{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:53:48Z","timestamp":1791266028490,"version":"4.1.0"},"reference-count":27,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T00:00:00Z","timestamp":1791244800000},"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. Comput. Neurosci."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Electroencephalogram (EEG) signal classification plays a pivotal role in brain-computer interfaces (BCIs) and the diagnosis of neurological disorders. However, the non-stationary, non-linear, and low signal-to-noise ratio characteristics of EEG signals make accurate feature extraction challenging. Conventional deep learning approaches often struggle to simultaneously capture localized transient patterns, deep temporal dependencies, and long-range contextual relationships within EEG signals.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>To address these challenges, an optimized end-to-end hybrid deep learning framework, termed CNN\u2013BiLSTM\u2013MHSA, is proposed for multiclass epileptic seizure recognition. The architecture sequentially integrates a multilayer one-dimensional Convolutional Neural Network (CNN) for localized feature extraction, a Bidirectional Long Short-Term Memory (BiLSTM) network for capturing forward and backward temporal dependencies, and a Multi-Head Self-Attention (MHSA) mechanism for dynamically emphasizing discriminative neurological events. Robust preprocessing includes Butterworth bandpass filtering, Z-score normalization, and data augmentation using Gaussian noise injection and amplitude scaling. The framework was evaluated on the UCI\/Kaggle Epileptic Seizure Recognition dataset comprising 11,500 EEG segments distributed across five classes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The proposed framework achieved an overall accuracy of 98.83%, precision of 98.95%, recall (sensitivity) of 98.84%, F1-score of 98.90%, specificity of 99.71%, balanced accuracy of 98.84%, ROC-AUC of 99.80%, and PR-AUC of 99.72%. Class-wise accuracies exceeded 99% for most categories, with one seizure class achieving perfect recognition. Comparative evaluation further demonstrated that the proposed framework outperformed several recent state-of-the-art EEG seizure classification methods.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The superior performance can be attributed to the complementary representation capabilities of the hybrid architecture. CNN layers effectively learn localized signal patterns, BiLSTM captures bidirectional temporal dependencies, and MHSA enhances the representation of long-range contextual relationships by assigning greater importance to informative temporal regions. The combination of robust preprocessing and data augmentation further contributes to improved generalization and classification reliability.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fncom.2026.1949040","type":"journal-article","created":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:21:33Z","timestamp":1791264093000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["An attention-enhanced hybrid CNN-BiLSTM framework for multiclass epileptic seizure recognition using EEG signals"],"prefix":"10.3389","volume":"20","author":[{"given":"Bobbinpreet","family":"Kaur","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, University Institute of Engineering (UIE) Chandigarh University","place":["Gharuan, Punjab, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sheenam","family":"Middha","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, University Institute of Engineering (UIE) Chandigarh University","place":["Gharuan, Punjab, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aleem","family":"Ali","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, University Institute of Engineering (UIE) Chandigarh University","place":["Gharuan, Punjab, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Noman","family":"Khan","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Sejong University","place":["Seoul, Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Reham","family":"Alhejaili","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Artificial Intelligence, College of Computer Science and Engineering, University of Jeddah","place":["Jeddah, Saudi Arabia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ahmed H.","family":"Almutairi","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Applied