Application of U-Net Neural Network for Automatic Spike Detection in EEG Recordings From Patients with Rolandic Epilepsy
The goal of the proposed study is to implement and test AI approaches to segmentation of benign epilepsy with centro-temporal spikes (BECTS) patterns in electroencephalo-gram (EEG) recordings. In order to do this we apply U-net-type deep neural network previously developed for electrocardiogram (ECG) segmentation. Our results shows potential applicability of this approach and can be used as a basis for further research in the field.
Authors
- Svetlana A. Gerasimova (ORCID: https://orcid.org/0009-0001-9377-0586)
- O. D. Elshina (ORCID: https://orcid.org/0000-0001-8280-6814)
- Tatiana A. Levanova (ORCID: https://orcid.org/0000-0003-2034-7346)
- Artem Sharkov (ORCID: https://orcid.org/0000-0002-0980-2638)
- Albina Lebedeva (ORCID: https://orcid.org/0000-0002-9194-4896)
- Nikolay Gromov (ORCID: https://orcid.org/0000-0002-9124-5526)
- Lev A. Smirnov (ORCID: https://orcid.org/0000-0002-2293-6534)
- Margarita I. Samburova (ORCID: https://orcid.org/0009-0001-9499-9403)
- Maria I. Yashanova
- V.Yu. Borisov (ORCID: https://orcid.org/0009-0002-6377-2756)
- A.M. Borisova (ORCID: https://orcid.org/0009-0007-3166-4060)
Institutions
- Privolzhsky Research Medical University (RU)
- N. I. Lobachevsky State University of Nizhny Novgorod (RU)
Publication Details
- Published
- 2024-09-19
- DOI
- https://doi.org/10.1109/cnn63506.2024.10705829
- Citations
- 1
- Primary Topic
- EEG and Brain-Computer Interfaces
- Type
- article
- Field-Weighted Citation Impact
- 0.34