Abstract
Atrial fibrillation is the most commonly encountered cardiac arrhythmia, increasing stroke risk and mortality. Paroxysmal atrial fibrillation (PAF) can be challenging to detect because arrhythmias occur intermittently. We have been able to classify PAF patients from sinus rhythm electrocardiograms (ECG), using a signal processing technique, Symmetric Projector Attractor Reconstruction, which transforms the ECG time-series into a quantifiable two-dimensional image termed an attractor. To optimise this methodology, we investigated the impact of varying parameters within the SPAR method, choice of lead, ECG sampling frequency and machine learning model choice. We determined that using a K nearest neighbours (KNN) model with 125 Hz ECG sampling frequency, using specific features that quantified the density of the attractor, gave a classification accuracy of 81.2%. When using a decision tree model it was found that the sensitivity was 72.5% which shows an obvious improvement over 30-day long term monitoring which has a sensitivity of 34%. The results of this paper present consideration for the application of a new method to the clinically relevant problem of aiding detection of PAF and conjectures as to the reasoning behind these results. Further investigation in larger cohorts is needed to fully elucidate these findings.
Data availability
The datasets analysed for this study can be found on the Physionet website (https:/physionet.org/content/ptb-xl/1.0.3) .
References
Gladstone, D. J. et al. Atrial fibrillation in patients with cryptogenic stroke. N. Engl. J. Med. 370, 2467–2477. https://doi.org/10.1056/NEJMoa1311376 (2014).
Sanna, T. et al. Cryptogenic stroke and underlying atrial fibrillation. N. Engl. J. Med. 370, 2478–2486. https://doi.org/10.1056/NEJMoa1313600 (2014).
Ding, M., Ebeling, M., Ziegler, L., Wennberg, A. & Modig, K. Time trends in atrial fibrillation-related stroke during 2001–2020 in sweden: a nationwide, observational study. Lancet Reg. Health - Europe. 28, 100596. https://doi.org/10.1016/j.lanepe.2023.100596 (2023).
Alonso, A., Almuwaqqat, Z. & Chamberlain, A. Mortality in atrial fibrillation. Is it changing? Trends Cardiovasc. Med. 31, 469–473. https://doi.org/10.1016/j.tcm.2020.10.010 (2021).
Kjerpeseth, L. J. et al. Prevalence and incidence rates of atrial fibrillation in Norway 2004–2014. Heart 107, 201–207. https://doi.org/10.1136/heartjnl-2020-316624 (2021).
Hindricks, G. et al. (2021). M.G.D. Bates, N.U. Zakirov, 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS), Eur. Heart J. 42 373–498. https://doi.org/10.1093/eurheartj/ehaa612.
Al-Makhamreh, H. et al. Paroxysmal and Non-Paroxysmal atrial fibrillation in middle Eastern patients: clinical features and the use of Medications. Analysis of the Jordan atrial fibrillation (JoFib) study. Int. J. Environ. Res. Public. Health. 19, 6173. https://doi.org/10.3390/ijerph19106173 (2022).
Tu, H. T., Spence, S., Kalman, J. M. & Davis, S. M. Twenty-eight day Holter monitoring is poorly tolerated and insensitive for paroxysmal atrial fibrillation detection in cryptogenic stroke. Intern. Med. J. 44, 505–508. https://doi.org/10.1111/imj.12410 (2014).
Krijthe, B. P. et al. Projections on the number of individuals with atrial fibrillation in the European Union, from 2000 to 2060. Eur. Heart J. 34, 2746–2751. https://doi.org/10.1093/eurheartj/eht280 (2013).
Colilla, S. et al. Estimates of current and future incidence and prevalence of atrial fibrillation in the U.S. Adult population. Am. J. Cardiol. 112, 1142–1147. https://doi.org/10.1016/j.amjcard.2013.05.063 (2013).
Lip, G. Y. H., Skjøth, F., Nielsen, P. B. & Larsen, T. B. 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. 125, 48–54. https://doi.org/10.1016/j.amjcard.2019.09.034 (2020).
Berge, T. et al. Systematic screening for atrial fibrillation in a 65-year-old population with risk factors for stroke: data from the Akershus cardiac examination 1950 study. EP Europace. 20, f299–f305. https://doi.org/10.1093/europace/eux293 (2018).
Moran, P.S., Teljeur, C., Ryan, M. & Smith, S. M. Systematic screening for the detection of atrial fibrillation. Cochrane Database Syst. Reviews. 2021 https://doi.org/10.1002/14651858.CD009586.pub3 (2016).
Huang, C. et al. A novel method for detection of the transition between atrial fibrillation and sinus rhythm. IEEE Trans. Biomed. Eng. 58, 1113–1119. https://doi.org/10.1109/TBME.2010.2096506 (2011).
Attia, Z. I. 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 394, 861–867. https://doi.org/10.1016/S0140-6736(19)31721-0 (2019).
Nandi, M., Venton, J. & Aston, P. J. A novel method to quantify arterial pulse waveform morphology: Attractor reconstruction for physiologists and clinicians. Physiol. Meas. 39, 104008. https://doi.org/10.1088/1361-6579/aae46a (2018).
Aston, P. J., Christie, M. I., Huang, Y. H. & Nandi, M. Beyond HRV: Attractor reconstruction using the entire cardiovascular waveform data for novel feature extraction. Physiol. Meas. 39, 024001. https://doi.org/10.1088/1361-6579/aaa93d (2018).
Lyle, J. V., Nandi, M. & Aston, P. J. Symmetric Projection Attractor Reconstruction: Sex differences in the ECG. Front. Cardiovasc. Med. 8, 709457. https://doi.org/10.3389/fcvm.2021.709457 (2021).
Bonet-Luz, E. et al. Symmetric Projection Attractor Reconstruction analysis of murine electrocardiograms: Retrospective prediction of Scn5a+/- genetic mutation attributable to Brugada syndrome. Heart Rhythm O2 1, 368–375. https://doi.org/10.1016/j.hroo.2020.08.007 (2020).
Huang, Y. H. et al. Detecting paroxysmal atrial fibrillation from normal sinus rhythm in equine athletes using Symmetric Projection Attractor Reconstruction and machine learning. Cardiovasc. Digit. Health J. 3, 96–106. https://doi.org/10.1016/j.cvdhj.2022.02.001 (2022).
Sasaki, N. Frequency analysis of atrial fibrillation from the specific ECG leads V7-V9: A lower DF in lead V9 is a marker of potential atrial remodelling. J. Cardiol. 66, 388–394. https://doi.org/10.1016/j.jjcc.2015.06.006 (2015).
Naderi, S. et al. The impact of age on the epidemiology of atrial fibrillation hospitalizations. Am. J. Med. 127 https://doi.org/10.1016/j.amjmed.2013.10.005 (2014). 158.e1-158.e7.
Wagner, P. et al. PTB-XL, a large publicly available electrocardiography dataset. Sci. Data. 7, 154. https://doi.org/10.1038/s41597-020-0495-6 (2020).
Wagner, P., Strodthoff, N., Bousseljot, R., Samek, W. & Schaeffter, T. PTB-XL, a large publicly available electrocardiography dataset (version 1.0.1). PhysioNet https://doi.org/10.13026/x4td-x982 (2020).
Goldberger, A. et al. Physionet, PhysioBank, PhysioToolkit, and PhysioNet (Components of a New Research Resource for Complex Physiologic Signals, 2000).
Ko, D. et al. Atrial fibrillation in women: epidemiology, pathophysiology, presentation, and prognosis. Nat. Rev. Cardiol. 13, 321–332. https://doi.org/10.1038/nrcardio.2016.45 (2016).
Piccini, J. P. et al. Incidence and prevalence of atrial fibrillation and associated mortality among medicare beneficiaries: 1993–2007. Circ. Cardiovasc. Qual. Outcomes. 5, 85–93. https://doi.org/10.1161/CIRCOUTCOMES.111.962688 (2012).
Takens, F. Detecting strange attractors in turbulence, in: (eds Rand, D. & Young, L. S.) Dynamical Systems and Turbulence, Warwick 1980. Lecture Notes in Mathematics, Vol 898., Springer, : 366–381. https://doi.org/10.1007/BFb0091924. (1981).
Nandi, M. & Aston, P. J. Extracting new information from old waveforms: Symmetric Projection Attractor Reconstruction: Where maths meets medicine. Exp. Physiol. 105, 1444–1451. https://doi.org/10.1113/EP087873 (2020).
Lyle, J. V. & Aston, P. J. Symmetric Projection Attractor Reconstruction: Embedding in higher dimensions. Chaos 31, 113135. https://doi.org/10.1063/5.0064450 (2021).
Aston, P. et al. Deep learning applied to attractor images derived from ECG signals for detection of genetic mutation. In Proceedings of International Conference in Computing in Cardiology, Singapore. https://doi.org/10.22489/CinC.2019.097 (2019).
Pizzuti, G. P., Cifaldi, S. & Nolfe, G. Digital sampling rate and ECG analysis. J. Biomed. Eng. 7, 247–250. https://doi.org/10.1016/0141-5425(85)90027-5 (1985).
Kwon, O. et al. Electrocardiogram sampling frequency range acceptable for heart rate variability analysis. Healthc. Inf. Res. 24, 198. https://doi.org/10.4258/hir.2018.24.3.198 (2018).
Acknowledgements
We thank Dr Jane Lyle for her help in developing the machine learning code at the beginning of the project.
Funding
This work was supported by a grant from the National Institute of Health Research Applied Research Collaboration Kent Surrey and Sussex.
Author information
Authors and Affiliations
Contributions
All authors attest they meet the current ICMJE criteria for authorship.
Corresponding author
Ethics declarations
Competing interests
Philip Aston and Manasi Nandi have a patent, WO2015121679A1, “Delay coordinate analysis of periodic data,” which covers the foundations of the Symmetric Projection Attractor Reconstruction (SPAR) method used in this paper. Steven Creasy, Gregory Lip, Gary Tse, and Kamalan Jeevaratnam have no competing interests to declare.
Ethics statement
The research in this paper was completed while adhering to the regulations of the University of Surrey review boards on human studies and animal care.
Patient consent statement
The data used in this paper was open source and was approved for publication by the Institutional Ethics Committee as anonymous data (PTB-2020-1).
Contribution to the field
Atrial fibrillation impacts a significant proportion of the adult population and is a serious risk factor for strokes. Detection is often difficult due to the brief nature of the episodes and as such long term monitoring has become the accepted method for diagnosis of atrial fibrillation. Studies have shown that long term monitoring is a fairly ineffective way to diagnose atrial fibrillation and so there is scope to improve the methodologies used. In this study we looked to apply a signal processing technique to electrocardiograms in conjunction with machine learning techniques to detect atrial fibrillation episodes after they had occurred. By investigating the impact of the sampling frequency of the electrocardiograms and optimising other parameters within the method we were able to not only detect atrial fibrillation after it had occurred but do so with a higher accuracy than current long term monitoring methods. This poses potential for a future change to the detection of atrial fibrillation for improved detection rate and also lower risk to the patients.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Creasy, S., Lip, G.Y.H., Tse, G. et al. Classification of paroxysmal atrial fibrillation using sinus rhythm electrocardiograms using the symmetric projection attractor reconstruction method. Sci Rep (2026). https://doi.org/10.1038/s41598-026-37491-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-37491-1
