{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,24]],"date-time":"2026-10-24T10:00:17Z","timestamp":1792836017408,"version":"4.5.0"},"reference-count":46,"publisher":"BMJ","issue":"1","license":[{"start":{"date-parts":[[2025,3,31]],"date-time":"2025-03-31T00:00:00Z","timestamp":1743379200000},"content-version":"unspecified","delay-in-days":89,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004063","name":"Knut and Alice Wallenberg foundation","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100004063","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100005760","name":"Gothenburg University","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005760","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["Open Heart"],"accepted":{"date-parts":[[2025,3,15]]},"published-print":{"date-parts":[[2025,1]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Atrial fibrillation (AF) is a common but often undiagnosed condition, increasing the risk of stroke and heart failure. Early detection is crucial, yet traditional methods struggle with AF\u2019s transient nature. This study investigates how augmenting ECG data with heart rate variability (HRV) and demographic data (age and sex) can improve AF detection.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We analysed 35 634 12-lead ECG recordings from three public databases (China Physiological Signal Challenge-Extra, PTB-XL and Georgia), each with physician-validated AF labels. A range of convolutional neural network models, including AlexNet, VGG-16, ResNet and transformers, were tested for AF prediction, enriched with HRV and demographic data to explore the effectiveness of the multimodal approach. Each data modality (ECG, HRV and demographic) was assessed for its contribution to model performance using fivefold cross-validation. Performance improvements were evaluated across key metrics, and saliency maps were generated to provide further insights into model behaviour and identify critical features in AF detection.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Integrating HRV and demographic data with ECG substantially improved performance. AlexNet and VGG-16 outperformed more complex models, achieving AUROC of 0.9617 (95% CI 0.95 to 0.97) and 0.9668 (95% CI 0.96 to 0.97), respectively. Adding HRV data showed the most significant improvement in sensitivity, with AlexNet increasing from 0.9117 to 0.9225 and VGG-16 from 0.9216 to 0.9225. Combining both HRV and demographic data led to further improvements, with AlexNet achieving a sensitivity of 0.9225 (up from 0.9192 with HRV) and VGG-16 reaching 0.9113 (up from 0.9097 with HRV). The combination of HRV and demographic data resulted in the highest gains in sensitivity and area under the receiver operating characteristic curve. Saliency maps confirmed the models identified key AF features, such as the absence of the P-wave, validating the multimodal approach.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>AlexNet and VGG-16 excelled in AF detection, with HRV data improving sensitivity, and demographic data providing additional benefits. These results highlight the potential of multimodal approaches, pending further clinical validation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/openhrt-2025-003185","type":"journal-article","created":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T02:32:35Z","timestamp":1743474755000},"page":"e003185","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":15,"title":["Integrating deep learning with ECG, heart rate variability and demographic data for improved detection of atrial fibrillation"],"prefix":"10.1136","volume":"12","author":[{"given":"Araz","family":"Rawshani","sequence":"first","affiliation":[{"name":"Departement of Clinical & Molecular Medicine, Institute of Medicine, Gothenburg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aidin","family":"Rawshani","sequence":"additional","affiliation":[{"name":"University of Gothenburg Institute of Medicine, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gustav","family":"Smith","sequence":"additional","affiliation":[{"name":"University of Gothenburg Institute of Medicine, Goteborg, Sweden"},{"name":"Department of Cardiology, Sahlgrenska University Hospital, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jan","family":"Boren","sequence":"additional","affiliation":[{"name":"University of Gothenburg Institute of Medicine, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Deepak L","family":"Bhatt","sequence":"additional","affiliation":[{"name":"Icahn School of Medicine at Mount Sinai, New York, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mats","family":"B\u00f6rjesson","sequence":"additional","affiliation":[{"name":"University of Gothenburg Institute of