Unlock instant, AI-driven research and patent intelligence for your innovation.

Epilepsy detection system based on CNN and Transform

A detection system and epilepsy technology, applied in the field of epilepsy detection systems based on CNN and Transformer, can solve problems such as single model, ignoring local features, and interference of various indicators of experimental accuracy, so as to improve accuracy, eliminate feature misalignment, and improve The effect of global awareness

Pending Publication Date: 2022-03-04
SHANDONG NORMAL UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, most of the EEG samples are manually observed by professional physicians, and the final conclusion is low, so the automatic recognition of EEG has become the main line of development in the medical field.
[0004] Most traditional epilepsy detection methods and deep learning methods only use a single model, for example, a single CNN model. The convolution operation is good at extracting local features, but it still has certain limitations in capturing global feature representations.
In a single Transformer, the self-attention module can capture long-distance feature dependencies, but ignores the details of local features
These single models will cause some interference to the experimental accuracy and various indicators

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
  • Epilepsy detection system based on CNN and Transform
  • Epilepsy detection system based on CNN and Transform
  • Epilepsy detection system based on CNN and Transform

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Such as figure 1 As shown, a CNN and Transformer-based epilepsy detection system includes:

[0050] An acquisition module configured to: acquire multi-channel EEG signals to be detected;

[0051] A preprocessing module, which is configured to: preprocess the multi-channel EEG signal to be detected;

[0052] The detection module is configured to: use CNN and Transformer models to obtain the epilepsy diagnosis result of the multi-channel EEG signal to be detected; specifically include: inputting the EEG signal slice into a pre-trained graph attention residual network In , the pre-trained CNN and Transformer models are used to extract local and global features of the EEG signals of each channel.

[0053] Further, the preprocessing of the multi-channel EEG signal to be detected includes:

[0054] Denoising is performed on the multi-channel EEG signal to be detected.

[0055] Further, the preprocessing of the multi-channel EEG signal to be detected also includes:

[005...

Embodiment 2

[0115] A computer-readable storage medium, wherein a plurality of instructions are stored, and the instructions are suitable for being loaded by a processor of a terminal device and executing the described epilepsy detection method based on CNN and Transformer, the method comprising the following steps:

[0116] Obtain the multi-channel EEG signal to be detected;

[0117] Preprocess the multi-channel EEG signal to be detected, slice the preprocessed EEG signal, and prepare for input to the model;

[0118] All read slices are input into the pre-trained CNN and Transformer models, and the epilepsy diagnosis results of multi-channel EEG signals to be detected are output.

Embodiment 3

[0120] A terminal device, including a processor and a computer-readable storage medium, the processor is used to implement instructions; the computer-readable storage medium is used to store multiple instructions, and the instructions are suitable for being loaded by the processor and executing the described one Based on the epilepsy detection method of CNN and Transformer, described method comprises the following steps:

[0121] Obtain the multi-channel EEG signal to be detected;

[0122] Preprocessing the multi-channel EEG signal to be detected, and slicing the preprocessed signal of each channel;

[0123] All read slices are input into the pre-trained CNN and Transformer models, and the epilepsy diagnosis results of multi-channel EEG signals to be detected are output.

[0124] Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may...

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 provides an epilepsy detection system based on CNN and Transform, comprising: an acquisition module configured to: acquire a multi-channel electroencephalogram signal to be detected; the pre-processing module is configured to pre-process the multi-channel electroencephalogram signals to be detected; the detection module is configured to obtain an epilepsy diagnosis result of the multi-channel electroencephalogram signal to be detected by utilizing a CNN (Convolutional Neural Network) and a Transform model; the method specifically comprises the steps that the electroencephalogram signal slices are input into a pre-trained graph attention residual network, and local and global feature extraction is conducted on the electroencephalogram signals of each channel through a pre-trained CNN and Transform model. According to the epilepsy detection method, the CNN model and the Transform model are combined to be applied to epilepsy detection for the first time, and the limitation of capturing global features in a single CNN model and the problem that details of local features are ignored in a single Transform model are solved.

Description

technical field [0001] The invention relates to the technical field of EEG signal processing, in particular to an epilepsy detection system based on CNN and Transformer. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] EEG is a graph obtained by amplifying and recording the spontaneous biopotential of the brain from the scalp through sophisticated electronic instruments, and is used to assist in the detection of epilepsy. Since the EEG can accurately record scattered slow waves, spikes or irregular spikes during epilepsy, the EEG is very accurate for the diagnosis of epilepsy, so the EEG is an indispensable method for the diagnosis of epilepsy. Inspection Method. Epilepsy patients need frequent review of EEG. A seizure is an abnormal firing of neurons in the brain that appears as "abnormal waves" on an EEG. This wave can occur during se...

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): A61B5/00A61B5/16
CPCA61B5/4094A61B5/7267A61B5/7207A61B5/168
Inventor 赵艳娜褚登雨张高波董长续薛明睿何佳桐郑元杰
Owner SHANDONG NORMAL UNIV