English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Background: Epilepsy is a neurological illness affecting the brain that makespeople more likely to experience frequent, spontaneous seizures. There has tobe an accurate automated method for measuring seizure frequency and severity inorder to assess the efficacy of pharmacological therapy for epilepsy. The drugquantities are often derived from patient reports which may cause significantissues owing to inadequate or inaccurate descriptions of seizures and theirfrequencies. Methods and materials: This study proposes a novel deep learningarchitecture based lightweight convolution transformer (LCT). The transformeris able to learn spatial and temporal correlated information simultaneouslyfrom the multi-channel electroencephalogram (EEG) signal to detect seizures atsmaller segment lengths. In the proposed model, the lack of translationequivariance and localization of ViT is reduced using convolution tokenization,and rich information from the transformer encoder is extracted by sequencepooling instead of the learnable class token. Results: Extensive experimentalresults demonstrate that the proposed model of cross-patient learning caneffectively detect seizures from the raw EEG signals. The accuracy and F1-scoreof seizure detection in the cross-patient case on the CHB-MIT dataset are shownto be 96.31% and 96.32%, respectively, at 0.5 sec segment length. In addition,the performance metrics show that the inclusion of inductive biases andattention-based pooling in the model enhances the performance and reduces thenumber of transformer encoder layers, which significantly reduces thecomputational complexity. In this research work, we provided a novel approachto enhance efficiency and simplify the architecture for multi-channel automatedseizure detection.

Lightweight Convolution Transformer for Cross-patient Seizure Detection in Multi-channel EEG Signals