Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{201929,
        author = {Amisha Dewangan and Nidhi Sharma Chandel and Dolly Verma},
        title = {A Survey on Deep Learning Ensemble Strategies for Multi-Class EEG Seizure and Brain Rhythm Classification},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {9510-9518},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=201929},
        abstract = {The detection of seizures and the classification of brain rhythm in the electroencephalography (EEG) system have drawn a lot of attention as it is a critical area in the diagnosis and follow up of neurological patients. EEG signals are very complex, nonlinear, and non-stationary, which complicates the task of effective multi-classification to a great extent. The latest advancements of deep learning and ensemble techniques have been demonstrated as exhibiting remarkable improvements in robustness and generalization as well as predictive accuracy. This survey provides a nice summary of publicly available EEG data, preprocessing techniques, feature engineering techniques, deep neural networks, and ensemble learning techniques applied to identify multi-class seizures and rhythms. Comparative analyses also indicate that the hybrid and the attention-based ensemble models outperform the standalone models. In addition, the primary challenges, which involve the imbalance of classes, inter-patient variability, complexity of computation, and limitations of real-time deployment, are also tackled. Finally, the research directions that are characterized by lightweight edge model, federated learning, and transformer-based ensembles are presented to help create trustful and clinically feasible EEG-based smart diagnostic systems.},
        keywords = {EEG Classification; Seizure Detection; Deep Learning; Ensemble Learning; Multi-Class Classification; Brain Rhythms},
        month = {May},
        }