FREE SHIPPING over $40
OVER 9000 FIVE-STAR REVIEWS
5% OFF ALL SUBSCRIPTIONS
100% HAPPINESS GUARANTEE
Analyze Diet
0
Home / Equine Research Bank / Function (Oxf) / ECG Restitution Analysis and Machine Learning to Detect Paro...
Function (Oxford, England)2020; 2(1); zqaa031; doi: 10.1093/function/zqaa031

ECG Restitution Analysis and Machine Learning to Detect Paroxysmal Atrial Fibrillation: Insight from the Equine Athlete as a Model for Human Athletes.

Abstract: Atrial fibrillation is the most frequent arrhythmia in both equine and human athletes. Currently, this condition is diagnosed via electrocardiogram (ECG) monitoring which lacks sensitivity in about half of cases when it presents in paroxysmal form. We investigated whether the arrhythmogenic substrate present between the episodes of paroxysmal atrial fibrillation (PAF) can be detected using restitution analysis of normal sinus-rhythm ECGs. In this work, ECG recordings were obtained during routine clinical work from control and horses with PAF. The extracted QT, TQ, and RR intervals were used for ECG restitution analysis. The restitution data were trained and tested using k-nearest neighbor (k-NN) algorithm with various values of neighbors k to derive a discrimination tool. A combination of QT, RR, and TQ intervals was used to analyze the relationship between these intervals and their effects on PAF. A simple majority vote on individual record (one beat) classifications was used to determine the final classification. The k-NN classifiers using two-interval measures were able to predict the diagnosis of PAF with area under the receiving operating characteristic curve close to 0.8 (RR, TQ with k ≥ 9) and 0.9 (RR, QT with k ≥ 21 or TQ, QT with k ≥ 25). By simultaneously using all three intervals for each beat and a majority vote, mean area under the curves of 0.9 were obtained for all tested k-values (3-41). We concluded that 3D ECG restitution analysis can potentially be used as a metric of an automated method for screening of PAF.
© The Author(s) 2020. Published by Oxford University Press on behalf of American Physiological Society.
Get Full Text
Publication Date: 2020-11-18 PubMed ID: 35330977PubMed Central: PMC8788737DOI: 10.1093/function/zqaa031Google Scholar: Lookup
The Equine Research Bank provides access to a large database of publicly available scientific literature. Inclusion in the Research Bank does not imply endorsement of study methods or findings by Mad Barn.
LinkedInCopy
  • Journal Article
  • Research Support
  • Non-U.S. Gov't

Summary

This research summary has been generated with artificial intelligence and may contain errors and omissions. Refer to the original study to confirm details provided. Submit correction.

This research paper investigates a new method of detecting paroxysmal atrial fibrillation (PAF) in horses and potentially humans by using ECG restitution analysis and machine learning. The researchers found that this methodology predicted PAF with high accuracy, indicating its potential as an alternative screening tool to standard electrocardiogram (ECG) monitoring.

Research Methodology & Data Collection

  • The researchers gathered ECG recordings from horses during routine clinical practices. These horses were either controls or afflicted with PAF.
  • The ECG recordings were then analyzed for patterns in the QT, TQ, and RR intervals – elements that are essential in diagnosing conditions related to heart rhythm.
  • Data from this analysis, known as ECG restitution analysis, were used for training a machine learning model.

Machine Learning Analysis

  • A k-nearest neighbor (k-NN) algorithm was used for the machine learning model. This model was trained and tested on the restitution data, with different values of neighbors used in an experiment to optimize the model’s performance.
  • The purpose of this model was to develop a tool that could distinguish between horses with and without PAF, based purely on the extracted ECG restitution data.

Results of the Analysis

  • The relationship between the QT, RR, and TQ intervals and their impact on PAF was analyzed.
  • The k-NN classifiers, when using two-interval measures, were found to predict PAF diagnosis with high accuracy – an area under the receiving operating characteristic curve close to 0.8 and 0.9.
  • Even better results were obtained when using all three interval measures simultaneously, and by using a majority voting system for final beat classification; in this case, the mean area under the curve was 0.9, indicative of a high level of diagnostic accuracy.

Conclusion

  • The study concluded that 3D ECG restitution analysis, paired with machine learning algorithms, could be used as a potential method for PAF screening.
  • The high level of accuracy suggests its viability as an automated method that may provide more precise detection in comparison to conventional ECG monitoring.
  • While the study was conducted on equine subjects, the findings could hold potential applicability for human athletes as well.

