{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T02:01:58Z","timestamp":1782439318985,"version":"3.54.5"},"reference-count":57,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T00:00:00Z","timestamp":1764892800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T00:00:00Z","timestamp":1764892800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100000274","name":"British Heart Foundation","doi-asserted-by":"publisher","award":["RG\/F\/22\/110078"],"award-info":[{"award-number":["RG\/F\/22\/110078"]}],"id":[{"id":"10.13039\/501100000274","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Artificial intelligence-enhanced electrocardiogram (AI-ECG) models have shown outstanding performance in diagnostic and prognostic tasks, yet their black-box nature hampers clinical adoption. Meanwhile, a growing demand for explainable AI in medicine underscores the need for transparent, trustworthy decision-making. Moving beyond post-hoc explainability techniques that have shown unreliable results, we focus on explicit representation learning using variational autoencoders (VAE) to capture inherently interpretable ECG features. While VAEs have demonstrated potential for ECG interpretability, the presumed performance-explainability trade-off remains underexplored, with many studies relying on complex, non-linear methods that obscure the morphological information of the features. In this work, we present a novel framework (VAE-SCAN) to model bi-directional, interpretable associations between ECG features and clinical factors. We also investigate how different representations affect ECG decoding performance across models with varying levels of explainability. Our findings demonstrate the cost introduced by intrinsic ECG interpretability, based on which we discuss potential implications and directions.<\/jats:p>","DOI":"10.1038\/s41746-025-02122-y","type":"journal-article","created":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T10:57:09Z","timestamp":1764932229000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["The cost of explainability in artificial intelligence-enhanced electrocardiogram models"],"prefix":"10.1038","volume":"8","author":[{"given":"Konstantinos","family":"Patlatzoglou","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Libor","family":"Pastika","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joseph","family":"Barker","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ewa","family":"Sieliwonczyk","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gul Rukh","family":"Khattak","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Boroumand","family":"Zeidaabadi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ant\u00f4nio H.","family":"Ribeiro","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James S.","family":"Ware","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicholas S.","family":"Peters","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Antonio Luiz P.","family":"Ribeiro","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel B.","family":"Kramer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonathan W.","family":"Waks","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arunashis","family":"Sau","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fu Siong","family":"Ng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,12,5]]},"reference":[{"key":"2122_CR1","doi-asserted-by":"publisher","first-page":"107187","DOI":"10.1016\/j.knosys.2021.107187","volume":"227","author":"X Liu","year":"2021","unstructured":"Liu, X., Wang, H., Li, Z. & Qin, L. Deep learning in ECG diagnosis: a review. Knowl. Based Syst. 227, 107187 (2021).","journal-title":"Knowl. Based Syst."},{"key":"2122_CR2","first-page":"1179","volume":"23","author":"S Somani","year":"2021","unstructured":"Somani, S. et al. Deep learning and the electrocardiogram: review of the current state-of-the-art. EP Eur. 23, 1179\u20131191 (2021).","journal-title":"EP Eur."