College, Northern Border University","place":["Arar, Saudi Arabia"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,10,6]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"596","DOI":"10.1007\/s44163-026-01340-x","article-title":"Low-latency six-channel Bi-LSTM for near-real-time epileptic seizure detection","volume":"6","author":"Afsari","year":"2026","journal-title":"Discov. Artif. Intell"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1680395","DOI":"10.3389\/fnhum.2025.1680395","article-title":"Using vision transformers for electrographic seizure classification to aid physician review of intracranial electroencephalography recordings","volume":"19","author":"Afzal","year":"2025","journal-title":"Front. Hum. Neurosci"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2502.15198","article-title":"Graph-based deep learning on stereo EEG for predicting seizure freedom in epilepsy patients","author":"Agaronyan","year":"2025","journal-title":"arXiv"},{"key":"B4","doi-asserted-by":"publisher","first-page":"104780","DOI":"10.1016\/j.bspc.2023.104780","article-title":"MP-SeizNet: A multi-path CNN Bi-LSTM network for seizure-type classification using EEG","volume":"84","author":"Albaqami","year":"2023","journal-title":"Biomed. Signal Process. Control"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1668358","DOI":"10.3389\/fncom.2025.1668358","article-title":"Advancing epileptic seizure recognition through bidirectional LSTM networks","volume":"19","author":"Al-Marzouki","year":"2025","journal-title":"Front. Comput. Neurosci"},{"key":"B6","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1016\/j.neuroscience.2019.10.034","article-title":"K-complexes detection in EEG signals using fractal and frequency features coupled with an ensemble classification model","volume":"422","author":"Al-Salman","year":"2019","journal-title":"Neuroscience"},{"key":"B7","doi-asserted-by":"publisher","first-page":"402","DOI":"10.1109\/ICESIC67389.2026.11496587","article-title":"\u201cAn attention-guided CNN-BiGRU framework for multi-class epileptic seizure classification using CEEMDAN-decomposed EEG signals,\u201d","author":"Arshath","year":"2026","journal-title":"2026 International conference on electronic systems and intelligent computing (ICESIC)"},{"key":"B8","doi-asserted-by":"publisher","first-page":"99","DOI":"10.3390\/e19030099","article-title":"Tunable-Q wavelet transform based multivariate sub-band fuzzy entropy with application to focal EEG signal analysis","volume":"19","author":"Bhattacharyya","year":"2017","journal-title":"Entropy"},{"key":"B9","doi-asserted-by":"publisher","first-page":"016018","DOI":"10.1088\/1741-2552\/ae33f8","article-title":"A multi-view neural framework with attention for epileptic seizure classification","volume":"23","author":"Feng","year":"2026","journal-title":"J. Neural Eng"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2407.20254","article-title":"EEGMamba: Bidirectional state space model with mixture of experts for EEG multi-task classification","author":"Gui","year":"2024","journal-title":"arXiv"},{"key":"B11","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1186\/s40708-026-00319-9","article-title":"Adaptive frequency band attention-guided CNN-BiLSTM for spatial-spectral-temporal EEG emotion recognition","volume":"13","author":"Hasan","year":"2026","journal-title":"Brain Inf"},{"key":"B12","doi-asserted-by":"publisher","first-page":"9579422","DOI":"10.1155\/2022\/9579422","article-title":"Epileptic seizure detection using a hybrid 1D CNN-machine learning approach from EEG data","volume":"2022","author":"Hassan","year":"2022","journal-title":"J. Healthc. Eng"},{"key":"B13","doi-asserted-by":"publisher","first-page":"030039","DOI":"10.1063\/5.0247920","article-title":"Automated epileptic seizure detection using convolutional neural networks on EEG signals","volume":"3237","author":"Kanamaneni","year":"2025","journal-title":"AIP Conf. Proc"},{"key":"B14","doi-asserted-by":"crossref","DOI":"10.52202\/085713-4837","article-title":"\u201cEvoBrain: dynamic multi-channel EEG graph modeling for time-evolving brain networks,\u201d","volume-title":"NeurIPS 2025 (Spotlight)","author":"Kotoge","year":"2025"},{"key":"B15","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1186\/s44147-023-00353-y","article-title":"Detection of epileptic seizure in EEG signals using machine