EEG (electroencephalogram) signal unsupervised feature learning method based on convolutional network and self encoding

A convolutional network and feature learning technology, applied in the field of convolutional networks, can solve problems such as difficulty in feature design and selection, new patients are unknown, etc.

Active Publication Date: 2018-07-06
XIAMEN UNIV
View PDF7 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Although experienced experts can identify epilepsy EEG and scholars have done a lot of research on the problem of epilepsy detection based on EEG, there are still many challenges in the automatic detection of epilepsy.
Difficulties in feature design and selection, and applicability of these approaches to new patients remains unknown

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
  • EEG (electroencephalogram) signal unsupervised feature learning method based on convolutional network and self encoding
  • EEG (electroencephalogram) signal unsupervised feature learning method based on convolutional network and self encoding
  • EEG (electroencephalogram) signal unsupervised feature learning method based on convolutional network and self encoding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] The following embodiments will further illustrate the present invention in conjunction with the accompanying drawings

[0051] Embodiments of the present invention include the following steps:

[0052] 1) The value of each dimension of the required EEG data sample is controlled between (0,1). Through 0-1 standardization, the sample data is linearly transformed, and the data is mapped to [0,1]. The conversion function is as follows:

[0053]

[0054] Where d is the value of one dimension of the input sample x, max(data) is the maximum value of each dimension in all samples, and min(data) is the minimum value of each dimension in all samples.

[0055] 2) Construct AE-CDNN model based on deep convolutional network and self-encoder. The framework structure of the deep convolutional network model (AE-CDNN) is shown in figure 1 . The model is mainly divided into two stages. The encoding stage operation includes sample input, convolutional layer, pooling layer (downsamp...

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 relates to an EEG (electroencephalogram) signal unsupervised feature learning method based on a convolutional network and self encoding and belongs to the convolutional network field. The method includes the following steps that: predictive processing is performed on the data of EEG data samples, the data are mapped into an interval of [0,1] through 0-1 standardization, namely, linear transformation; an AE-CDNN model is constructed on the basis of a deep convolutional network and self-coding, training data are used to train the AE-CDNN model; and the dimensionality of the trainedAE-CDNN model is decreased to a low dimensionality through using test data, and therefore, the classification of the EEG data can be benefitted. The AE-CDNN model is constructed on the basis of the deep convolutional network and self-coding and is adopted to perform unsupervised feature learning of epilepsy EEG signals; feature learning and classification are performed on two public epilepsy EEGdata sets; and the AE-CDNN model can be well applied to the extraction of epilepsy brain signal features.

Description

technical field [0001] The invention relates to a convolutional network, in particular to an unsupervised feature learning method for EEG signals based on a convolutional network and self-encoding. Background technique [0002] Epilepsy is a non-infectious chronic brain disease that affects people of all ages. There are currently about 50 million epilepsy patients worldwide. It has become one of the most common neurological diseases worldwide [1] . Epilepsy causes cognitive dysfunction (such as loss of consciousness, consciousness, etc.) to the patient, which will bring great physical harm to the patient (such as fracture and injury during the seizure), and it will also bring great psychological harm to the patient due to humiliation and discrimination. pain of. Because epileptic seizures can cause irreversible damage to the brain of patients and have the characteristics of unprovoked repeated seizures, how to accurately analyze and prevent epilepsy is of great significanc...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04G06N3/08G06K9/00
CPCG06N3/088G06N3/045G06F2218/08G06F2218/12
Inventor 张仲楠温廷羲
Owner XIAMEN 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