},{"key":"2122_CR3","doi-asserted-by":"publisher","first-page":"e000193","DOI":"10.1136\/bmjmed-2022-000193","volume":"2","author":"A Sau","year":"2023","unstructured":"Sau, A. & Ng, F. S. \u00ad\u00ad The emerging role of artificial intelligence enabled electrocardiograms in healthcare. BMJ Med. 2, e000193 (2023).","journal-title":"BMJ Med."},{"key":"2122_CR4","doi-asserted-by":"publisher","first-page":"1461","DOI":"10.1038\/s41591-024-02961-4","volume":"30","author":"C-S Lin","year":"2024","unstructured":"Lin, C.-S. et al. AI-enabled electrocardiography alert intervention and all-cause mortality: a pragmatic randomized clinical trial. Nat. Med. 30, 1461\u20131470 (2024).","journal-title":"Nat. Med."},{"key":"2122_CR5","doi-asserted-by":"publisher","first-page":"168","DOI":"10.1109\/RBME.2018.2885714","volume":"12","author":"P Bizopoulos","year":"2019","unstructured":"Bizopoulos, P. & Koutsouris, D. Deep learning in cardiology. IEEE Rev. Biomed. Eng. 12, 168\u2013193 (2019).","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"2122_CR6","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1016\/j.hrthm.2020.08.021","volume":"18","author":"H Bleijendaal","year":"2021","unstructured":"Bleijendaal, H. et al. Computer versus cardiologist: is a machine learning algorithm able to outperform an expert in diagnosing a phospholamban p.Arg14del mutation on the electrocardiogram?. Heart Rhythm 18, 79\u201387 (2021).","journal-title":"Heart Rhythm"},{"key":"2122_CR7","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2023.101805","volume":"99","author":"S Ali","year":"2023","unstructured":"Ali, S. et al. Explainable artificial intelligence (XAI): what we know and what is left to attain Trustworthy Artificial Intelligence. Inf. Fusion 99, 101805 (2023).","journal-title":"Inf. Fusion"},{"key":"2122_CR8","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1038\/s41746-024-01130-8","volume":"7","author":"SV Kalmady","year":"2024","unstructured":"Kalmady, S. V. et al. Development and validation of machine learning algorithms based on electrocardiograms for cardiovascular diagnoses at the population level. Npj Digit. Med. 7, 133 (2024).","journal-title":"Npj Digit. Med."},{"key":"2122_CR9","doi-asserted-by":"publisher","first-page":"e791","DOI":"10.1016\/S2589-7500(24)00172-9","volume":"6","author":"A Sau","year":"2024","unstructured":"Sau, A. et al. Artificial intelligence-enabled electrocardiogram for mortality and cardiovascular risk estimation: a model development and validation study. Lancet Digit. Health. 6, e791\u2013e802 (2024).","journal-title":"Lancet Digit. Health."},{"key":"2122_CR10","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-023-01989-3","volume":"23","author":"G Baj","year":"2023","unstructured":"Baj, G. et al. Comparison of discrimination and calibration performance of ECG-based machine learning models for prediction of new-onset atrial fibrillation. BMC Med. Res. Methodol. 23, 169 (2023).","journal-title":"BMC Med. Res. Methodol."},{"key":"2122_CR11","doi-asserted-by":"publisher","first-page":"85","DOI":"10.3389\/fdgth.2021.671015","volume":"3","author":"U Mahmood","year":"2021","unstructured":"Mahmood, U. et al. Detecting spurious correlations with sanity tests for artificial intelligence guided radiology systems. Front. Digit. Health 3, 85 (2021).","journal-title":"Front. Digit. Health"},{"key":"2122_CR12","doi-asserted-by":"publisher","first-page":"360","DOI":"10.1038\/s41591-020-0791-x","volume":"26","author":"X Han","year":"2020","unstructured":"Han, X. et al. Deep learning models for electrocardiograms are susceptible to adversarial attack. Nat. Med. 26, 360\u2013363 (2020).","journal-title":"Nat. Med."