DBTR-AGF: A Dual-Branch Transformer–Recurrent Network with Adaptive Gating for Epileptic Seizure Prediction

DBTR-AGF: A Dual-Branch Transformer–Recurrent Network with Adaptive Gating for Epileptic Seizure Prediction | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article DBTR-AGF: A Dual-Branch Transformer–Recurrent Network with Adaptive Gating for Epileptic Seizure Prediction Sobhan Khalifeh, Hojat Ghimatgar, Mojtaba Mansorinejada, Samira soleimani nasab This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8276098/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract Epilepsy is a chronic neurological disorder marked by unpredictable seizures, creating a pressing need for reliable early-warning systems. Despite advances in deep learning for EEG-based prediction, many models show weak patient-independent generalization, elevated false-positive rates, and sensitivity to limited data. We present DBTR-AGF (Dual-Branch Transformer–Recurrent with Adaptive Gating Fusion), which unifies a temporal branch (Transformer → BiLSTM) that captures global-to-local temporal dependencies with a spectral branch (Transformer → Temporal Convolutional Network) that extracts multi-scale spectral structure efficiently. A learnable adaptive-gating module fuses both representations on a per-sample basis. To improve robustness and reduce spurious alarms, we pair comprehensive regularization (Mixup, label smoothing, L2 weight decay) with Hidden Markov Model–based temporal smoothing at inference. On the CHB-MIT scalp-EEG dataset under a clinically motivated leave-one-subject-out protocol, DBTR-AGF attains 99.02% accuracy, 98.95% sensitivity, 99.91% specificity, and 0.12/h false-positive rate, outperforming recent baselines. Performance degrades by only 4.7% when training data are halved, indicating robustness in low-data regimes. These results suggest DBTR-AGF is a strong candidate for practical, patient-independent seizure forecasting. Biological sciences/Computational biology and bioinformatics Health sciences/Neurology Biological sciences/Neuroscience Electroencephalography Seizure Prediction DBTR-AGF Transformer Networks Temporal-Spectral Fusion Patient-Independent Generalization Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 30 Apr, 2026 Reviewers invited by journal 23 Apr, 2026 Editor assigned by journal 21 Apr, 2026 Editor invited by journal 09 Dec, 2025 Submission checks completed at journal 05 Dec, 2025 First submitted to journal 05 Dec, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8276098","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":633602694,"identity":"f1bbcdab-0c9e-4569-b7ff-de5c30aa61d6","order_by":0,"name":"Sobhan Khalifeh","email":"","orcid":"","institution":"Persian Gulf University","correspondingAuthor":false,"prefix":"","firstName":"Sobhan","middleName":"","lastName":"Khalifeh","suffix":""},{"id":633602695,"identity":"a90654af-04d2-48af-b143-f76ba6b436b7","order_by":1,"name":"Hojat Ghimatgar","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIiWNgGAWjYDACCcYGCIMdSjMwE6UlAcjgOUC0FhAB0iKRQKS75KObWzf8/FErbz7z7eHXPAx28gzsvA/wajG8c7DtZk/CccM5t/PSrHkYkg0bmNkN8GuZkdh2gyfhGOMM6RwzYx4G5gQGZjb8DgNpufkn4Zj9DMkzIC31hLXISyS23eZJqEmcIcFj/JiH4TBhLQYgLTJpB5Jn8OSlMc4xOG7YRtCWGenPbr6xqbOdwX728Ic3FdXy/PzHCNhyAEwdBmIeNgkGYFgRsANoSwOYqgNpYf5ASPUoGAWjYBSMTAAA9OlAiidgrVgAAAAASUVORK5CYII=","orcid":"","institution":"Persian Gulf University","correspondingAuthor":true,"prefix":"","firstName":"Hojat","middleName":"","lastName":"Ghimatgar","suffix":""},{"id":633602696,"identity":"66ecebb2-1494-4203-84d8-dc81f4745c56","order_by":2,"name":"Mojtaba Mansorinejada","email":"","orcid":"","institution":"Persian Gulf University","correspondingAuthor":false,"prefix":"","firstName":"Mojtaba","middleName":"","lastName":"Mansorinejada","suffix":""},{"id":633602697,"identity":"8d16566e-ca45-4c98-8f99-d46b5af8e8d8","order_by":3,"name":"Samira soleimani nasab","email":"","orcid":"","institution":"Persian Gulf University","correspondingAuthor":false,"prefix":"","firstName":"Samira","middleName":"soleimani","lastName":"nasab","suffix":""}],"badges":[],"createdAt":"2025-12-04 06:23:42","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8276098/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8276098/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108493403,"identity":"84795cac-b640-486f-bc8a-d1c64125e0c4","added_by":"auto","created_at":"2026-05-05 10:00:18","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":634563,"visible":true,"origin":"","legend":"","description":"","filename":"RevisedManuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8276098/v1_covered_de9fbebd-f0f9-45fd-9df6-a5e7329d670b.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"DBTR-AGF: A Dual-Branch Transformer–Recurrent Network with Adaptive Gating for Epileptic Seizure Prediction","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Electroencephalography, Seizure Prediction, DBTR-AGF, Transformer Networks, Temporal-Spectral Fusion, Patient-Independent Generalization","lastPublishedDoi":"10.21203/rs.3.rs-8276098/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8276098/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Epilepsy is a chronic neurological disorder marked by unpredictable seizures, creating a pressing need for reliable early-warning systems. Despite advances in deep learning for EEG-based prediction, many models show weak patient-independent generalization, elevated false-positive rates, and sensitivity to limited data. We present DBTR-AGF (Dual-Branch Transformer–Recurrent with Adaptive Gating Fusion), which unifies a temporal branch (Transformer → BiLSTM) that captures global-to-local temporal dependencies with a spectral branch (Transformer → Temporal Convolutional Network) that extracts multi-scale spectral structure efficiently. A learnable adaptive-gating module fuses both representations on a per-sample basis. To improve robustness and reduce spurious alarms, we pair comprehensive regularization (Mixup, label smoothing, L2 weight decay) with Hidden Markov Model–based temporal smoothing at inference. On the CHB-MIT scalp-EEG dataset under a clinically motivated leave-one-subject-out protocol, DBTR-AGF attains 99.02% accuracy, 98.95% sensitivity, 99.91% specificity, and 0.12/h false-positive rate, outperforming recent baselines. Performance degrades by only 4.7% when training data are halved, indicating robustness in low-data regimes. These results suggest DBTR-AGF is a strong candidate for practical, patient-independent seizure forecasting.","manuscriptTitle":"DBTR-AGF: A Dual-Branch Transformer–Recurrent Network with Adaptive Gating for Epileptic Seizure Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-04 10:47:25","doi":"10.21203/rs.3.rs-8276098/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"145133463452478411813084165218814902348","date":"2026-04-30T07:20:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-23T12:04:10+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-21T11:17:04+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-12-10T03:51:18+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-05T11:10:01+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-12-05T10:57:24+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"04676805-1dec-42ee-88d7-ed4d7814c590","owner":[],"postedDate":"May 4th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"145133463452478411813084165218814902348","date":"2026-04-30T07:20:51+00:00","index":85,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67436271,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":67436272,"name":"Health sciences/Neurology"},{"id":67436273,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-05-04T10:47:25+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-04 10:47:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8276098","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8276098","identity":"rs-8276098","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}