EUVIMEDObjevte EUVIMED
Open access · CC BY 4.0

An integrated multi-objective deep learning model for EEG epilepsy detection and lobe-level localization

R. Manjupriya, A. Anny Leema

Array · 2026-12 · 32 · 101240

DOI: 10.1016/j.array.2026.101240 ↗

At a glance

Journal article in Array, 2026-12. Written by R. Manjupriya and 1 co-authors. 50 references. The full text is freely available (open access, CC BY 4.0).

AbstraktBibliografické údajeCiteRelated studies

Abstract (original language)

Epilepsy is a recurrent neurological condition defined by abnormal brain activity that requires fast and accurate seizure detection for optimal treatment using electroencephalography (EEG), the standard diagnostic tool. While recent research has enhanced EEG-based seizure detection, most approaches rely on generic classification rather than tackling a significant clinical challenge: accurate localization of seizure-generating brain areas. This limits their usefulness in personalized treatment plans and surgical planning. To address this, we present a novel framework that combines an Optuna-tuned Wave-Gated Recurrent Unit (OW-GRU) with a Bell Elliptic Fuzzy Logic System (BE-FLS) for temporal EEG pattern recognition and spatial localization. It leverages Stochastic Variance Gradient-based Honey Badger Optimization (SVGL-HBO) to select features efficiently. The proposed strategy provides strong performance on benchmark EEG datasets by tackling problems such as noise contamination, imbalanced data, and weakly periodic seizure patterns, with 98.1% detection accuracy, 99.2% precision,98.5% sensitivity, and 98.9% localization accuracy. These findings show that the framework has the potential to enable real-time clinical decision-making, shorten diagnostic delays, and improve treatment planning for people with epilepsy.

Cite

APA

Manjupriya, R., & Leema, A. A. (2026). An integrated multi-objective deep learning model for EEG epilepsy detection and lobe-level localization. Array, 32, 101240. https://doi.org/10.1016/j.array.2026.101240

Vancouver

Manjupriya R, Leema AA. An integrated multi-objective deep learning model for EEG epilepsy detection and lobe-level localization. Array. 2026;32:101240. doi:10.1016/j.array.2026.101240

BibTeX
@article{Manjupriya2026,
  author = {Manjupriya, R. and Leema, A. Anny},
  title = {An integrated multi-objective deep learning model for EEG epilepsy detection and lobe-level localization},
  journal = {Array},
  year = {2026},
  volume = {32},
  pages = {101240},
  doi = {10.1016/j.array.2026.101240}
}
Start for free →