Deep Convolutional Neural Network based on Acceleration Data for the Detection of Generalized Tonic-Clonic Seizures

Hyosung Joo1, Jihwan Woo1

  • 1University of Ulsan

Details

12:00 - 13:45 | Mon 6 Nov | Auditorium Foyer, E1/E2, Upper Atrium Space | MLunch_Break.28

Session: Lunch, Posters and POC Technologies Demonstrations – Session I

Abstract

Monitoring devices for detecting seizures are very helpful for early intervention in daily life. Several studies have suggested methods for detecting generalized tonic-clonic seizures (GTCs) using an accelerometer placed on the patient's wrist. However, these studies need to be improved in terms of accuracy. In this study, we employed the convolutional neural network (CNN) for detecting the GTCs based on a time-frequency spectrogram of the acceleration data. The result shows that the CNN-based GTCs detection approach could achieve more accurate detection compared to the previous detection methods.