{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T13:55:43Z","timestamp":1790862943718,"version":"4.1.0"},"reference-count":30,"publisher":"AIP Publishing","issue":"10","funder":[{"name":"The Brain Science and Brain-like Intelligence Technology-National Science and Technology Major Project","award":["2021ZD0201600"],"award-info":[{"award-number":["2021ZD0201600"]}]}],"content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,10,1]]},"abstract":"<jats:p>Discriminating interictal and ictal brain states is important for both seizure-state classification and the characterization of seizure-related network reorganization in drug-resistant epilepsy. Intracranial electroencephalography (iEEG) provides high-resolution recordings of brain activity, but its sparse, irregular, and patient-specific electrode geometry requires models that go beyond Euclidean representations. In this work, we propose a multivariate weighted joint recurrence network coupled with a graph diffusion convolution network for binary interictal\/ictal classification. Each iEEG segment is first embedded into a reconstructed phase space and transformed into a recurrence-based functional network, in which weighted edges quantify nonlinear dynamical coupling between implanted contacts. A graph diffusion convolution mechanism is then applied to capture higher-order dependencies across the implanted brain topology. On a nine-patient subset of the Hospital of the University of Pennsylvania iEEG Epilepsy Dataset, the proposed framework achieved an average accuracy of 93.4% and outperformed standard graph convolutional and Euclidean deep-learning baselines. In addition, recurrence networks derived from ictal and interictal periods showed state-dependent differences in weighted topological organization, suggesting seizure-related reconfiguration of nonlinear brain-network dynamics. These results indicate that recurrence-based graph diffusion provides an effective and interpretable nonlinear framework for seizure-state classification and for modeling ictal network reorganization in iEEG.<\/jats:p>","DOI":"10.1063\/5.0344726","type":"journal-article","created":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T12:56:37Z","timestamp":1790859397000},"update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":0,"title":["Graph diffusion convolution on multivariate weighted joint recurrence networks for seizure-state classification in intracranial EEG"],"prefix":"10.1063","volume":"36","author":[{"given":"Qing","family":"Cai","sequence":"first","affiliation":[{"name":"School of Artificial Intelligence, Tiangong University 1 , Tianjin,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yushi","family":"Hao","sequence":"additional","affiliation":[{"name":"School of Electrical and Information Engineering, Tianjin University 2 , Tianjin,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9108-6182","authenticated-orcid":false,"given":"Jianpeng","family":"An","sequence":"additional","affiliation":[{"name":"School of Electrical and Information Engineering, Tianjin University 2 , Tianjin,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6981-0165","authenticated-orcid":false,"given":"Chao","family":"Ma","sequence":"additional","affiliation":[{"name":"School of Electrical and Information Engineering, Tianjin University 2 , Tianjin,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3591-5898","authenticated-orcid":false,"given":"Mengyu","family":"Li","sequence":"additional","affiliation":[{"name":"School of Electrical and Information Engineering, Tianjin University 2 , Tianjin,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9551-202X","authenticated-orcid":false,"given":"Zhongke","family":"Gao","sequence":"additional","affiliation":[{"name":"School of Electrical and Information Engineering, Tianjin University 2 , Tianjin,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"317","published-online":{"date-parts":[[2026,10,1]]},"reference":[{"key":"2026100108555848400_c1","doi-asserted-by":"publisher","first-page":"357","DOI":"10.1016\/S1474-4422(18)30454-X","article-title":"Global, regional, and national burden of epilepsy, 1990\u20132016: A systematic analysis for the global burden of disease study 2016","volume":"18","year":"2019","journal-title":"Lancet Neurol."