Local field potential signal generation method and system based on improved diffusion model

By combining the Transformer encoder and GRU diffusion model, the fidelity and stability problems of local field potential signal generation in the prior art are solved, and high-fidelity, conditionally controlled LFP signal generation is realized, which is suitable for multi-channel synchronization and behavior-related neural signal simulation.

CN120492795APending Publication Date: 2025-08-15SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
0 Cites 2 Cited by

Patent Information

Application Number
CN202510427310.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-07
Publication Date
2025-08-15

AI Technical Summary

Technical Problem

When generating local field potential signals, the prior art has problems with low fidelity and stability, making it difficult to take into account the global timing mode and local dynamic details at the same time, as well as the lack of support for behavioral conditions and multi-channel data.

Method used

A diffusion model based on Transformer encoder and GRU is adopted, and a high-fidelity, conditional controllable LFP time series is generated through multi-head self-attention calculation, latent variable sampling, continuous ODE solution and discrete GRU rule update, combined with forward noise addition and reverse denoising sampling of the diffusion model.

Benefits of technology

The fidelity and physiological correlation of the generated signals are improved, making the synthetic LFP closer to real records in terms of statistical characteristics, spectral characteristics and behavior-related changes, and achieving high fidelity, controllable conditions and stable structural neural signal generation.

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

Abstract

The invention discloses a local field potential signal generation method and system based on an improved diffusion model. The method comprises the following steps: acquiring and preprocessing an original noise-containing signal, timestamp information, a multi-channel behavior signal or a stimulation signal; performing multi-head self-attention calculation on the preprocessed data based on a Transform encoder to obtain a potential state of the LFP sequence in a potential space; carrying out potential variable sampling according to the potential state, carrying out continuous ODE solution and discrete GRU rule updating on the obtained initial potential variable, and generating a potential state sequence evolved along with time; performing forward noise addition and reverse noise reduction sampling on the potential state sequence based on a diffusion model to obtain a final potential sequence; and mapping a potential sequence finally obtained by the diffusion model back to an LFP signal of a preset dimension, and outputting LFP time sequence data generated by simulation. According to the invention, high-fidelity, controllable-condition and stable-structure neural signal generation is realized.
Need to check novelty before this filing date? Find Prior Art