. Goto S, et al. Artificial intelligence to predict needs for urgent revascularization from 12-leads electrocardiography in emergency patients. PLoS ONE. 2019;14:e0210103. doi: 10.1371/journal.pone.0210103.
    1. Rautaharju PM, Surawicz B, Gettes LS. AHA/ACCF/HRS Recommendations for the Standardization and Interpretation of the Electrocardiogram: Part IV: The ST Segment, T and U Waves, and the QT Interval A Scientific Statement From the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society Endorsed by the International Society for Computerized Electrocardiology. J. Am. Coll. Cardiol. 2009;53:982–991. doi: 10.1016/j.jacc.2008.12.014.
    1. Luo S, Johnston P. A review of electrocardiogram filtering. J. Electrocardiol. 2010;43:486–496. doi: 10.1016/j.jelectrocard.2010.07.007.
    1. Nascimento BR, Brant LCC, Marino BCA, Passaglia LG, Ribeiro ALP. Implementing myocardial infarction systems of care in low/middle-income countries. Heart. 2019;105:20. doi: 10.1136/heartjnl-2018-313398.
    1. Macfarlane P, et al. Methodology of ECG interpretation in the Glasgow program. Methods Inf. Med. 1990;29:354–361. doi: 10.1055/s-0038-1634799.
    1. Macfarlane PW, Latif S. Automated serial ECG comparison based on the Minnesota code. J. Electrocardiol. 1996;29:29–34. doi: 10.1016/S0022-0736(96)80016-1.
    1. Prineas, R. J., Crow, S. & Zhang, Z. M. The Minnesota Code Manual of Electrocardiographic Findings (Springer Science & Business Media, 2009).
    1. Veloso, A., Meira, W. Jr. & Zaki, M. J. Lazy associative classification. In Proc. 6th International Conference on Data Mining (ICDM) 645–654 (IEEE Computer Society, 2006).
    1. Cohen J. A coefficient of agreement for nominal scales. Educ. Psychological Meas. 1960;20:37–46. doi: 10.1177/001316446002000104.
    1. Kligfield P, et al. Recommendations for the standardization and interpretation of the electrocardiogram. J. Am. Coll. Cardiol. 2007;49:1109. doi: 10.1016/j.jacc.2007.01.024.
    1. He, K., Zhang, X., Ren, S. & Sun, J. Identity mappings in deep residual networks. In Computer Vision—ECCV 2016 (eds Leibe, B. et al.) 630–645 (Springer International Publishing, 2016).
    1. Ioffe, S. & Szegedy, C. Batch normalization: accelerating deep network training by reducing internal covariate shift. In Proc. 32nd International Conference on Machine Learning (eds Bach, F. & Blei, D.) 448–456 (, PMLR, 2015).
    1. Srivastava N, Hinton GE, Krizhevsky A, Sutskever I, Salakhutdinov R. Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 2014;15:1929–1958.
    1. Hutchison, D. et al. Evaluation of pooling operations in convolutional architectures for object recognition. In Artificial Neural Networks—ICANN 2010 Vol. 6354 (eds Diamantaras, K. et al.) 92–101 (Springer, Berlin, 2010).
    1. Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. In Proc. 3rd International Conference for Learning Representations (ICLR) (eds Bengio, Y. & LeCun, Y.) (Conference Track Proceedings, San Diego, CA, USA, 2015). .
    1. He, K., Zhang, X., Ren, S. & Sun, J. Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In Proc. IEEE International Conference on Computer Vision 1026–1034 (2015).
    1. Saito T, Rehmsmeier M. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS ONE. 2015;10:e0118432. doi: 10.1371/journal.pone.0118432.

Source: PubMed

Sponsorit ja yhteistyökumppanit

Sairaudet

Huumeiden interventiot

3
Tilaa