},{"key":"2026100108555848400_c2","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1056\/NEJM200002033420503","article-title":"Early identification of refractory epilepsy","volume":"342","year":"2000","journal-title":"N. Engl. J. Med."},{"key":"2026100108555848400_c3","doi-asserted-by":"publisher","first-page":"1955","DOI":"10.1016\/j.clinph.2007.06.006","article-title":"Increasing synchronization may promote seizure termination: Evidence from status epilepticus","volume":"118","year":"2007","journal-title":"Clin. Neurophysiol."},{"key":"2026100108555848400_c4","doi-asserted-by":"publisher","first-page":"918","DOI":"10.1016\/j.clinph.2006.12.002","article-title":"Small-world networks and epilepsy: Graph theoretical analysis of intracerebrally recorded mesial temporal lobe seizures","volume":"118","year":"2007","journal-title":"Clin. Neurophysiol."},{"key":"2026100108555848400_c5","doi-asserted-by":"publisher","first-page":"493","DOI":"10.1007\/s12264-016-0035-8","article-title":"Advances of intracranial electroencephalography in localizing the epileptogenic zone","volume":"32","year":"2016","journal-title":"Neurosci. Bull."},{"key":"2026100108555848400_c6","doi-asserted-by":"publisher","first-page":"056019","DOI":"10.1088\/1741-2552\/ac90ed","article-title":"Addressing spatial bias in intracranial EEG functional connectivity analyses for epilepsy surgical planning","volume":"19","year":"2022","journal-title":"J. Neural Eng."},{"key":"2026100108555848400_c7","doi-asserted-by":"publisher","first-page":"188","DOI":"10.1177\/155005940503600309","article-title":"Approximate entropy of the electroencephalogram in healthy awake subjects and absence epilepsy patients","volume":"36","year":"2005","journal-title":"Clin. EEG Neurosci."},{"key":"2026100108555848400_c8","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1007\/BF01140588","article-title":"Phase space topography and the Lyapunov exponent of electrocorticograms in partial seizures","volume":"2","year":"1990","journal-title":"Brain Topogr."},{"key":"2026100108555848400_c9","doi-asserted-by":"publisher","first-page":"237","DOI":"10.1016\/j.physrep.2006.11.001","article-title":"Recurrence plots for the analysis of complex systems","volume":"438","year":"2007","journal-title":"Phys. Rep."},{"key":"2026100108555848400_c10","doi-asserted-by":"publisher","first-page":"214","DOI":"10.1016\/j.physleta.2004.07.066","article-title":"Multivariate recurrence plots","volume":"330","year":"2004","journal-title":"Phys. Lett. A"},{"key":"2026100108555848400_c11","doi-asserted-by":"publisher","first-page":"033025","DOI":"10.1088\/1367-2630\/12\/3\/033025","article-title":"Recurrence networks\u2014A novel paradigm for nonlinear time series analysis","volume":"12","year":"2010","journal-title":"New J. Phys."},{"key":"2026100108555848400_c12","doi-asserted-by":"publisher","first-page":"18998","DOI":"10.1038\/s41598-022-23656-1","article-title":"Graph-generative neural network for EEG-based epileptic seizure detection via discovery of dynamic brain functional connectivity","volume":"12","year":"2022","journal-title":"Sci. Rep."},{"key":"2026100108555848400_c13","first-page":"7865","article-title":"On over-squashing in message passing neural networks: The impact of width, depth, and topology","year":"2023"},{"key":"2026100108555848400_c14","doi-asserted-by":"publisher","first-page":"E5321","DOI":"10.1073\/pnas.1401752111","article-title":"Network dynamics of the brain and influence of the epileptic seizure onset zone","volume":"111","year":"2014","journal-title":"Proc. Natl. Acad. Sci. U.S.A."