},{"key":"2122_CR13","doi-asserted-by":"publisher","first-page":"e745","DOI":"10.1016\/S2589-7500(21)00208-9","volume":"3","author":"M Ghassemi","year":"2021","unstructured":"Ghassemi, M., Oakden-Rayner, L. & Beam, A. L. The false hope of current approaches to explainable artificial intelligence in health care. Lancet Digit. Health 3, e745\u2013e750 (2021).","journal-title":"Lancet Digit. Health"},{"key":"2122_CR14","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1016\/j.inffus.2019.12.012","volume":"58","author":"A Barredo Arrieta","year":"2020","unstructured":"Barredo Arrieta, A. et al. Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion 58, 82\u2013115 (2020).","journal-title":"Inf. Fusion"},{"key":"2122_CR15","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.dsp.2017.10.011","volume":"73","author":"G Montavon","year":"2018","unstructured":"Montavon, G., Samek, W. & M\u00fcller, K.-R. Methods for interpreting and understanding deep neural networks. Digit. Signal Process. 73, 1\u201315 (2018).","journal-title":"Digit. Signal Process."},{"key":"2122_CR16","unstructured":"Doshi-Velez, F. & Kim, B. Towards a rigorous science of interpretable machine learning. Preprint at https:\/\/arxiv.org\/abs\/1702.08608 (2017)."},{"key":"2122_CR17","unstructured":"Higgins, I. et al. \u03b2-VAE: learning basic visual concepts with a constrained variational framework. (2017)."},{"key":"2122_CR18","unstructured":"Higgins, I. et al. SCAN: Learning Hierarchical Compositional Visual Concepts. Preprint at http:\/\/arxiv.org\/abs\/1707.03389 (2018)."},{"key":"2122_CR19","doi-asserted-by":"publisher","first-page":"e0260612","DOI":"10.1371\/journal.pone.0260612","volume":"16","author":"J-H Jang","year":"2021","unstructured":"Jang, J.-H., Kim, T. Y., Lim, H.-S. & Yoon, D. Unsupervised feature learning for electrocardiogram data using the convolutional variational autoencoder. PLoS ONE 16, e0260612 (2021).","journal-title":"PLoS ONE"},{"key":"2122_CR20","unstructured":"Oren, Y. & Gilbai, K. This ECG Does Not Exist: Using Variational Autoencoders to Generate Periodic ECG Signals. Stanford University CS230 Project Report (2022). Available at: http:\/\/cs230.stanford.edu\/projects_fall_2022\/reports\/133.pdf."},{"key":"2122_CR21","doi-asserted-by":"crossref","unstructured":"Dakshit, S. & Prabhakaran, B. CVAE-based generator for variable length synthetic ECG. In 2023 IEEE 11th International Conference on Healthcare Informatics (ICHI) 235\u2013244 (IEEE, Houston, TX, USA, 2023).","DOI":"10.1109\/ICHI57859.2023.00040"},{"key":"2122_CR22","doi-asserted-by":"publisher","first-page":"390","DOI":"10.1093\/ehjdh\/ztac038","volume":"3","author":"RR Van De Leur","year":"2022","unstructured":"Van De Leur, R. R. et al. Improving explainability of deep neural network-based electrocardiogram interpretation using variational auto-encoders. Eur. Heart J. Digit. Health 3, 390\u2013404 (2022).","journal-title":"Eur. Heart J. Digit. Health"},{"key":"2122_CR23","doi-asserted-by":"publisher","first-page":"638191","DOI":"10.3389\/fgene.2021.638191","volume":"12","author":"VV Kuznetsov","year":"2021","unstructured":"Kuznetsov, V. V., Moskalenko, V. A., Gribanov, D. V. & Zolotykh, N. Yu. Interpretable feature generation in ECG using a variational autoencoder. Front. Genet. 12, 638191 (2021).","journal-title":"Front. Genet."