{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,2]],"date-time":"2026-09-02T11:14:57Z","timestamp":1788347697596,"version":"build-2803163510"},"reference-count":38,"publisher":"IOP Publishing","issue":"5","license":[{"start":{"date-parts":[[2026,9,2]],"date-time":"2026-09-02T00:00:00Z","timestamp":1788307200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,9,2]],"date-time":"2026-09-02T00:00:00Z","timestamp":1788307200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Biomed. Phys. Eng. Express"],"published-print":{"date-parts":[[2026,10,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Cardiovascular diseases remain a major global health burden, making accurate electrocardiogram (ECG) analysis essential for timely diagnosis. While deep learning-based methods have made significant progress in automating ECG classification, they are often limited\n                    <jats:bold>by<\/jats:bold>\n                    the variability of ECG signals, the frequent presence of multiple cardiac abnormalities, and the inadequate integration of diverse clinical data. This study proposes Deformable Lead-Path and Multi-Modal Fusion Network (DLM-Net): a robust deep learning framework for multi-label ECG classification, designed to address these challenges. DLM-Net incorporates several key innovations. First, the Lead-Specific Deformable Path applies deformable convolutions independently to each lead. This enables adaptive modeling of non-rigid and complex waveform variations. It is also enhanced by a high-frequency module, which captures fine-grained temporal details. Secondly, a spatial path based on a ResNet-like backbone and feature pyramid network fuses multi-scale representations in order to effectively capture both global and local ECG patterns. In parallel, the lead path uses a lead attention Transformer to explicitly model inter-lead dependencies. Finally, ECG features from the spatial and lead paths are integrated with clinical tabular data through a multimodal fusion module, in which compact patient-level tabular embeddings provide complementary contextual information for ECG representation. Extensive experiments on three public multi-label ECG datasets\u2014 Physikalisch-Technische Bundesanstalt-XL, CPSC2018 and Chapman\u2014demonstrate that DLM-Net achieves competitive and stable performance across datasets, with label-wise accuracies of 89.11%, 96.05% and 97.82%, respectively. Overall, DLM-Net provides an effective and flexible framework for multi-label ECG classification, offering improved accuracy and robustness across diverse datasets and demonstrating the effectiveness of the proposed architecture in handling complex multi-label diagnostic tasks in retrospective settings.\n                  <\/jats:p>","DOI":"10.1088\/2057-1976\/ae9e39","type":"journal-article","created":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T22:48:17Z","timestamp":1787698097000},"page":"055002","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A deformable lead-attention fusion network for multi-label ECG classification integrating clinical metadata"],"prefix":"10.1088","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-6510-076X","authenticated-orcid":true,"given":"Enjun","family":"Zhou","sequence":"first","affiliation":[{"name":"University of Shanghai for Science and Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-9925-5215","authenticated-orcid":false,"given":"Xiangyong","family":"Kong","sequence":"additional","affiliation":[{"name":"University of Shanghai for Science and Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-1823-9025","authenticated-orcid":true,"given":"Hao","family":"Wang","sequence":"additional","affiliation":[{"name":"University of Shanghai for Science and Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-3620-6071","authenticated-orcid":false,"given":"Chenrui","family":"Bai","sequence":"additional","affiliation":[{"name":"University of Shanghai for Science and Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,9,2]]},"reference":[{"key":"bpexae9e39bib1","doi-asserted-by":"publisher","first-page":"2019","DOI":"10.1038\/s41591-022-01990-1","type":"journal-article","article-title":"The global burden of disease study at 30 years","volume":"28","author":"Murray","year":"2022","journal-title":"Nat. Med."},{"key":"bpexae9e39bib2","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1016\/j.ccep.2010.10.012","type":"journal-article","article-title":"Overview of basic mechanisms of cardiac arrhythmia","volume":"3","author":"Antzelevitch","year":"2011","journal-title":"Card. Electrophysiol. Clin."},{"key":"bpexae9e39bib3","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1016\/j.bspc.2018.03.003","type":"journal-article","article-title":"A survey on ECG analysis","volume":"43","author":"Berkaya","year":"2018","journal-title":"Biomed. Signal Process. Control"},{"key":"bpexae9e39bib4","doi-asserted-by":"publisher","first-page":"9460","DOI":"10.3390\/app11209460","type":"journal-article","article-title":"Cardiac arrhythmia classification based on one-dimensional morphological features","volume":"11","author":"Lee","year":"2021","journal-title":"Appl. Sci."