Home // SIGNAL 2025, The Tenth International Conference on Advances in Signal, Image and Video Processing // View article

A New 1D-CNN Paradigm for Onset Detection of Absence Seizures in Children

Authors:
Maxime Yochum
Amar Kachenoura
Matthieu Aud’hui
Fabrice Wendling
Anna Kaminska
Rima Nabbout
Mathieu Kuchenbuch
Pascal Benquet

Keywords: Surface EEG; Childhood Absence Epilepsy; Onset Detection; 1D-CNN.

Abstract:
This study presents a One-Dimensional Convolutional Neural Network (1D-CNN)-based algorithm for the early detection of childhood absence seizures in ElectroEncephaloGraphy (EEG) traces. This detection aims to enable timely sensory interventions, such as acoustic or visual stimulation, to potentially abort seizures. The algorithm was evaluated using a clinical setting with full EEG data and a reduced number of electrodes version of the data to show its suitability in a normal child environment. On the clinical EEG database of 117 patients, the model achieved promising results, including a Sensitivity of 0.859, Precision of 0.819, F1-score of 0.837, and a mean detection delay of 0.522 seconds. The performance remained satisfactory when using fewer electrodes, with a Sensitivity of 0.837, Precision of 0.808, F1-score of 0.820, and similar detection delays. These results demonstrate the method’s robustness and feasibility for clinical applications, as well as its potential to be embedded in wearable devices for continuous, real-time seizure monitoring and intervention in children with absence epilepsy.

Pages: 3 to 7

Copyright: Copyright (c) IARIA, 2025

Publication date: March 9, 2025

Published in: conference

ISSN: 2519-8432

ISBN: 978-1-68558-245-6

Location: Lisbon, Portugal

Dates: from March 9, 2025 to March 13, 2025