},{"key":"2122_CR24","doi-asserted-by":"publisher","unstructured":"Van Steenkiste, T., Deschrijver, D. & Dhaene, T. Generating an Explainable ECG Beat Space With Variational Auto-Encoders. Preprint at https:\/\/doi.org\/10.48550\/arXiv.1911.04898 (2019).","DOI":"10.48550\/arXiv.1911.04898"},{"key":"2122_CR25","doi-asserted-by":"publisher","unstructured":"Van Steenkiste, T., Deschrijver, D. & Dhaene, T. Interpretable ECG beat embedding using disentangled variational auto-encoders. In 2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS) 373\u2013378 https:\/\/doi.org\/10.1109\/CBMS.2019.00081 (2019).","DOI":"10.1109\/CBMS.2019.00081"},{"key":"2122_CR26","doi-asserted-by":"publisher","first-page":"104587","DOI":"10.1016\/j.bspc.2023.104587","volume":"82","author":"Y Xia","year":"2023","unstructured":"Xia, Y., Wang, W. & Wang, K. ECG signal generation based on conditional generative models. Biomed. Signal Process. Control 82, 104587 (2023).","journal-title":"Biomed. Signal Process. Control"},{"key":"2122_CR27","unstructured":"Gyawali, P. K., Horacek, B. M., Sapp, J. L. & Wang, L. Learning disentangled representation from 12-lead electrograms: application in localizing the origin of ventricular tachycardia. Preprint at https:\/\/arxiv.org\/abs\/1808.01524 (2018)."},{"key":"2122_CR28","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-024-01170-0","volume":"7","author":"L Pastika","year":"2024","unstructured":"Pastika, L. et al. Artificial intelligence-enhanced electrocardiography derived body mass index as a predictor of future cardiometabolic disease. Npj Digit. Med. 7, 1\u201316 (2024).","journal-title":"Npj Digit. Med."},{"key":"2122_CR29","doi-asserted-by":"crossref","unstructured":"Crook, B., Schl\u00fcter, M. & Speith, T. Revisiting the performance-explainability trade-off in explainable artificial intelligence (XAI). In 2023 IEEE 31st International Requirements Engineering Conference Workshops (REW) 316\u2013324 (IEEE, Hannover, Germany, 2023).","DOI":"10.1109\/REW57809.2023.00060"},{"key":"2122_CR30","unstructured":"Burgess, C. P. et al. Understanding disentangling in \u03b2-VAE. Preprint at http:\/\/arxiv.org\/abs\/1804.03599 (2018)."},{"key":"2122_CR31","unstructured":"Bonheme, L. & Grzes, M. How do variational autoencoders learn? Insights from representational similarity. Preprint at http:\/\/arxiv.org\/abs\/2205.08399 (2022)."},{"key":"2122_CR32","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCEP.119.007284","volume":"12","author":"ZI Attia","year":"2019","unstructured":"Attia, Z. I. et al. Age and sex estimation using artificial intelligence from standard 12-lead ECGs. Circ. Arrhythm. Electrophysiol. 12, e007284 (2019).","journal-title":"Circ. Arrhythm. Electrophysiol."},{"key":"2122_CR33","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-25351-7","volume":"12","author":"EM Lima","year":"2021","unstructured":"Lima, E. M. et al. Deep neural network-estimated electrocardiographic age as a mortality predictor. Nat. Commun. 12, 5117 (2021).","journal-title":"Nat. Commun."},{"key":"2122_CR34","doi-asserted-by":"publisher","first-page":"886","DOI":"10.1038\/s41591-020-0870-z","volume":"26","author":"S Raghunath","year":"2020","unstructured":"Raghunath, S. et al. Prediction of mortality from 12-lead electrocardiogram voltage data using a deep neural network. Nat. Med. 26, 886\u2013891 (2020).","journal-title":"Nat. Med."},{"key":"2122_CR35","doi-asserted-by":"publisher","first-page":"240","DOI":"10.1016\/0735-6757(90)90331-S","volume":"8","author":"J Jones","year":"1990","unstructured":"Jones, J., Srodulski, Z. M. & Romisher, S. The aging electrocardiogram. Am. J. Emerg. Med. 8, 240\u2013245 (1990).","journal-title":"Am. J. Emerg. Med."},{"key":"2122_CR36","doi-asserted-by":"publisher","DOI":"10.1186\/s40001-022-00880-z","volume":"27","author":"C Prajapati","year":"2022","unstructured":"Prajapati, C., Koivum\u00e4ki, J., Pekkanen-Mattila, M. & Aalto-Set\u00e4l\u00e4, K. Sex differences in heart: from basics to clinics. Eur. J. Med. Res. 27, 241 (2022).","journal-title":"Eur. J. Med. Res."},{"key":"2122_CR37","doi-asserted-by":"publisher","first-page":"582","DOI":"10.1016\/j.amjmed.2019.08.056","volume":"133","author":"C Polcwiartek","year":"2020","unstructured":"Polcwiartek, C. et al. Long-term prognostic value of less-stringent electrocardiographic Q waves and fourth universal definition of myocardial infarction Q waves. Am. J. Med. 133, 582\u2013589.e7 (2020).","journal-title":"Am. J. Med."