Cite This Article

APA
Huang YH, Alexeenko V, Tse G, Huang CL, Marr CM, Jeevaratnam K. (2020). ECG Restitution Analysis and Machine Learning to Detect Paroxysmal Atrial Fibrillation: Insight from the Equine Athlete as a Model for Human Athletes. Function (Oxf), 2(1), zqaa031. https://doi.org/10.1093/function/zqaa031

Publication

Function (Oxford, England)
ISSN: 2633-8823
NlmUniqueID: 101770668
Country: England
Language: English
Volume: 2
Issue: 1
Pages: zqaa031

Researcher Affiliations

Huang, Ying H
  • Faculty of Health and Medical Sciences, University of Surrey, Guildford, GU2 7AL, UK.
Alexeenko, Vadim
  • Faculty of Health and Medical Sciences, University of Surrey, Guildford, GU2 7AL, UK.
Tse, Gary
  • Tianjin Key Laboratory of Ionic-Molecular Function of Cardiovascular Disease, Department of Cardiology, Tianjin Institute of Cardiology, the Second Hospital of Tianjin Medical University, Tianjin, China.
Huang, Christopher L-H
  • Faculty of Health and Medical Sciences, University of Surrey, Guildford, GU2 7AL, UK.
  • Physiological Laboratory, University of Cambridge, Cambridge, CB2 1QW, UK.
Marr, Celia M
  • Rossdales Equine Hospital and Diagnostic Centre, Exning, CB8 7NN, Suffolk, UK.
Jeevaratnam, Kamalan
  • Faculty of Health and Medical Sciences, University of Surrey, Guildford, GU2 7AL, UK.
  • Physiological Laboratory, University of Cambridge, Cambridge, CB2 1QW, UK.

MeSH Terms

  • Humans
  • Horses
  • Animals
  • Atrial Fibrillation / diagnosis
  • Electrocardiography / methods
  • Heart Rate
  • Machine Learning