},{"key":"bpexae9e39bib5","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2021.103344","type":"journal-article","article-title":"Application of an EMG interference filtering method to dynamic ECGs based on an adaptive wavelet-wiener filter and adaptive moving average filter","volume":"72","author":"Li","year":"2022","journal-title":"Biomed. Signal Process. Control"},{"key":"bpexae9e39bib6","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2022.103583","type":"journal-article","article-title":"Electrocardiogram signal filtering using circulant singular spectrum analysis and cascaded Savitzky-Golay filter","volume":"75","author":"Chaitanya","year":"2022","journal-title":"Biomed. Signal Process. Control"},{"key":"bpexae9e39bib7","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.108764","type":"journal-article","article-title":"Elicitation of fetal ECG from abdominal recordings using blind source separation techniques and robust set membership affine projection algorithm for signal quality enhancement","volume":"178","author":"Diwan","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"bpexae9e39bib8","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","type":"journal-article","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"Lecun","year":"1998","journal-title":"Proc. IEEE"},{"key":"bpexae9e39bib9","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1145\/3065386","type":"journal-article","article-title":"ImageNet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Commun. ACM"},{"key":"bpexae9e39bib10","doi-asserted-by":"publisher","first-page":"770","DOI":"10.1109\/CVPR.2016.90","type":"conference-proceedings","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"bpexae9e39bib11","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","type":"preprint"},{"key":"bpexae9e39bib12","doi-asserted-by":"publisher","first-page":"395","DOI":"10.1109\/ICCCS49078.2020.9118570","type":"conference-proceedings","article-title":"Image classification and detection of cigarette combustion cone based on inception resnet V2","author":"Deng","year":"2020"},{"key":"bpexae9e39bib13","article-title":"Attention is all you need","author":"Vaswani","year":"2023","type":"preprint"},{"key":"bpexae9e39bib14","doi-asserted-by":"publisher","first-page":"1471","DOI":"10.1007\/s11760-020-01681-9","type":"journal-article","article-title":"Optimized time\u2013frequency features and semi-supervised SVM to heartbeat classification","volume":"14","author":"Lekhal","year":"2020","journal-title":"Signal Image Video Process."},{"key":"bpexae9e39bib15","doi-asserted-by":"publisher","first-page":"399","DOI":"10.1016\/j.measurement.2008.08.004","type":"journal-article","article-title":"Classification of ECG complexes using self-organizing CMAC","volume":"42","author":"Wen","year":"2009","journal-title":"Measurement"},{"key":"bpexae9e39bib16","doi-asserted-by":"publisher","first-page":"892","DOI":"10.1109\/JBHI.2014.2332001","type":"journal-article","article-title":"Electrocardiogram classification using reservoir computing with logistic regression","volume":"19","author":"Escalona-Moran","year":"2015","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"bpexae9e39bib17","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1038\/s41591-018-0268-3","type":"journal-article","article-title":"Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network","volume":"25","author":"Hannun","year":"2019","journal-title":"Nat. Med."},{"key":"bpexae9e39bib18","doi-asserted-by":"publisher","author":"Natarajan","year":"2020","DOI":"10.22489\/CinC.2020.107","type":"book"},{"key":"bpexae9e39bib19","doi-asserted-by":"publisher","first-page":"278","DOI":"10.1016\/j.compbiomed.2018.06.002","type":"journal-article","article-title":"Automated diagnosis of arrhythmia using combination of CNN and LSTM techniques with variable length heart beats","volume":"102","author":"Oh","year":"2018","journal-title":"Comput. Biol. Med."},{"key":"bpexae9e39bib20","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2022.104106","type":"journal-article","article-title":"Post-processing refined ECG delineation based on 1D-UNet","volume":"79","author":"Chen","year":"2023","journal-title":"Biomed. Signal Process. Control"},{"key":"bpexae9e39bib21","doi-asserted-by":"publisher","first-page":"7161","DOI":"10.1109\/JBHI.2025.3566531","type":"journal-article","article-title":"Enhancing ECG classification in cardiac diagnostics: a novel approach using adaptive focal cross-entropy loss function","volume":"29","author":"Monday","year":"2025","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"bpexae9e39bib22","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2025.107835","type":"journal-article","article-title":"Multimodal prediction of catheter ablation outcomes in patients with persistent atrial fibrillation","volume":"191","author":"Zhou","year":"2025","journal-title":"Neural Netw."},{"key":"bpexae9e39bib23","doi-asserted-by":"publisher","first-page":"891","DOI":"10.1038\/s44161-025-00679-1","type":"journal-article","article-title":"Multimodal AI to forecast arrhythmic death in hypertrophic cardiomyopathy","volume":"4","author":"Lai","year":"2025","journal-title":"Nat. Cardiovasc. Res."