},{"key":"2122_CR38","doi-asserted-by":"publisher","first-page":"521","DOI":"10.1161\/CIRCOUTCOMES.110.959023","volume":"4","author":"EZ Gorodeski","year":"2011","unstructured":"Gorodeski, E. Z. et al. Use of hundreds of electrocardiographic biomarkers for prediction of mortality in postmenopausal women. Circ. Cardiovasc. Qual. Outcomes 4, 521\u2013532 (2011).","journal-title":"Circ. Cardiovasc. Qual. Outcomes"},{"key":"2122_CR39","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-020-15432-4","volume":"11","author":"AH Ribeiro","year":"2020","unstructured":"Ribeiro, A. H. et al. Automatic diagnosis of the 12-lead ECG using a deep neural network. Nat. Commun. 11, 1760 (2020).","journal-title":"Nat. Commun."},{"key":"2122_CR40","doi-asserted-by":"publisher","first-page":"422","DOI":"10.1161\/CIRCULATIONAHA.114.008727","volume":"133","author":"ALP Ribeiro","year":"2016","unstructured":"Ribeiro, A. L. P. et al. Cardiovascular health in Brazil: trends and perspectives. Circulation 133, 422\u2013433 (2016).","journal-title":"Circulation"},{"key":"2122_CR41","doi-asserted-by":"publisher","unstructured":"Delaney, A. M., Brophy, E. & Ward, T. E. Synthesis of realistic ECG using generative adversarial networks. Preprint at https:\/\/doi.org\/10.48550\/arXiv.1909.09150 (2019).","DOI":"10.48550\/arXiv.1909.09150"},{"key":"2122_CR42","unstructured":"Locatello, F. et al. Weakly-Supervised Disentanglement Without Compromises. In Proc. 37th International Conference on Machine Learning, 6348\u20136359 (PMLR, 2020)."},{"key":"2122_CR43","doi-asserted-by":"publisher","first-page":"107798","DOI":"10.1016\/j.cmpb.2023.107798","volume":"242","author":"HF Stabenau","year":"2023","unstructured":"Stabenau, H. F. & Waks, J. W. BRAVEHEART: open-source software for automated electrocardiographic and vectorcardiographic analysis. Comput. Methods Prog. Biomed. 242, 107798 (2023).","journal-title":"Comput. Methods Prog. Biomed."},{"key":"2122_CR44","unstructured":"Alvarez-Melis, D. & Jaakkola, T. S. Towards robust interpretability with self-explaining neural networks. In Proc. 32nd International Conference on Neural Information Processing Systems, 7786\u20137795 (Curran Associates Inc., Red Hook, NY, USA, 2018)."},{"key":"2122_CR45","unstructured":"Koh, P. W. et al. Concept Bottleneck Models. In Proc. 37th International Conference on Machine Learning, 5338\u20135348 (PMLR, 2020)."},{"key":"2122_CR46","doi-asserted-by":"publisher","first-page":"0221","DOI":"10.34133\/hds.0221","volume":"5","author":"K Yang","year":"2025","unstructured":"Yang, K. et al. ECG-LM: understanding electrocardiogram with a large language model. Health Data Sci. 5, 0221 (2025).","journal-title":"Health Data Sci."},{"key":"2122_CR47","doi-asserted-by":"publisher","unstructured":"Lan, X. et al. GEM: empowering MLLM for grounded ECG understanding with time series and images. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2503.06073 (2025).","DOI":"10.48550\/arXiv.2503.06073"},{"key":"2122_CR48","unstructured":"Yu, H., Guo, P. & Sano, A. ECG Semantic Integrator (ESI): A Foundation ECG Model Pretrained with LLM-Enhanced Cardiological Text. Trans. Mach. Learn. Res. https:\/\/openreview.net\/forum?id=giEbq8Khcf (2024)."},{"key":"2122_CR49","doi-asserted-by":"publisher","first-page":"988","DOI":"10.1109\/78.492552","volume":"44","author":"F Gustafsson","year":"1996","unstructured":"Gustafsson, F. Determining the initial states in forward-backward filtering. IEEE Trans. Signal Process. 44, 988\u2013992 (1996).","journal-title":"IEEE Trans. Signal Process."