References

This article includes 41 references
  1. Kirchhof P, Benussi S, Kotecha D. 2016 ESC Guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Europace 2016;18(11):1609–1678.
    pubmed: 27567465
  2. Chandra N, Bastiaenen R, Papadakis M, Sharma S. Sudden cardiac death in young athletes: practical challenges and diagnostic dilemmas. J Am Coll Cardiol 2013;61(10):1027–1040.
    pubmed: 23473408
  3. Sharma S, Drezner JA, Baggish A. International recommendations for electrocardiographic interpretation in athletes. J Am Coll Cardiol 2017;69(8):1057–1075.
    pubmed: 28231933doi: 10.1016/j.jacc.2017.01.015google scholar: lookup
  4. Nishida K, Michael G, Dobrev D, Nattel S. Animal models for atrial fibrillation: clinical insights and scientific opportunities. Europace 2010;12(2):160–172.
    pubmed: 19875395
  5. Frydrychowski P, Michałek M, Sławuta A, Noszczyk-Nowak A. Large animals as models of atrial fibrillation. Adv Clin Exp Med 2020;29(6):757–767.
    pubmed: 32603556
  6. Howlett PJ, Hatch FS, Alexeenko V, Jabr RI, Leatham EW, Fry CH. Diagnosing paroxysmal atrial fibrillation: are biomarkers the solution to this elusive arrhythmia?. Biomed Res Int 2015;2015(2314–6141 (Electronic)):910267.
    pmc: PMC4502272pubmed: 26229966
  7. Thijs V. Atrial fibrillation detection fishing for an irregular heartbeat before and after stroke. Stroke 2017;48(10):2671–2677.
    pubmed: 28916671
  8. Censi F, Calcagnini G, Mattei E, Gargaro A, Biancalana G, Capucci A. Simulation of monitoring strategies for atrial arrhythmia detection. Ann Ist Super Sanita 2013;49(2):176–182.
    pubmed: 23771262
  9. Tu HT, Spence S, Kalman JM, Davis SM. Twenty-eight day Holter monitoring is poorly tolerated and insensitive for paroxysmal atrial fibrillation detection in cryptogenic stroke. Intern Med J 2014;44(5):505–508.
    pubmed: 24816310
  10. Attia ZI, Noseworthy PA, Lopez-Jimenez F. 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(10201):861–867.
    pubmed: 31378392doi: 10.1016/s0140-6736(19)31721-0google scholar: lookup
  11. Li M, Chadda KR, Matthews GDK, Marr CM, Huang CL-H, Jeevaratnam K. Cardiac electrophysiological adaptations in the equine athlete-restitution analysis of electrocardiographic features. PLoS One 2018;13(3):e0194008.
    pmc: PMC5844547pubmed: 29522557
  12. Wyse DG, Van Gelder IC, Ellinor PT. Lone atrial fibrillation: does it exist?. J Am Coll Cardiol 2014;63(17):1715–1723.
    pmc: PMC4008692pubmed: 24530673
  13. Fossa AA, Wisialowski T, Magnano A. Dynamic beat-to-beat modeling of the QT-RR interval relationship: analysis of QT prolongation during alterations of autonomic state versus human ether a-go-go-related gene inhibition. J Pharmacol Exp Ther 2005;312(1):1–11.
    pubmed: 15306635
  14. Cunningham P, Delany SJ. k-Nearest neighbour classifiers. Mult Classif Syst 2007;34(8):1–17.
  15. Alexeenko V, Fraser JA, Dolgoborodov A. The application of Lempel-Ziv and Titchener complexity analysis for equine telemetric electrocardiographic recordings. Sci Rep 2019;9(1).
    pmc: PMC6385502pubmed: 30796330doi: 10.1038/s41598-019-38935-7google scholar: lookup
  16. Alexeenko V, Fraser JA, Bowen M, Huang CLH, Marr CM, Jeevaratnam K. The complexity of clinically-normal sinus-rhythm ECGs is decreased in equine athletes with a diagnosis of paroxysmal atrial fibrillation. Sci Rep 2020;10(1).
    pmc: PMC7176685pubmed: 32321950doi: 10.1038/s41598-020-63343-7google scholar: lookup
  17. R Core Team. R: A Language and Environment for Statistical Computing. .
  18. Meek S. ABC of clinical electrocardiography: Introduction. I—Leads, rate, rhythm, and cardiac axis. BMJ 2002;324(7334):415–418.
    pmc: PMC1122339pubmed: 11850377
  19. Kligfield P, Okin PM. Prevalence and clinical implications of improper filter settings in routine electrocardiography. Am J Cardiol 2007;99(5):711–713.
    pubmed: 17317378
  20. Leroux AA, Detilleux J, Sandersen CF. Prevalence and risk factors for cardiac diseases in a hospital-based population of 3,434 horses (1994-2011). J Vet Intern Med 2013;27(6):1563–1570.
    pubmed: 24112454
  21. Ohmura H, Hiraga A, Takahashi T, Kai M, Jones JH. Risk factors for atrial fibrillation during racing in slow-finishing horses. J Am Vet Med Assoc 2003;223(1):84–88.
    pubmed: 12839069
  22. He Haibo, Garcia EA. Learning from Imbalanced Data. IEEE Trans Knowl Data Eng 2009;21(9):1263–1284.
  23. Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP. SMOTE: synthetic minority over-sampling technique. J Artif Intell Res 2002;16(8):321–357.
  24. Kuhn M, Johnson K. Applied Predictive Modeling. Vol 26. New York: Springer; 2013.
  25. Zou KH, O’Malley AJ, Mauri L. Receiver-operating characteristic analysis for evaluating diagnostic tests and predictive models. Circulation 2007;115(5):654–657.
    pubmed: 17283280
  26. Wilcoxon F. Individual comparisons by ranking methods. In: Kotz S, Johnson NL, eds. Breakthroughs in Statistics. New York: Springer, 1992:196–202.
  27. Goldberger AL, Amaral LAN, Glass L. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. Circulation 2000;101(23):215–220.
    pubmed: 10851218
  28. Elhaj FA, Salim N, Harris AR, Swee TT, Ahmed T. Arrhythmia recognition and classification using combined linear and nonlinear features of ECG signals. Comput Methods Programs Biomed 2016;127:52–63.
    pubmed: 27000289
  29. Oh SL, Ng EYK, Tan RS, Acharya UR. Automated diagnosis of arrhythmia using combination of CNN and LSTM techniques with variable length heart beats. Comput Biol Med 2018;102(April):278–287.
    pubmed: 29903630
  30. Lown M, Brown M, Brown C. Machine learning detection of atrial fibrillation using wearable technology. PLoS One 2020;15(1):e0227401.
    pmc: PMC6980577pubmed: 31978173
  31. Van Steenkiste G, van Loon G, Crevecoeur G. Transfer learning in ECG classification from human to horse using a novel parallel neural network architecture. Sci Rep 2020;10(1):1–12.
    pmc: PMC6957496pubmed: 31932667
  32. Milani-Nejad N, Janssen PML. Small and large animal models in cardiac contraction research: advantages and disadvantages. Pharmacol Ther 2014;141(3):235–249.
    pmc: PMC3947198pubmed: 24140081
  33. Ohmura H, Jones JH. Changes in heart rate and heart rate variability as a function of age in Thoroughbred horses. J equine Sci 2017;28(3):99–103.
    pmc: PMC5608962pubmed: 28955161
  34. Marr CM, Bowen M. Cardiology of the Horse. Edinburgh, UK: Saunders-Elsevier; 2010.
    doi: 10.1016/c2009-0-48166-2google scholar: lookup
  35. Fagard R. Athlete’s heart. Heart 2003;89(12):1455–1461.
    pmc: PMC1767992pubmed: 14617564
  36. Tanaka H, Monahan KD, Seals DR. Age-predicted maximal heart rate revisited. J Am Coll Cardiol 2001;37(1):153–156.
    pubmed: 11153730
  37. Munoz ML, van Roon A, Riese H. Validity of (ultra-)short recordings for heart rate variability measurements. PLoS One 2015;10(9):e0138921.
    pmc: PMC4586373pubmed: 26414314
  38. Nicolson WB, McCann GP, Brown PD. A novel surface electrocardiogram–based marker of ventricular arrhythmia risk in patients with ischemic cardiomyopathy. J Am Heart Assoc 2012;1(4):1–10.
    pmc: PMC3487358pubmed: 23130163
  39. Fossa AA. Beat-to-beat ECG restitution: a review and proposal for a new biomarker to assess cardiac stress and ventricular tachyarrhythmia vulnerability. Ann Noninvasive Electrocardiol 2017;22(5):1–11.
    pmc: PMC6931610pubmed: 28497858
  40. Fossa AA, Zhou M, Robinson A, Purkayastha J, Martin P. Use of ECG restitution (beat-to-beat QT-TQ interval analysis) to assess arrhythmogenic risk of QTc prolongation with guanfacine. Ann Noninvasive Electrocardiol 2014;19(6):582–594.
    pmc: PMC6931917pubmed: 25200912
  41. Hesselkilde EZ, Carstensen H, Haugaard MM. Effect of flecainide on atrial fibrillatory rate in a large animal model with induced atrial fibrillation. BMC Cardiovasc Disord 2017;17(1):289.
    pmc: PMC5723027pubmed: 29221440

