Multimodal detection of epilepsy with deep neural networks
作者：
Highlights：
• We employ pretrained models, i.e., VGG16, EfficientNet, etc. for detecting epilepsy.
• We introduce a multimodal deep neural network with a gated multimodal unit.
• We run a series of ablation experiments to show the robustness of the architecture.
• We study five cases for classification using the dataset of the University of Bonn.
• Our proposed architecture obtains comparable performance to existing research works.
摘要
•We employ pretrained models, i.e., VGG16, EfficientNet, etc. for detecting epilepsy.•We introduce a multimodal deep neural network with a gated multimodal unit.•We run a series of ablation experiments to show the robustness of the architecture.•We study five cases for classification using the dataset of the University of Bonn.•Our proposed architecture obtains comparable performance to existing research works.
论文关键词：Epilepsy,Deep learning,Single-channel EEG,Short-time fourier transform,Gated multimodal unit
论文评审过程：Received 27 January 2022, Revised 1 October 2022, Accepted 9 October 2022, Available online 14 October 2022, Version of Record 21 October 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.119010
