Digital electroencephalograph

By employing a feature extraction method with a uniform time window and sliding step size in invasive brain-computer interfaces, combined with independent LSTM networks and dynamic normalization processing, the temporal inconsistency and non-stationarity issues in multimodal neural signal decoding are resolved, achieving high-precision multi-task behavior prediction and robust decoding.

CN121943344APending Publication Date: 2026-05-01SHANGHAI STAIRMED TECHNOLOGY CO LTD
0 Cites 2 Cited by

Patent Information

Application Number
CN202512039796.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-30
Publication Date
2026-05-01

AI Technical Summary

Technical Problem

Existing invasive brain-computer interfaces lack a unified time frame, cross-modal normalization, and robust real-time fusion strategies for decoding multimodal neural signals, resulting in limited decoding accuracy and robustness. Furthermore, deep learning models are limited in flexibility and performance in multi-task learning.

Method used

We employ a unified time window and sliding step size to extract features from LFP and pulse signals, construct a fusion feature tensor with complementary information, and decode it through an independent Long Short-Term Memory (LSTM) network. Combined with dynamic normalization and an independent optimizer, we achieve multi-task behavior prediction.

Benefits of technology

It achieves efficient fusion and stable decoding of multimodal neural signals, improves the robustness and flexibility of decoding, adapts to the personalized training needs of different tasks, reduces model maintenance costs, and provides online adaptive learning capabilities.

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

Abstract

The invention discloses a digital electroencephalograph, and belongs to the technical field of brain-computer interface signal processing. An electroencephalogram signal processor of the digital electroencephalograph acquires digital neural signals of a plurality of channels for a target object, wherein the digital neural signals comprise LFP signals and pulse signals; performing feature extraction on the two signals based on a unified time window and a sliding step length, and integrating a time-aligned feature matrix into a first fusion feature tensor; performing dynamic normalization processing on the first fusion feature tensor, wherein the statistical magnitude is updated along with the change of the input data; and n parallel LSTM networks are used to decode the normalized second fusion feature tensor to obtain n decoding results of different dimensions, the n LSTM networks are mutually independent, and each LSTM network has an independent optimizer and an independent parameter. According to the method, multi-modal features can be aligned in time, signal drift is compensated, decoding robustness is improved, and inter-task optimization decoupling and personalized training are achieved through independent parameters and an optimizer.
Need to check novelty before this filing date? Find Prior Art