Deep domain adaptation-based seizure detection method

Wenlong Qiu

doi:10.1117/12.2685889

28 July 2023 Deep domain adaptation-based seizure detection method

Wenlong Qiu

Proceedings Volume 12756, 3rd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2023); 127563H (2023) https://doi.org/10.1117/12.2685889
Event: 2023 3rd International Conference on Applied Mathematics, Modelling and Intelligent Computing (CAMMIC 2023), 2023, Tangshan, China

Abstract

Electroencephalogram (EEG) seizure examination is an important part of epilepsy diagnosis and evaluation. In this paper, a Deep domain adaptation-based seizure detection method is proposed for the automatic detection of epilepsy by transfer learning. First, a Butterworth bandpass filter is used for denoising and the EEG signals in the frequency of 0.5-50Hz are extracted. The filtered EEG signals were then sliced in 4-second segments. In the second step, the EEG signal is Pretreatment processed, and the two-dimensional spectrogram signal is obtained by short-time Fourier variation in a segment every 4 seconds at the same time. Scalp EEG at CHB-MIT was tested for a total of 915.34h, including 29.06 minutes of seizure data. The average sensitivity and specificity were 96.77% and 96.48%, respectively.

Citation Download Citation

Wenlong Qiu "Deep domain adaptation-based seizure detection method", Proc. SPIE 12756, 3rd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2023), 127563H (28 July 2023); https://doi.org/10.1117/12.2685889

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available