},{"key":"bpexae9e39bib24","doi-asserted-by":"publisher","first-page":"154","DOI":"10.1038\/s41597-020-0495-6","type":"journal-article","article-title":"PTB-XL, a large publicly available electrocardiography dataset","volume":"7","author":"Wagner","year":"2020","journal-title":"Sci. Data"},{"key":"bpexae9e39bib25","doi-asserted-by":"publisher","first-page":"1368","DOI":"10.1166\/jmihi.2018.2442","type":"journal-article","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 Health Inform."},{"key":"bpexae9e39bib26","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1038\/s41597-020-0386-x","type":"journal-article","article-title":"A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients","volume":"7","author":"Zheng","year":"2020","journal-title":"Sci. Data"},{"key":"bpexae9e39bib27","doi-asserted-by":"publisher","first-page":"764","DOI":"10.1109\/ICCV.2017.89","type":"conference-proceedings","article-title":"Deformable convolutional networks","author":"Dai","year":"2017"},{"key":"bpexae9e39bib28","doi-asserted-by":"publisher","first-page":"7132","DOI":"10.1109\/CVPR.2018.00745","type":"conference-proceedings","article-title":"Squeeze-and-excitation networks","author":"Hu","year":"2018"},{"key":"bpexae9e39bib29","doi-asserted-by":"publisher","first-page":"936","DOI":"10.1109\/CVPR.2017.106","type":"conference-proceedings","article-title":"Feature pyramid networks for object detection","author":"Lin","year":"2017"},{"key":"bpexae9e39bib30","doi-asserted-by":"publisher","first-page":"2999","DOI":"10.1109\/ICCV.2017.324","type":"conference-proceedings","article-title":"Focal loss for dense object detection","author":"Lin","year":"2017"},{"key":"bpexae9e39bib31","doi-asserted-by":"publisher","first-page":"692","DOI":"10.1109\/TNNLS.2023.3332217","type":"journal-article","article-title":"Spectral cross-domain neural network with soft-adaptive threshold spectral enhancement","volume":"36","author":"Liu","year":"2025","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"bpexae9e39bib32","doi-asserted-by":"publisher","first-page":"1336","DOI":"10.1007\/s10618-020-00690-z","type":"journal-article","article-title":"TEASER: early and accurate time series classification","volume":"34","author":"Sch\u00e4fer","year":"2020","journal-title":"Data Min. Knowl. Discov."},{"key":"bpexae9e39bib33","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1109\/JBHI.2023.3327025","type":"journal-article","article-title":"ST-ReGE: a novel spatial-temporal residual graph convolutional network for CVD","volume":"28","author":"Zhang","year":"2024","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"bpexae9e39bib34","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2024.108315","type":"journal-article","article-title":"CPSS: fusing consistency regularization and pseudo-labeling techniques for semi-supervised deep cardiovascular disease detection using all unlabeled electrocardiograms","volume":"254","author":"Shi","year":"2024","journal-title":"Comput. Methods Programs Biomed."},{"key":"bpexae9e39bib35","doi-asserted-by":"publisher","first-page":"3791","DOI":"10.1109\/JBHI.2022.3173655","type":"journal-article","article-title":"Semi-supervised learning for automatic atrial fibrillation detection in 24-hour holter monitoring","volume":"26","author":"Zhang","year":"2022","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"bpexae9e39bib36","doi-asserted-by":"publisher","DOI":"10.1016\/j.medj.2025.100668","type":"journal-article","article-title":"Clinician-artificial intelligence collaboration: a win-win solution for efficiency and reliability in atrial fibrillation diagnosis","volume":"6","author":"Zhang","year":"2025","journal-title":"Med"},{"key":"bpexae9e39bib37","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1007\/s11263-019-01228-7","type":"journal-article","article-title":"Grad-CAM: visual explanations from deep networks via gradient-based localization","volume":"128","author":"Selvaraju","year":"2019","journal-title":"Int. J. Comput. Vis."},{"key":"bpexae9e39bib38","doi-asserted-by":"publisher","first-page":"414","DOI":"10.1016\/j.medj.2024.02.006","type":"journal-article","article-title":"Artificial-intelligence-based risk prediction and mechanism discovery for atrial fibrillation using heart beat-to-beat intervals","volume":"5","author":"Lin","year":"2024","journal-title":"Med"}],"container-title":["Biomedical Physics &amp; Engineering Express"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,2]],"date-time":"2026-09-02T10:32:32Z","timestamp":1788345152000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/ae9e39"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,2]]},"references-count":38,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,9,2]]},"published-print":{"date-parts":[[2026,10,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2057-1976\/ae9e39","relation":{},"ISSN":["2057-1976"],"issn-type":[{"value":"2057-1976","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,2]]},"assertion":[{"value":"A deformable lead-attention fusion network for multi-label ECG classification integrating clinical metadata","name":"article_title","label":"Article Title"},{"value":"Biomedical Physics & Engineering Express","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2026-02-10","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-25","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-09-02","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}