},{"key":"2122_CR50","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1126\/science.1127647","volume":"313","author":"GE Hinton","year":"2006","unstructured":"Hinton, G. E. & Salakhutdinov, R. R. Reducing the dimensionality of data with neural networks. Science 313, 504\u2013507 (2006).","journal-title":"Science"},{"key":"2122_CR51","doi-asserted-by":"publisher","first-page":"208","DOI":"10.1109\/RBME.2022.3154893","volume":"16","author":"K Nezamabadi","year":"2023","unstructured":"Nezamabadi, K., Sardaripour, N., Haghi, B. & Forouzanfar, M. Unsupervised ECG analysis: a review. IEEE Rev. Biomed. Eng. 16, 208\u2013224 (2023).","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"2122_CR52","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1038\/s41591-018-0240-2","volume":"25","author":"ZI Attia","year":"2019","unstructured":"Attia, Z. I. et al. Screening for cardiac contractile dysfunction using an artificial intelligence\u2013enabled electrocardiogram. Nat. Med. 25, 70\u201374 (2019).","journal-title":"Nat. Med."},{"key":"2122_CR53","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1093\/ehjdh\/ztac042","volume":"3","author":"A Sau","year":"2022","unstructured":"Sau, A. et al. Artificial intelligence-enabled electrocardiogram to distinguish cavotricuspid isthmus dependence from other atrial tachycardia mechanisms. Eur. Heart J. - Digit. Health 3, 405\u2013414 (2022).","journal-title":"Eur. Heart J. - Digit. Health"},{"key":"2122_CR54","doi-asserted-by":"publisher","first-page":"1798","DOI":"10.1109\/TPAMI.2013.50","volume":"35","author":"Y Bengio","year":"2013","unstructured":"Bengio, Y., Courville, A. & Vincent, P. Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35, 1798\u20131828 (2013).","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"2122_CR55","unstructured":"Locatello, F. et al. Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations. In Proc. 36th International Conference on Machine Learning, 4114\u20134124 (PMLR, 2019)."},{"key":"2122_CR56","unstructured":"Honke, G. et al. Representation learning for improved interpretability and classification accuracy of clinical factors from EEG. Preprint at http:\/\/arxiv.org\/abs\/2010.15274 (2020)."},{"key":"2122_CR57","doi-asserted-by":"crossref","unstructured":"McDermott, M. B., Zhang, H., Hansen, L. H., Angelotti, G. & Gallifant, J. A closer look at AUROC and AUPRC under class imbalance. In Proc. 38th International Conference on Neural Information Processing Systems, Vol. 37 44102\u201344163 (Curran Associates Inc., Red Hook, NY, USA, 2024).","DOI":"10.52202\/079017-1400"}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02122-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02122-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02122-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T10:57:21Z","timestamp":1764932241000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02122-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,5]]},"references-count":57,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["2122"],"URL":"https:\/\/doi.org\/10.1038\/s41746-025-02122-y","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,5]]},"assertion":[{"value":"10 March 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 October 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"5 December 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing non-financial interests, but the following competing financial interests: J.W.W. was previously on the advisory board for Heartcor solutions LLC and previously received research support from Anumana, J.S.W. has consulted for MyoKardia, Inc., Pfizer, Foresite Labs, and Health Lumen, and receives research support from Bristol Myers-Squibb. F.S.N., A.S., L.P., and B.Z. are inventors on a patent application on a platform for AI-ECG analysis. The remaining authors declare no Competing Financial Interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"747"}}