{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,13]],"date-time":"2026-02-13T07:01:01Z","timestamp":1770966061006,"version":"3.50.1"},"reference-count":33,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T00:00:00Z","timestamp":1761868800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Temporal lobe epilepsy (TLE) represents a significant neurological disorder with complex genetic underpinnings. This study aimed to develop an interpretable deep learning diagnostic model for TLE and identify disease-associated markers.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Using RNA-seq and microarray data from 287 samples collected from eight GEO datasets, we constructed multiple machine learning algorithms including Deep Neural Networks (DNN), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Logistic Regression (LR), and K-Nearest Neighbors (KNN) to distinguish TLE from normal. SHapley Additive exPlanations (SHAP) and Kolmogorov-Arnold Networks (KAN) were employed to interpret the model and identify key genes associated with TLE pathogenesis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>After comparative analysis, a Deep Neural Network (DNN) model with 10 optimized genetic features achieved perfect diagnostic performance (AUC = 1.000, accuracy = 1.000). SHAP interpretation identified DEPDC5, STXBP1, GABRG2, SLC2A1, and LGI1 as the most significant TLE-associated genes. The KAN model revealed complex nonlinear relationships between these genes and TLE status, providing mathematical expressions that capture their contributions. To facilitate clinical application, we developed an online diagnostic platform that delivers interpretable predictions based on gene expression values.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>This study advances our understanding of TLE pathogenesis and provides a transparent, interpretable diagnostic model, which combines with traditional diagnostic methods may significantly improve the accuracy of TLE diagnosis, serving as a supplementary tool for clinical assessment.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2025.1655338","type":"journal-article","created":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T07:14:47Z","timestamp":1761894887000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Construction of a diagnostic model for temporal lobe epilepsy using interpretable deep learning: disease-associated markers identification"],"prefix":"10.3389","volume":"8","author":[{"given":"Tianyu","family":"Wang","sequence":"first","affiliation":[]},{"given":"Aowen","family":"Wang","sequence":"additional","affiliation":[]},{"given":"Minwei","family":"Zhu","sequence":"additional","affiliation":[]},{"given":"Wenhao","family":"Jiang","sequence":"additional","affiliation":[]},{"given":"Mingrui","family":"Li","sequence":"additional","affiliation":[]},{"given":"Shi","family":"Yan","sequence":"additional","affiliation":[]},{"given":"Yifu","family":"Shu","sequence":"additional","affiliation":[]},{"given":"Shengkun","family":"Yu","sequence":"additional","affiliation":[]},{"given":"Zhiguo","family":"Lin","sequence":"additional","affiliation":[]},{"given":"Zhibin","family":"Han","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,10,31]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"892","DOI":"10.1111\/epi.17165","article-title":"The global cost of epilepsy: a systematic review and extrapolation","volume":"63","author":"Begley","year":"2022","journal-title":"Epilepsia"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1038\/nrneurol.2010.199","article-title":"Advances in MRI for \u201ccryptogenic\u201d epilepsies","volume":"7","author":"Bernasconi","year":"2011","journal-title":"Nat. Rev. Neurol."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"2984","DOI":"10.1093\/brain\/awu259","article-title":"Glutamatergic neuron-targeted loss of LGI1 epilepsy gene results in seizures","volume":"137","author":"Boillot","year":"2014","journal-title":"Brain"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"106288","DOI":"10.1016\/j.cmpb.2021.106288","article-title":"The importance of being external. methodological insights for the external validation of machine learning models in medicine","volume":"208","author":"Cabitza","year":"2021","journal-title":"Comput. Methods Prog. Biomed."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"12","DOI":"10.3389\/fneur.2021.806506","article-title":"Epileptic phenotypes associated with SNAREs and related synaptic vesicle exocytosis machinery","volume":"12","author":"Cali","year":"2022","journal-title":"Front. Neurol."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"313","DOI":"10.1002\/ana.25655","article-title":"Randomized placebo-controlled trial of intravenous immunoglobulin in autoimmune LGI1\/CASPR2 epilepsy","volume":"87","author":"Dubey","year":"2020","journal-title":"Ann. Neurol."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"201","DOI":"10.1038\/nrn1624","article-title":"Molecular motors and mechanisms of directional transport in neurons","volume":"6","author":"Hirokawa","year":"2005","journal-title":"Nat. Rev. Neurosci."},{"key":"ref9","article-title":"Kolmogorov-Arnold Networks: A Critical Assessment of Claims, Performance, and Practical Viability","author":"Hou","year":"2024"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"12618","DOI":"10.1038\/s41598-017-12574-2","article-title":"Knockout of the epilepsy gene Depdc5 in mice causes severe embryonic dysmorphology with hyperactivity of mTORC1 signalling","volume":"7","author":"Hughes","year":"2017","journal-title":"Sci. Rep."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"464","DOI":"10.1001\/jamaneurol.2015.4996","article-title":"Evidence on use of neuroimaging for surgical treatment of temporal lobe epilepsy: a systematic review","volume":"73","author":"Jones","year":"2016","journal-title":"JAMA Neurol."