Medicine, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1677-7215","authenticated-orcid":false,"given":"Johan","family":"Engdahl","sequence":"additional","affiliation":[{"name":"Department of Cardiology, Sahlgrenska University Hospital, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peter","family":"Kelly","sequence":"additional","affiliation":[{"name":"Department of Cardiology, Sahlgrenska University Hospital, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Antros","family":"Louca","sequence":"additional","affiliation":[{"name":"University of Gothenburg Institute of Medicine, Goteborg, Sweden"},{"name":"Department of Molecular and Clinical Medicine, Gothenburg University, Gothenburg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Truls","family":"Ramunddal","sequence":"additional","affiliation":[{"name":"Department of Cardiology, Sahlgrenska University Hospital, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-0567-2020","authenticated-orcid":false,"given":"Erik","family":"Andersson","sequence":"additional","affiliation":[{"name":"Department of Clinical and Molecular Medicine, University of Gothenburg Institute of Medicine, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3875-8621","authenticated-orcid":false,"given":"Elmir","family":"Omerovic","sequence":"additional","affiliation":[{"name":"Department of Cardiology, Sahlgrenska University Hospital, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zacharias","family":"Mandalenakis","sequence":"additional","affiliation":[{"name":"Department of Cardiology, Sahlgrenska University Hospital, Goteborg, Sweden"},{"name":"Department of Molecular and Clinical Medicine, Gothenburg University, Gothenburg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3785-8380","authenticated-orcid":false,"given":"Vibha","family":"Gupta","sequence":"additional","affiliation":[{"name":"University of Gothenburg Institute of Medicine, Goteborg, Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2025,3,31]]},"reference":[{"key":"2025081710122812000_12.1.e003185.1","doi-asserted-by":"publisher","DOI":"10.1161\/CIR.0000000000000568"},{"key":"2025081710122812000_12.1.e003185.2","doi-asserted-by":"publisher","DOI":"10.1161\/CIR.0000000000001193"},{"key":"2025081710122812000_12.1.e003185.3","doi-asserted-by":"crossref","first-page":"299","DOI":"10.1093\/ehjqcco\/qcw031","article-title":"Characteristics and outcomes of atrial fibrillation patients with or without specific symptoms: results from the PREFER in AF registry","volume":"2","author":"Bakhai","year":"2016","journal-title":"Eur Heart J Qual Care Clin Outcomes"},{"key":"2025081710122812000_12.1.e003185.4","doi-asserted-by":"publisher","DOI":"10.1016\/j.amjmed.2014.11.026"},{"key":"2025081710122812000_12.1.e003185.5","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCRESAHA.120.316340"},{"key":"2025081710122812000_12.1.e003185.6","doi-asserted-by":"crossref","first-page":"2478","DOI":"10.1038\/s41591-022-02079-5","article-title":"AI detection of cardiac dysfunction from consumer watch ECG recordings","volume":"28","author":"Attia","year":"2022","journal-title":"Nat Med"},{"key":"2025081710122812000_12.1.e003185.7","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa1901183"},{"key":"2025081710122812000_12.1.e003185.8","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/0022-0736(92)90123-H","article-title":"Diagnosis of atrial fibrillation from surface electrocardiograms based on computer-detected atrial activity","volume":"25","author":"Slocum","year":"1992","journal-title":"J Electrocardiol"},{"key":"2025081710122812000_12.1.e003185.9","doi-asserted-by":"crossref","first-page":"274","DOI":"10.1016\/j.bspc.2015.01.007","article-title":"Rate-independent detection of atrial fibrillation by statistical modeling of atrial activity","volume":"18","author":"Ladavich","year":"2015","journal-title":"Biomed Signal Process Control"},{"key":"2025081710122812000_12.1.e003185.10","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2010.2096506"},{"key":"2025081710122812000_12.1.e003185.11","doi-asserted-by":"publisher","DOI":"10.1007\/BF02345439"},{"key":"2025081710122812000_12.1.e003185.12","unstructured":"Larburu N , Lopetegi T , Romero I . Comparative study of algorithms for Atrial Fibrillation detection. in 2011 Computing in Cardiology. 2011;265\u20138."