A three-dimensional convolution method for imagined language electroencephalography topography decoding

By constructing a frequency-band specific rectangular BEAM map using a three-dimensional convolution method and designing a dual-branch network, the problem of insufficient decoding accuracy in traditional methods is solved, achieving efficient decoding of the spatial structure of EEG signals and improving classification accuracy.

CN120995220BActive Publication Date: 2025-12-23CHANGCHUN UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202511476192.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-10-16
Publication Date
2025-12-23
Estimated Expiration
2045-10-16

AI Technical Summary

Technical Problem

Existing technologies for EEG decoding of imaginary language suffer from problems related to signal level and feature representation complexity. In particular, the geometric mismatch and single-dimensional feature extraction of the traditional BEAM method lead to a decrease in decoding accuracy, making it difficult to effectively mine the spatial structural information of EEG signals.

Method used

A three-dimensional convolutional method is adopted, and a rectangular BEAM graph is constructed by frequency band-specific kriging interpolation. Combining three-dimensional continuous spatial-frequency band and spatial-temporal representations, a two-branch three-dimensional convolutional network is designed. Attention weights are used for feature fusion decoding to improve decoding accuracy.

Benefits of technology

It improves the ability to recognize language-based imagery EEG patterns and significantly enhances classification accuracy, especially in decoding multi-band and multi-time-step EEG signals.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A three-dimensional convolution method for imagined language electroencephalography (EEG) topography decoding. It solves the problem of the deficiency of the existing convolutional neural network method in modeling ability for multi-rhythm EEG topography, which leads to a significant decline in decoding accuracy in imagined language. In the aspect of rectangular BEAM reconstruction, a frequency-specific adaptive variogram kriging interpolation model is built, which automatically selects the optimal variogram function according to the spatial variation characteristics of the EEG power of each frequency band, and constructs a high-resolution BEAM map sequence under multi-frequency and multi-time steps, taking into account the spatial continuity and the adaptability of the CNN structure. A double-branch three-dimensional convolutional neural network is designed to extract high-dimensional features from spatial-time and spatial-frequency tensors, respectively, to explore the multi-scale structure and dynamic expression ability of neural activity. The discriminative ability of the two feature vector branches is dynamically weighted to improve the recognition ability of the classifier for the EEG pattern of language imagination.
Need to check novelty before this filing date? Find Prior Art