},{"key":"2026100108555848400_c15"},{"key":"2026100108555848400_c16"},{"key":"2026100108555848400_c17","doi-asserted-by":"publisher","first-page":"2248","DOI":"10.1093\/brain\/awad007","article-title":"Quantitative approaches to guide epilepsy surgery from intracranial EEG","volume":"146","year":"2023","journal-title":"Brain"},{"key":"2026100108555848400_c18","first-page":"366","article-title":"Detecting strange attractors in turbulence","year":"2006"},{"key":"2026100108555848400_c19","doi-asserted-by":"publisher","first-page":"3403","DOI":"10.1103\/PhysRevA.45.3403","article-title":"Determining embedding dimension for phase-space reconstruction using a geometrical construction","volume":"45","year":"1992","journal-title":"Phys. Rev. A"},{"key":"2026100108555848400_c20","doi-asserted-by":"publisher","first-page":"1134","DOI":"10.1103\/PhysRevA.33.1134","article-title":"Independent coordinates for strange attractors from mutual information","volume":"33","year":"1986","journal-title":"Phys. Rev. A"},{"key":"2026100108555848400_c21","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1109\/TAU.1967.1161901","article-title":"The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms","volume":"15","year":"1967","journal-title":"IEEE Trans. Audio Electroacoust."},{"key":"2026100108555848400_c22","doi-asserted-by":"publisher","first-page":"1059","DOI":"10.1016\/j.neuroimage.2009.10.003","article-title":"Complex network measures of brain connectivity: Uses and interpretations","volume":"52","year":"2010","journal-title":"NeuroImage"},{"key":"2026100108555848400_c23","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1016\/S0169-7552(98)00110-X","article-title":"The anatomy of a large-scale hypertextual web search engine","volume":"30","year":"1998","journal-title":"Comput. Netw. ISDN Syst."},{"key":"2026100108555848400_c24","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1016\/j.nicl.2014.04.006","article-title":"Patient-specific connectivity pattern of epileptic network in frontal lobe epilepsy","volume":"4","year":"2014","journal-title":"NeuroImage: Clin."},{"key":"2026100108555848400_c25","doi-asserted-by":"publisher","first-page":"102293","DOI":"10.1016\/j.bspc.2020.102293","article-title":"Epilepsy prediction through optimized multidimensional sample entropy and Bi-LSTM","volume":"64","year":"2021","journal-title":"Biomed. Signal Process. Control"},{"key":"2026100108555848400_c26","unstructured":"X.\n            Yao\n          , Q.Cheng, and G.Zhang, \u201cAutomated classification of seizures against nonseizures: A deep learning approach,\u201d arXiv:1906.02745 (2019)."},{"key":"2026100108555848400_c27","doi-asserted-by":"publisher","first-page":"056013","DOI":"10.1088\/1741-2552\/aace8c","article-title":"EEGNet: A compact convolutional neural network for EEG-based brain\u2013computer interfaces","volume":"15","year":"2018","journal-title":"J. Neural Eng."},{"key":"2026100108555848400_c28","doi-asserted-by":"publisher","first-page":"5391","DOI":"10.1002\/hbm.23730","article-title":"Deep learning with convolutional neural networks for EEG decoding and visualization","volume":"38","year":"2017","journal-title":"Hum. Brain Mapp."},{"key":"2026100108555848400_c29","doi-asserted-by":"publisher","first-page":"120209","DOI":"10.1016\/j.neuroimage.2023.120209","article-title":"LMDA-Net: A lightweight multi-dimensional attention network for general EEG-based brain-computer interfaces and interpretability","volume":"276","year":"2023","journal-title":"NeuroImage"},{"key":"2026100108555848400_c30","unstructured":"T. N.\n            Kipf\n           and M.Welling, \u201cVariational graph auto-encoders,\u201d arXiv:1611.07308 (2016)."}],"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\/5.0344726\/21158421\/103105_1_5.0344726.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/aip\/cha\/article-pdf\/doi\/10.1063\/5.0344726\/21158421\/103105_1_5.0344726.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T12:56:39Z","timestamp":1790859399000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/cha\/article\/36\/10\/103105\/3405742\/Graph-diffusion-convolution-on-multivariate"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,1]]},"references-count":30,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2026,10,1]]}},"URL":"https:\/\/doi.org\/10.1063\/5.0344726","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":"103105"}}