},{"key":"2025081710122812000_12.1.e003185.13","doi-asserted-by":"crossref","first-page":"8101","DOI":"10.1007\/s00521-019-04318-2","article-title":"A novel hybrid network of fusing rhythmic and morphological features for atrial fibrillation detection on mobile ECG signals","volume":"32","author":"Fan","year":"2020","journal-title":"Neural Comput & Applic"},{"key":"2025081710122812000_12.1.e003185.14","doi-asserted-by":"crossref","first-page":"6071","DOI":"10.3390\/s22166071","article-title":"Atrioventricular Synchronization for Detection of Atrial Fibrillation and Flutter in One to Twelve ECG Leads Using a Dense Neural Network Classifier","volume":"22","author":"Jekova","year":"2022","journal-title":"Sensors (Basel)"},{"key":"2025081710122812000_12.1.e003185.15","doi-asserted-by":"crossref","first-page":"6848","DOI":"10.3390\/s21206848","article-title":"Application of Dense Neural Networks for Detection of Atrial Fibrillation and Ranking of Augmented ECG Feature Set","volume":"21","author":"Krasteva","year":"2021","journal-title":"Sensors (Basel)"},{"key":"2025081710122812000_12.1.e003185.16","doi-asserted-by":"crossref","first-page":"102462","DOI":"10.1016\/j.bspc.2021.102462","article-title":"Atrial fibrillation detection with and without atrial activity analysis using lead-I mobile ECG technology","volume":"66","author":"Tuboly","year":"2021","journal-title":"Biomed Signal Process Control"},{"key":"2025081710122812000_12.1.e003185.17","doi-asserted-by":"crossref","first-page":"657304","DOI":"10.3389\/fphys.2021.657304","article-title":"An Interpretable Hand-Crafted Feature-Based Model for Atrial Fibrillation Detection","volume":"12","author":"Rouhi","year":"2021","journal-title":"Front Physiol"},{"key":"2025081710122812000_12.1.e003185.18","unstructured":"Zhao Z . Transforming ECG Diagnosis:An In-depth Review of Transformer-based DeepLearning Models in Cardiovascular Disease Detection, 2023. Available: https:\/\/doi.org\/10.48550\/arXiv.2306.01249"},{"key":"2025081710122812000_12.1.e003185.19","doi-asserted-by":"publisher","DOI":"10.1016\/s0140-6736(19)31721-0"},{"key":"2025081710122812000_12.1.e003185.20","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1038\/s41746-023-00791-1","article-title":"Generalizable and robust deep learning algorithm for atrial fibrillation diagnosis across geography, ages and sexes","volume":"6","author":"Biton","year":"2023","journal-title":"NPJ Digit Med"},{"key":"2025081710122812000_12.1.e003185.21","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCULATIONAHA.120.047829"},{"key":"2025081710122812000_12.1.e003185.22","doi-asserted-by":"crossref","first-page":"108312","DOI":"10.1016\/j.ymssp.2021.108312","article-title":"An accurate and adaptative cutterhead torque prediction method for shield tunneling machines via adaptative residual long-short term memory network","volume":"165","author":"Jin","year":"2022","journal-title":"Mech Syst Signal Process"},{"key":"2025081710122812000_12.1.e003185.23","doi-asserted-by":"crossref","first-page":"105446","DOI":"10.1016\/j.knosys.2019.105446","article-title":"Automated detection of atrial fibrillation and atrial flutter in ECG signals based on convolutional and improved Elman neural network","volume":"193","author":"Wang","year":"2020","journal-title":"Knowl Based Syst"},{"key":"2025081710122812000_12.1.e003185.24","doi-asserted-by":"crossref","unstructured":"Mousavi S , Afghah F , Razi A , et al . ECGNET: learning where to attend for detection of atrial fibrillation with deep visual attention. 2019 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI); Chicago, IL, USA, 2019:1\u20134. doi:10.1109\/BHI.2019.8834637","DOI":"10.1109\/BHI.2019.8834637"},{"key":"2025081710122812000_12.1.e003185.25","doi-asserted-by":"crossref","first-page":"12818","DOI":"10.1038\/s41598-021-92172-5","article-title":"A new deep learning algorithm of 12-lead electrocardiogram for identifying atrial fibrillation during sinus rhythm","volume":"11","author":"Baek","year":"2021","journal-title":"Sci Rep"},{"key":"2025081710122812000_12.1.e003185.26","doi-asserted-by":"crossref","first-page":"103378","DOI":"10.1016\/j.compbiomed.2019.103378","article-title":"Accurate detection of atrial fibrillation from 12-lead ECG using deep neural network","volume":"116","author":"Cai","year":"2020","journal-title":"Comput Biol Med"},{"key":"2025081710122812000_12.1.e003185.27","doi-asserted-by":"crossref","unstructured":"Couceiro RJ dos S , Carvalho P , Henriques J , et al . In international conference on biomedical and health informatics (eds. In: Zhang Y-T , Carvalho P , Magjarevic R , eds. 2019:205\u201310. doi:10.1007\/978-981-10-4505-9_46","DOI":"10.1007\/978-981-10-4505-9_46"},{"key":"2025081710122812000_12.1.e003185.28","doi-asserted-by":"crossref","unstructured":"Chen T , Guestrin C . In proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining 785\u2013794. 