Citations

This article has been cited 9 times.
  1. Kapusniak A, Lara NM, Hitchens PL, Bailey S, Nath L, Franklin S. Use of Artificial Intelligence to Detect Cardiac Rhythm Disturbances in Athletes: A Scoping Review. J Vet Intern Med 2025 Nov-Dec;39(6):e70257.
    doi: 10.1111/jvim.70257pubmed: 41017277google scholar: lookup
  2. Akbarein H, Taaghi MH, Mohebbi M, Soufizadeh P. Applications and Considerations of Artificial Intelligence in Veterinary Sciences: A Narrative Review. Vet Med Sci 2025 May;11(3):e70315.
    doi: 10.1002/vms3.70315pubmed: 40173266google scholar: lookup
  3. Flanders WH, Moïse NS, Otani NF. Use of machine learning and Poincaré density grid in the diagnosis of sinus node dysfunction caused by sinoatrial conduction block in dogs. J Vet Intern Med 2024 May-Jun;38(3):1305-1324.
    doi: 10.1111/jvim.17071pubmed: 38682817google scholar: lookup
  4. Chung CT, Lee S, King E, Liu T, Armoundas AA, Bazoukis G, Tse G. Clinical significance, challenges and limitations in using artificial intelligence for electrocardiography-based diagnosis. Int J Arrhythmia 2022;23(1):24.
    doi: 10.1186/s42444-022-00075-xpubmed: 36212507google scholar: lookup
  5. Stracina T, Ronzhina M, Redina R, Novakova M. Golden Standard or Obsolete Method? Review of ECG Applications in Clinical and Experimental Context. Front Physiol 2022;13:867033.
    doi: 10.3389/fphys.2022.867033pubmed: 35547589google scholar: lookup
  6. Huang YH, Lyle JV, Ab Razak AS, Nandi M, Marr CM, Huang CL, Aston PJ, Jeevaratnam K. Detecting paroxysmal atrial fibrillation from normal sinus rhythm in equine athletes using Symmetric Projection Attractor Reconstruction and machine learning. Cardiovasc Digit Health J 2022 Apr;3(2):96-106.
    doi: 10.1016/j.cvdhj.2022.02.001pubmed: 35493267google scholar: lookup
  7. Ramírez J, Tinker A. Ventricular Restitution Predicts Paroxysmal Atrial Fibrillation in Horses. Function (Oxf) 2021;2(1):zqaa038.
    doi: 10.1093/function/zqaa038pubmed: 35330978google scholar: lookup
  8. Kjeldsen ST, Nissen SD, Buhl R, Hopster-Iversen C. Paroxysmal Atrial Fibrillation in Horses: Pathophysiology, Diagnostics and Clinical Aspects. Animals (Basel) 2022 Mar 10;12(6).
    doi: 10.3390/ani12060698pubmed: 35327097google scholar: lookup
  9. Tse G, Hao G, Lee S, Zhou J, Zhang Q, Du Y, Liu T, Cheng SH, Wong WT. Measures of repolarization variability predict ventricular arrhythmogenesis in heptanol-treated Langendorff-perfused mouse hearts. Curr Res Physiol 2021;4:125-134.
    doi: 10.1016/j.crphys.2021.04.001pubmed: 34746832google scholar: lookup
Subscribe to our newsletter
Join 99,344 horse owners like you who receive our email updates on horse health and nutrition.