Electroencephalogram signal denoising method based on one-dimensional residual convolutional neural network

A convolutional neural network, EEG technology, applied in instruments, computing, character and pattern recognition, etc., can solve the problems of signal delay analysis, signal distortion, inability to meet EEG noise reduction requirements, etc., to improve the signal-to-noise ratio and Root mean square error, reduce denoising time, avoid the effect of gradient explosion

Active Publication Date: 2019-05-21
SHAANXI NORMAL UNIV
View PDF7 Cites 32 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These filtering methods are all offline filtering, so the signal may be distorted during the transmission process, and it will also cause signal delay...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Electroencephalogram signal denoising method based on one-dimensional residual convolutional neural network
  • Electroencephalogram signal denoising method based on one-dimensional residual convolutional neural network
  • Electroencephalogram signal denoising method based on one-dimensional residual convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] Take 20 sample data selected from the EEG collection on the physiionet website as an example, the sampling frequency is 256Hz, and the signal within the time period of 0 to 1 minute is used as input. Each sample data represents a 15360×23 matrix, each row represents the voltage signal of an electrode channel within 1 minute, and there are 23 electrode channels in total. The EEG signal denoising method based on one-dimensional residual convolutional neural network consists of the following steps (such as figure 1 shown):

[0044] (1) Select EEG samples

[0045] Select m EEG samples from the Physionet database and mark them as EEG samples E D×N , where D is the number of channels of the EEG signal, N is the number of sampling points of the EEG signal, in the EEG sample E D×N Among them, D is 23, N is 60s×256Hz sampling points, and this data is sent as input to the one-dimensional residual convolution network. Before feeding into a 1D residual convolutional network for...

Embodiment 2

[0070] Take 20 sample data selected from the EEG collection on the physiionet website as an example, the sampling frequency is 256Hz, and the signal within the time period of 0 to 1 minute is used as input. Each sample data represents a 15360×23 matrix, each row represents the voltage signal of an electrode channel within 1 minute, and there are 23 electrode channels in total. The EEG signal denoising method based on one-dimensional residual convolutional neural network consists of the following steps:

[0071] (1) Select EEG samples

[0072] This step is the same as in Example 1.

[0073] (2) Constructing noisy EEG signal samples

[0074] This step is the same as in Example 1.

[0075] (3) Divide the network training set and test set

[0076] The data enhancement method is used to expand the EEG samples and the noisy EEG signal samples, and divide them into network training set and test set respectively. The data enhancement method is as follows: all the EEG signals in t...

Embodiment 3

[0082] Take 20 sample data selected from the EEG collection on the physiionet website as an example, the sampling frequency is 256Hz, and the signal within the time period of 0 to 1 minute is used as input. Each sample data represents a 15360×23 matrix, each row represents the voltage signal of an electrode channel within 1 minute, and there are 23 electrode channels in total. The EEG signal denoising method based on one-dimensional residual convolutional neural network consists of the following steps:

[0083] (1) Select EEG samples

[0084] This step is the same as in Example 1.

[0085] (2) Constructing noisy EEG signal samples

[0086] This step is the same as in Example 1.

[0087] (3) Divide the network training set and test set

[0088]The data enhancement method is used to expand the EEG samples and the noisy EEG signal samples, and divide them into network training set and test set respectively. The data enhancement method is as follows: all the EEG signals in th...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an electroencephalogram signal denoising method based on a one-dimensional residual convolutional neural network. The electroencephalogram signal denoising method comprises thesteps of selecting an electroencephalogram sample, constructing a noisy electroencephalogram signal sample, dividing a network training set and a test set, constructing the one-dimensional residual convolutional neural network, training the one-dimensional residual convolutional neural network and reconstructing a denoised electroencephalogram signal; according to the invention, a one-dimensionalresidual convolutional neural network formed by connecting residual networks is constructed; a convolutional layer and an activation layer are introduced, so that the learning ability of a neural network is enhanced, accurate mapping and real-time denoising of noise signals to brain signals are established, neurons smaller than 0 are removed by using a linear rectification unit layer function after the convolutional layer, effective characteristics are screened out, and the defect of gradient explosion is avoided; signal de-noising is divided into a model training process and a de-noising process, the signal-to-noise ratio and the root-mean-square error of signal de-noising are improved, the de-noising time is shortened, the de-noising efficiency and quality of electroencephalogram signals are improved, and the method can be applied to the technical field of signal processing preprocessing and de-noising processing.

Description

technical field [0001] The invention belongs to the technical field of electroencephalogram signal processing, and in particular relates to an electroencephalogram signal denoising method of a one-dimensional residual convolutional neural network. technical background [0002] Electroencephalogram (Electroencephalogram, EEG) is the response of the electrical activity in the brain nerve cells on the cerebral cortex to some electrodes placed on the scalp in multiple areas of the brain, but EEG is a highly random nonlinear non-stationary The signal contains very complex components, and the signal amplitude is microvolt level, the strength is very weak, and it is easily interfered by other physiological signals of the human body or non-physiological signals such as space electromagnetic noise. It hinders the follow-up research analysis and application of EEG signals. Therefore, it is of great theoretical and practical significance to develop related methods to remove the artifa...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/00G06K9/62
Inventor 吴晓军孙维彤张玉梅路纲
Owner SHAANXI NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap