{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T13:55:38Z","timestamp":1790862938633,"version":"4.1.0"},"reference-count":28,"publisher":"AIP Publishing","issue":"10","funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["DMS-2235451"],"award-info":[{"award-number":["DMS-2235451"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000893","name":"Simons Foundation","doi-asserted-by":"publisher","award":["MPS-NITMB-00005320"],"award-info":[{"award-number":["MPS-NITMB-00005320"]}],"id":[{"id":"10.13039\/100000893","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,10,1]]},"abstract":"<jats:p>Electrocardiogram (ECG) signals exhibit complex, nonlinear, and nonstationary dynamics that are not fully captured by conventional time series analysis methods. In this study, we propose a novel framework for atrial fibrillation (AF) detection that integrates a beat-level recurrence representation with supervised metric learning. ECG recordings are first segmented into beat-centered waveforms that serve as effective dynamical states, avoiding the inaccurate estimation of time delay and embedding dimension from short recordings. Rather than constructing conventional recurrence plots from delay-embedded trajectories, the proposed method forms a beat-to-beat dissimilarity matrix using the Jensen\u2013Shannon distance between normalized ECG beat segments. This representation captures variations in beat morphology and rhythm irregularity and provides information complementary to heart rate variability descriptors and recurrence quantification features. A contrastive Gaussian-mixture variational autoencoder learns an eight-component latent distribution with two components assigned to each of the four rhythm classes. Reconstruction, Gaussian-mixture regularization, supervised classification, clustering, and contrastive objectives jointly promote informative, compact, and class-discriminative latent representations. Class-conditional samples drawn from the learned mixture provide balanced synthetic latent representations for classifier training. Evaluated on the four-class PhysioNet\/CinC 2017 dataset, the proposed framework achieves an overall accuracy of approximately 94%, with F1 scores of 0.97, 0.95, 0.92, and 0.86 for normal sinus rhythm, AF, other rhythm, and noise, respectively. These results demonstrate the potential of combining beat-level nonlinear dynamics, structured generative modeling, and metric learning for AF detection from short ECG recordings.<\/jats:p>","DOI":"10.1063\/5.0337012","type":"journal-article","created":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T12:56:14Z","timestamp":1790859374000},"update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":0,"title":["Generative recurrence analysis and metric learning for detection of atrial fibrillation from short ECG data"],"prefix":"10.1063","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0482-2530","authenticated-orcid":false,"given":"Chunyu","family":"Wu","sequence":"first","affiliation":[{"name":"School of Systems Science and Industrial Engineering, State University of New York 1 , Binghamton, New York 13902,","place":["USA"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-6188-1012","authenticated-orcid":false,"given":"Suwetha Jahan","family":"Maheswary","sequence":"additional","affiliation":[{"name":"School of Systems Science and Industrial Engineering, State University of New York 1 , Binghamton, New York 13902,","place":["USA"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3386-3446","authenticated-orcid":false,"given":"Rakesh","family":"Srivastava","sequence":"additional","affiliation":[{"name":"School of Computing Science Engineering and Artificial Intelligence, VIT Bhopal University 2 , Sehore 466114, Madhya Pradesh,","place":["India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3262-5067","authenticated-orcid":false,"given":"Pon","family":"Harshavardhanan","sequence":"additional","affiliation":[{"name":"School of Computing Science Engineering and Artificial Intelligence, VIT Bhopal University 2 , Sehore 466114, Madhya Pradesh,","place":["India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1352-0385","authenticated-orcid":false,"given":"Swagat","family":"Samantaray","sequence":"additional","affiliation":[{"name":"Deevia Software (India) Pvt. Ltd 3 , Bengaluru 560102, Karnataka,","place":["India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9667-0313","authenticated-orcid":false,"given":"Zimo","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Systems Science and Industrial Engineering, State University of New York 1 , Binghamton, New York 13902,","place":["USA"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7810-7279","authenticated-orcid":false,"given":"Changqing","family":"Cheng","sequence":"additional","affiliation":[{"name":"School of Systems Science and Industrial Engineering, State University of New York 1 , Binghamton, New York 13902,","place":["USA"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"317","published-online":{"date-parts":[[2026,10,1]]},"reference":[{"key":"2026100108520200500_c1","doi-asserted-by":"publisher","first-page":"16016","DOI":"10.1038\/nrdp.2016.16","article-title":"Atrial fibrillation","volume":"2","year":"2016","journal-title":"Nat. Rev. Dis. Primers"},{"key":"2026100108520200500_c2","doi-asserted-by":"publisher","first-page":"113560","DOI":"10.1016\/j.eswa.2020.113560","article-title":"Pattern recognition and automatic identification of early-stage atrial fibrillation","volume":"158","year":"2020","journal-title":"Expert Syst. Appl."},{"key":"2026100108520200500_c3","doi-asserted-by":"publisher","first-page":"093104","DOI":"10.1063\/5.0006537","article-title":"Intrinsic recurrence quantification analysis of nonlinear and nonstationary short-term time series","volume":"30","year":"2020","journal-title":"Chaos"},{"key":"2026100108520200500_c4","doi-asserted-by":"publisher","first-page":"606","DOI":"10.1136\/hrt.2006.107573","article-title":"A national survey of the prevalence, incidence, primary care burden and treatment of atrial fibrillation in Scotland","volume":"93","year":"2007","journal-title":"Heart"},{"key":"2026100108520200500_c5","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1161\/CIRCRESAHA.120.316340","article-title":"Epidemiology of atrial fibrillation in the 21st century","volume":"127","year":"2020","journal-title":"Circ. Res."},{"key":"2026100108520200500_c6","doi-asserted-by":"publisher","first-page":"68","DOI":"10.1109\/TBME.2023.3292383","article-title":"Image-decomposition-enhanced deep learning for detection of rotor cores in cardiac fibrillation","volume":"71","year":"2024","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"2026100108520200500_c7","doi-asserted-by":"publisher","first-page":"1053","DOI":"10.1080\/0740817X.2014.999180","article-title":"Time series forecasting for nonlinear and non-stationary processes: A review and comparative study","volume":"47","year":"2015","journal-title":"IIE Trans."},{"key":"2026100108520200500_c8","doi-asserted-by":"publisher","first-page":"1246746","DOI":"10.3389\/fphys.2023.1246746","article-title":"Deep learning for ECG arrhythmia detection and classification: An overview of progress for period 2017\u20132023","volume":"14","year":"2023","journal-title":"Front. Physiol."},{"key":"2026100108520200500_c9","doi-asserted-by":"publisher","first-page":"056206","DOI":"10.1103\/PhysRevE.82.056206","article-title":"Forecasting the evolution of nonlinear and nonstationary systems using recurrence-based local Gaussian process models","volume":"82","year":"2010","journal-title":"Phys. Rev. E"},{"key":"2026100108520200500_c10","first-page":"1383","article-title":"Gaussian mixture variational autoencoder with contrastive learning for multi-label classification","year":"2022"},{"key":"2026100108520200500_c11","doi-asserted-by":"publisher","first-page":"2466","DOI":"10.1073\/pnas.012579499","article-title":"Fractal dynamics in physiology: Alterations with disease and aging","volume":"99","year":"2002","journal-title":"Proc. Natl. Acad. Sci."},{"key":"2026100108520200500_c12","doi-asserted-by":"publisher","first-page":"278","DOI":"10.1016\/j.compbiomed.2018.06.002","article-title":"Automated diagnosis of arrhythmia using combination of CNN and LSTM techniques with variable length heart beats","volume":"102","year":"2018","journal-title":"Comput. Biol. Med."},{"key":"2026100108520200500_c13","doi-asserted-by":"publisher","first-page":"1147","DOI":"10.1088\/0967-3334\/32\/8\/010","article-title":"Prediction of paroxysmal atrial fibrillation using recurrence plot-based features of the RR-interval signal","volume":"32","year":"2011","journal-title":"Physiol. Meas."},{"key":"2026100108520200500_c14","doi-asserted-by":"publisher","first-page":"1105","DOI":"10.1016\/j.medengphy.2008.01.008","article-title":"Predicting termination of atrial fibrillation based on the structure and quantification of the recurrence plot","volume":"30","year":"2008","journal-title":"Med. Eng. Phys."},{"key":"2026100108520200500_c15","doi-asserted-by":"publisher","first-page":"965","DOI":"10.1152\/jappl.1994.76.2.965","article-title":"Dynamical assessment of physiological systems and states using recurrence plot strategies","volume":"76","year":"1994","journal-title":"J. Appl. Physiol."},{"key":"2026100108520200500_c16","doi-asserted-by":"publisher","first-page":"549","DOI":"10.1109\/TBME.2003.810705","article-title":"Epileptic seizure prediction and control","volume":"50","year":"2003","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"2026100108520200500_c17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1155\/2018\/6920420","article-title":"A review on the nonlinear dynamical system analysis of electrocardiogram signal","volume":"2018","year":"2018","journal-title":"J. Healthc. Eng."},{"key":"2026100108520200500_c18","first-page":"1","article-title":"AF classification from a short single lead ECG recording: The PhysioNet\/computing in cardiology challenge 2017","volume-title":"2017 Computing in Cardiology (CinC)","year":"2017"},{"key":"2026100108520200500_c19","doi-asserted-by":"publisher","first-page":"101712","DOI":"10.1016\/j.softx.2024.101712","article-title":"BioSPPy: A Python toolbox for physiological signal processing","volume":"26","year":"2024","journal-title":"SoftwareX"},{"key":"2026100108520200500_c20","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1109\/TBME.1985.325532","article-title":"A real-time QRS detection algorithm","volume":"BME-32","year":"1985","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"2026100108520200500_c21","article-title":"Temperature schedules for self-supervised contrastive methods on long-tail data","year":"2022"},{"key":"2026100108520200500_c22","first-page":"1","article-title":"Comparing feature-based classifiers and convolutional neural networks to detect arrhythmia from short segments of ECG","volume-title":"2017 Computing in Cardiology (CinC)","year":"2017"},{"key":"2026100108520200500_c23","first-page":"1","article-title":"Atrial fibrillation detection and ECG classification based on convolutional recurrent neural network","volume-title":"2017 Computing in Cardiology (CinC)","year":"2017"},{"key":"2026100108520200500_c24","doi-asserted-by":"publisher","first-page":"2461","DOI":"10.1109\/JBHI.2020.2981526","article-title":"Deep multi-scale fusion neural network for multi-class arrhythmia detection","volume":"24","year":"2020","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"2026100108520200500_c25","doi-asserted-by":"publisher","first-page":"102672","DOI":"10.1016\/j.bspc.2021.102672","article-title":"Stacking segment-based CNN with SVM for recognition of atrial fibrillation from single-lead ECG recordings","volume":"68","year":"2021","journal-title":"Biomed. Signal Process. Control"},{"key":"2026100108520200500_c26","doi-asserted-by":"publisher","first-page":"1362185","DOI":"10.3389\/fphys.2024.1362185","article-title":"Automated identification of atrial fibrillation from single-lead ECGs using multi-branching ResNet","volume":"15","year":"2024","journal-title":"Front. Physiol."},{"key":"2026100108520200500_c27","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.compbiomed.2016.05.006","article-title":"Heterogeneous recurrence analysis of heartbeat dynamics for the identification of sleep apnea events","volume":"75","year":"2016","journal-title":"Comput. Biol. Med."},{"key":"2026100108520200500_c28","doi-asserted-by":"publisher","first-page":"2700109","DOI":"10.1109\/JTEHM.2013.2273354","article-title":"Wireless wearable multisensory suite and real-time prediction of obstructive sleep apnea episodes","volume":"1","year":"2013","journal-title":"IEEE J. Transl. Eng. Health Med."}],"container-title":["Chaos: An Interdisciplinary Journal of Nonlinear Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/aip\/cha\/article-pdf\/doi\/10.1063\/5.0337012\/21159213\/103106_1_5.0337012.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/aip\/cha\/article-pdf\/doi\/10.1063\/5.0337012\/21159213\/103106_1_5.0337012.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T12:56:16Z","timestamp":1790859376000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/cha\/article\/36\/10\/103106\/3405793\/Generative-recurrence-analysis-and-metric-learning"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,1]]},"references-count":28,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2026,10,1]]}},"URL":"https:\/\/doi.org\/10.1063\/5.0337012","relation":{},"ISSN":["1054-1500","1089-7682"],"issn-type":[{"value":"1054-1500","type":"print"},{"value":"1089-7682","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,10]]},"published":{"date-parts":[[2026,10,1]]},"article-number":"103106"}}