A brain and myoelectric signal decoding method based on time-frequency fusion and double-branch network

By combining time-frequency fusion and dual-branch network methods with global and local features, the problem of information loss in EEG and sEMG signal fusion is solved, achieving more efficient motion classification and auxiliary control for rehabilitation of neurological diseases.

CN119760529BActive Publication Date: 2025-11-28BEIJING SONGGUO BRAIN MACHINE TECHNOLOGY CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202411724583.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-11-28
Publication Date
2025-11-28
Estimated Expiration
2044-11-28

AI Technical Summary

Technical Problem

Existing technologies fail to fully consider the characteristics of time-frequency maps in EEG and sEMG signal fusion, resulting in the failure to extract more effective frequency bands and the discarding of a large amount of global information during the frequency band selection process, thus limiting the comprehensiveness of the fusion effect.

Method used

We adopt a time-frequency fusion and dual-branch network approach to capture global information by stacking time-frequency maps and extract local information by introducing Rayleigh entropy. We design a dual-stream hybrid residual network and combine attention mechanism and ConvLSTM model to optimize feature representation.

Benefits of technology

It improves the effectiveness of feature extraction and the comprehensiveness of information retention of EEG and sEMG signals, enhances the accuracy and stability of action classification, and is suitable for rehabilitation and auxiliary control of neurological diseases.

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

Abstract

The application claims a brain and muscle electrical signal decoding method based on time-frequency fusion and double-branch network, comprising the following steps: designing a brain and muscle electrical synchronous experiment paradigm; carrying out denoising and normalization processing on the collected brain and muscle electrical signals; using wavelet coherence analysis to screen the electroencephalogram channel and using short-time Fourier transform to obtain time-frequency characteristics; fusing in the following two ways: one is to stack by channel to obtain global characteristics, and the other is to use the Renyi entropy to obtain the significant feature band of each channel and splice into a new time-frequency graph; designing a double-branch network to extract features, using a time-domain residual module and a channel attention module to extract the features of the spliced time-frequency graph, using a 3D residual module and a ConvLSTM module to capture the dynamic changes and space-time characteristics of the stacked time-frequency graph; finally, splicing and fusing the features extracted by the two branches, using a classification module to calculate errors and classify decoding, effectively improving the accuracy and stability of the brain and muscle electrical fusion decoding.
Need to check novelty before this filing date? Find Prior Art