Epilepsy detection system based on information bottleneck of dual-modal EEG signals

Through the dual-modal EEG signal information bottleneck system, the one-dimensional and two-dimensional convolutional neural networks and two-way long and short-term memory networks are used to integrate time series and spectrum features, and the problem of low detection accuracy in the existing technology is solved, and the problem of low detection accuracy is realized. Higher accuracy in epilepsy EEG signal recognition and diagnostic auxiliary effect.

CN115363531BActive Publication Date: 2024-12-10SHANDONG NORMAL UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211006502.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-22
Publication Date
2024-12-10
Estimated Expiration
2042-08-22

AI Technical Summary

Technical Problem

The existing machine learning-based epilepsy detection scheme fails to effectively consider time dependence and bimodal information, resulting in low detection accuracy and single-module technology fails to fully utilize the combined advantages of time series and two-dimensional spectrum.

Method used

The dual-modal EEG signal information bottleneck system is adopted to extract features through one-dimensional and two-dimensional convolutional neural networks, and the two-way long and short-term memory network modeling time relationships are used to combine the dual-modal information bottleneck network fusion characteristics to realize the epilepsy EEG signal. Accurate identification.

Benefits of technology

It improves the accuracy of epilepsy EEG signal recognition, assists doctors in diagnosis, eliminates irrelevant features, and combines the advantages of time series and spectrum, which significantly improves the accuracy of detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115363531B_ABST
    Figure CN115363531B_ABST
Patent Text Reader

Abstract

The present invention discloses an epilepsy detection system based on a dual-modal EEG signal information bottleneck, including an EEG signal acquisition module, a preprocessing module, a first extraction module, a second extraction module, a fusion module and a detection module. Based on the first and second extraction modules, one-dimensional and two-dimensional convolutional neural networks are used to extract features of EEG signals, and a bidirectional long short-term memory network is used to model the time relationship of the extracted features, and output one-dimensional and two-dimensional features of EEG signals; based on the fusion module, the one-dimensional and two-dimensional features of EEG signals are fused using a trained dual-modal information bottleneck network, and the fused features are output, and then the epilepsy detection results are output through the detection module. The present invention extracts EEG signal features in two modes, time series and spectrogram dimensions, and uses the dual-modal information bottleneck to obtain and share the most relevant information for epilepsy detection from each mode, thereby improving the accuracy of epilepsy EEG signal recognition and assisting doctors in diagnosing epilepsy.
Need to check novelty before this filing date? Find Prior Art