learning and deep learning techniques","volume":"71","author":"Kunekar","year":"2024","journal-title":"J. Eng. Appl. Sci"},{"key":"B16","doi-asserted-by":"publisher","first-page":"109675","DOI":"10.1016\/j.bspc.2026.109675","article-title":"Improved detection of epileptic seizures via EEG signals and texture analysis of recurrence plots","volume":"118","author":"Lashkari","year":"2026","journal-title":"Biomed. Signal Process. Control"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1439607","DOI":"10.3389\/fphys.2024.1439607","article-title":"Multi-branch fusion graph neural network based on multi-head attention for childhood seizure detection","volume":"15","author":"Li","year":"2024","journal-title":"Front. Physiol"},{"key":"B18","doi-asserted-by":"publisher","first-page":"2750001","DOI":"10.1142\/S0129065727500018","article-title":"Unified multi-class electroencephalogram artifact recognition using machine learning classifiers","volume":"36","author":"Martins","year":"2026","journal-title":"Int. J. Neural Syst"},{"key":"B19","doi-asserted-by":"publisher","first-page":"781","DOI":"10.3390\/bioengineering9120781","article-title":"Supervised machine learning and deep learning techniques for epileptic seizure recognition using EEG signals-a systematic literature review","volume":"9","author":"Nafea","year":"2022","journal-title":"Bioengineering"},{"key":"B20","unstructured":"Classification of epileptic seizure dataset using different machine learning algorithms and PCA feature reduction technique\n          \n          47\n          60\n          \n            \n              Nahzat\n              S.\n            \n            \n              Yaganoglu\n              M.\n            \n          \n          J. Investig. Eng. Technol\n          4\n          2021"},{"key":"B21","doi-asserted-by":"publisher","first-page":"848","DOI":"10.3390\/brainsci14080848","article-title":"Detection of anxiety-based epileptic seizures in EEG signals using fuzzy features and parrot optimization-tuned LSTM","volume":"14","author":"Palanisamy","year":"2024","journal-title":"Brain Sci"},{"key":"B22","doi-asserted-by":"publisher","first-page":"202","DOI":"10.1016\/j.neunet.2020.01.017","article-title":"EEG based multi-class seizure type classification using convolutional neural network and transfer learning","volume":"124","author":"Raghu","year":"2020","journal-title":"Neural Netw"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1577","DOI":"10.1007\/s11760-023-02871-x","article-title":"Epileptic seizure detection using scalogram-based hybrid CNN model on EEG signals","volume":"18","author":"Sadam","year":"2024","journal-title":"Signal Image Video Process"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/EMBC58623.2025.11252697","article-title":"FEMBA: efficient and scalable EEG analysis with a bidirectional mamba foundation model","volume":"2025","author":"Tegon","year":"2025","journal-title":"Camera-ready version, EMBC"},{"key":"B25","doi-asserted-by":"publisher","first-page":"22615","DOI":"10.1038\/s41598-026-51418-w","article-title":"A novel lexicon dictionary and CNN-LSTM employed hybrid approach for sentiment detection of Covid-19 vaccines","volume":"16","author":"Tiwari","year":"2026","journal-title":"Sci. Rep"},{"key":"B26","doi-asserted-by":"publisher","first-page":"13441","DOI":"10.1038\/s41598-026-41636-7","article-title":"A time-frequency cross-attention network model for epileptic seizure detection","volume":"16","author":"Wang","year":"2026","journal-title":"Sci. Rep"},{"key":"B27","doi-asserted-by":"publisher","first-page":"2550080","DOI":"10.1142\/S0129065725500807","article-title":"Lightweight seizure prediction model based on kernel-enhanced global temporal attention","volume":"36","author":"Zhai","year":"2026","journal-title":"Int. J. Neural Syst"}],"container-title":["Frontiers in Computational Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2026.1949040\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:21:34Z","timestamp":1791264094000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2026.1949040\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,6]]},"references-count":27,"alternative-id":["10.3389\/fncom.2026.1949040"],"URL":"https:\/\/doi.org\/10.3389\/fncom.2026.1949040","relation":{},"ISSN":["1662-5188"],"issn-type":[{"value":"1662-5188","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,6]]},"article-number":"1949040"}}