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1111\/epi.13326","article-title":"Parental anxiety in childhood epilepsy: a systematic review","volume":"57","author":"Jones","year":"2016","journal-title":"Epilepsia"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"1009","DOI":"10.1001\/jamaneurol.2016.0449","article-title":"Molecular pathogenic basis for GABRG2 mutations associated with a spectrum of epilepsy syndromes, from generalized absence epilepsy to Dravet syndrome","volume":"73","author":"Kang","year":"2016","journal-title":"JAMA Neurol."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"e203","DOI":"10.1111\/epi.13222","article-title":"The role of SLC2A1 mutations in myoclonic astatic epilepsy and absence epilepsy, and the estimated frequency of GLUT1 deficiency syndrome","volume":"56","author":"Larsen","year":"2015","journal-title":"Epilepsia"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"767","DOI":"10.1186\/s12967-024-05387-1","article-title":"GABRG2 mutations in genetic epilepsy with febrile seizures plus: structure, roles, and molecular genetics","volume":"22","author":"Li","year":"2024","journal-title":"J. Transl. Med."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"106842","DOI":"10.1016\/j.nbd.2025.106842","article-title":"A missense variant in DEPDC5 resulted in abnormal morphology and increased seizure susceptibility and mortality through regulating mTOR signaling","volume":"207","author":"Liu","year":"2025","journal-title":"Neurobiol. Dis."},{"key":"ref17","author":"Liu","year":""},{"key":"ref18","author":"Liu","year":""},{"key":"ref19","doi-asserted-by":"publisher","first-page":"4768","DOI":"10.5555\/3295222.3295230","article-title":"A unified approach to interpreting model predictions","volume":"30","author":"Lundberg","year":"2017","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"1820","DOI":"10.1111\/j.1528-1167.2011.03163.x","article-title":"STXBP1-related encephalopathy presenting as infantile spasms and generalized tremor in three patients","volume":"52","author":"Mignot","year":"2011","journal-title":"Epilepsia"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1829","DOI":"10.1038\/s41467-024-46150-w","article-title":"Data leakage inflates prediction performance in connectome-based machine learning models","volume":"15","author":"Rosenblatt","year":"2024","journal-title":"Nat. Commun."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"348","DOI":"10.1136\/bmj.g254","article-title":"Drug treatment of epilepsy in adults","volume":"348","author":"Schmidt","year":"2014","journal-title":"BMJ (Online)."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1093\/brain\/aww272","article-title":"De novo GABRG2 mutations associated with epileptic encephalopathies","volume":"140","author":"Shen","year":"2017","journal-title":"Brain"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1177\/15357597251320748","article-title":"Illuminating LGI1 seizure mechanisms using novel optical techniques","volume":"25","author":"Stafstrom","year":"2025","journal-title":"Epilepsy Curr."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1212\/WNL.0000000000201612","article-title":"Erratum: natural history study of STXBP1-developmental and epileptic encephalopathy into adulthood","volume":"101","author":"Stamberger","year":"2023","journal-title":"Neurology"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"e221","DOI":"10.1212\/WNL.0000000000200715","article-title":"Natural history study of STXBP1-developmental and epileptic encephalopathy into adulthood","volume":"99","author":"Stamberger","year":"2022","journal-title":"Neurology"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1016\/j.yped.2011.12.020","article-title":"The long-term outcome of adult epilepsy surgery, patterns of seizure remission, and relapse: a cohort study","volume":"2013","author":"Stockman","year":"2013","journal-title":"Yearb. Pediatr."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1017","DOI":"10.1152\/physrev.1991.71.4.1017","article-title":"Neuropeptide hormones as neurotrophic factors","volume":"71","author":"Strand","year":"1991","journal-title":"Physiol. Rev."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"1831","DOI":"10.1093\/brain\/awn113","article-title":"Paroxysmal exercise-induced dyskinesia and epilepsy is due to mutations in SLC2A1, encoding the glucose transporter GLUT1","volume":"131","author":"Suls","year":"2008","journal-title":"Brain"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"689","DOI":"10.1016\/S0140-6736(18)32596-0","article-title":"Epilepsy in adults","volume":"393","author":"Thijs","year":"2019","journal-title":"Lancet"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1016\/j.neuron.2020.05.036","article-title":"Snareopathies: diversity in mechanisms and symptoms","volume":"107","author":"Verhage","year":"2020","journal-title":"Neuron"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1016\/j.seizure.2016.11.030","article-title":"Epilepsy-associated genes","volume":"44","author":"Wang","year":"2017","journal-title":"Seizure"},{"key":"ref33","author":"Xu","year":"2024"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1655338\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T07:14:48Z","timestamp":1761894888000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1655338\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,31]]},"references-count":33,"alternative-id":["10.3389\/frai.2025.1655338"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1655338","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,31]]},"article-number":"1655338"}}