2016 doi:10.1145\/2939672.2939785","DOI":"10.1145\/2939672.2939785"},{"key":"2025081710122812000_12.1.e003185.29","doi-asserted-by":"crossref","first-page":"124003","DOI":"10.1088\/1361-6579\/abc960","article-title":"Classification of 12-lead ECGs: the PhysioNet\/Computing in Cardiology Challenge 2020","volume":"41","author":"Perez Alday","year":"2020","journal-title":"Physiol Meas"},{"key":"2025081710122812000_12.1.e003185.30","doi-asserted-by":"crossref","first-page":"154","DOI":"10.1038\/s41597-020-0495-6","article-title":"PTB-XL, a large publicly available electrocardiography dataset","volume":"7","author":"Wagner","year":"2020","journal-title":"Sci Data"},{"key":"2025081710122812000_12.1.e003185.31","doi-asserted-by":"crossref","first-page":"1368","DOI":"10.1166\/jmihi.2018.2442","article-title":"An Open Access Database for Evaluating the Algorithms of Electrocardiogram Rhythm and Morphology Abnormality Detection","volume":"8","author":"Liu","year":"2018","journal-title":"j med imaging hlth inform"},{"key":"2025081710122812000_12.1.e003185.32","unstructured":"Gent P , Farah H , Nes N , et al . Heart rate analysis for human factors: development and validation of an open source toolkit for noisy naturalistic heart rate data. 6th Humanist Conference. Proc. Of; 2018"},{"key":"2025081710122812000_12.1.e003185.33","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"2025081710122812000_12.1.e003185.34","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"2025081710122812000_12.1.e003185.35","unstructured":"Simonyan K , Zisserman A . Very Deep Convolutional Networks for Large-Scale Image Recognition, 2015. Available: https:\/\/doi.org\/10.48550\/arXiv.1409.1556"},{"key":"2025081710122812000_12.1.e003185.36","doi-asserted-by":"crossref","unstructured":"He K , Zhang X , Ren S , et al . Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Las Vegas, NV, USA, 2016 doi:10.1109\/CVPR.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"2025081710122812000_12.1.e003185.37","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0268-3"},{"key":"2025081710122812000_12.1.e003185.38","doi-asserted-by":"crossref","unstructured":"Szegedy C , Ioffe S , Vanhoucke V , et al . Inception-v4, inception-resnet and the impact of residual connections on learning. 2016. Available: https:\/\/doi.org\/10.48550\/arXiv.1602.07261","DOI":"10.1609\/aaai.v31i1.11231"},{"key":"2025081710122812000_12.1.e003185.39","doi-asserted-by":"crossref","unstructured":"Long J , Shelhamer E , Darrell T . Fully convolutional networks for semantic segmentation. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Boston, MA, USA. doi:10.1109\/CVPR.2015.7298965","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"2025081710122812000_12.1.e003185.40","doi-asserted-by":"crossref","unstructured":"Cho K , et al . Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation, 2014. Available: https:\/\/doi.org\/10.48550\/arXiv.1406.1078","DOI":"10.3115\/v1\/D14-1179"},{"key":"2025081710122812000_12.1.e003185.41","unstructured":"Vaswani A , et al . Attention is all you need."},{"key":"2025081710122812000_12.1.e003185.42","unstructured":"TensorFlow A . tf.keras.callbacks.ReduceLROnPlateau | TensorFlow v2.16.1, Available: https:\/\/www.tensorflow.org\/api_docs\/python\/tf\/keras\/callbacks\/ReduceLROnPlateau"},{"key":"2025081710122812000_12.1.e003185.43","doi-asserted-by":"crossref","unstructured":"Akiba T , Sano S , Yanase T , et al . In proceedings of the 25th acm sigkdd international conference on knowledge discovery & data mining 2623\u20132631. 2019 doi:10.1145\/3292500.3330701","DOI":"10.1145\/3292500.3330701"},{"key":"2025081710122812000_12.1.e003185.44","doi-asserted-by":"crossref","unstructured":"Khorsheed MS , Karim AA . Analyzing the impact of transfer learning on explanation accuracy in deep learning-based ECG recognition systems. Open Engineering 2024;14. doi:10.1515\/eng-2024-0066","DOI":"10.1515\/eng-2024-0066"},{"key":"2025081710122812000_12.1.e003185.45","doi-asserted-by":"crossref","first-page":"149","DOI":"10.1038\/s41746-022-00689-4","article-title":"Integrated multimodal artificial intelligence framework for healthcare applications","volume":"5","author":"Soenksen","year":"2022","journal-title":"NPJ Digit Med"},{"key":"2025081710122812000_12.1.e003185.46","doi-asserted-by":"crossref","unstructured":"Stoyanov T , Krasteva V , Naydenov Naydenov S , et al . Deep transfer learning for detection of atrial fibrillation using holter ecg color maps. 2024 Computing in Cardiology Conference; 2024 doi:10.22489\/CinC.2024.283","DOI":"10.22489\/CinC.2024.283"}],"container-title":["Open Heart"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/openhrt-2025-003185","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,17]],"date-time":"2025-08-17T17:12:38Z","timestamp":1755450758000},"score":1,"resource":{"primary":{"URL":"https:\/\/openheart.bmj.com\/lookup\/doi\/10.1136\/openhrt-2025-003185"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1]]},"references-count":46,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,3,31]]},"published-print":{"date-parts":[[2025,1]]}},"alternative-id":["10.1136\/openhrt-2025-003185"],"URL":"https:\/\/doi.org\/10.1136\/openhrt-2025-003185","relation":{},"ISSN":["2053-3624"],"issn-type":[{"value":"2053-3624","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1]]}}}