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

Ultra-lightweight convolutional neural network (CNN)-based electrocardiograph (ECG) signal classification method

A technology of convolutional neural network and signal classification, which is applied in the field of ECG signal classification based on ultra-lightweight convolutional neural network, which can solve the problem of lowering the order of magnitude, the large amount of parameters in the fully connected layer, and the large amount of calculation in the conventional convolutional layer, etc. problem, to achieve the effect of reducing the amount of calculation and reducing the pressure of calculation

Active Publication Date: 2020-11-20
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF7 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Aiming at the above-mentioned deficiencies in the prior art, an ECG signal classification method based on an ultra-lightweight convolutional neural network provided by the present invention solves the problem of the large amount of parameters of the conventional fully-connected layer in the CNN used by the end-to-end detection algorithm and the conventional The problem of the huge amount of computation in the convolutional layer can reduce the order of magnitude of the computational complexity of the algorithm to a level suitable for devices with extremely limited storage resources

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
  • Ultra-lightweight convolutional neural network (CNN)-based electrocardiograph (ECG) signal classification method
  • Ultra-lightweight convolutional neural network (CNN)-based electrocardiograph (ECG) signal classification method
  • Ultra-lightweight convolutional neural network (CNN)-based electrocardiograph (ECG) signal classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0122] Embodiment 1: the ECG signal classification model of the training process in step S2 includes: the first two classification models and the first five classification models;

[0123] Such as figure 2 As shown, the first two classification model includes: the first DSCEMP convolutional layer, the second DSCEMP convolutional layer, the first LDSCP fully connected layer, the first fully connected layer and the first softmax layer;

[0124] The first DSCEMP convolutional layer, the second DSCEMP convolutional layer, the first LDSCP fully connected layer, the first fully connected layer and the first softmax layer are sequentially connected; the input end of the first DSCEMP convolutional layer is used as the first The input end of a two classification model; The output end of the first softmax layer is used as the output end of the first two classification model;

[0125] Such as image 3 As shown, the first five classification models include: the third DSCEMP convolution...

Embodiment 2

[0184] Embodiment 2: the ECG signal classification model of the training process in step S2 comprises: the second two classification models and the second five classification models;

[0185] The second two classification model includes: the fifth DSCEMP convolutional layer, the sixth DSCEMP convolutional layer, the second LDSCP fully connected layer, the third fully connected layer and the third softmax layer;

[0186] The fifth DSCEMP convolutional layer, the sixth DSCEMP convolutional layer, the second LDSCP fully connected layer, the third fully connected layer and the third softmax layer are sequentially connected; the input end of the fifth DSCEMP convolutional layer is used as the first The input terminal of the binary classification model; the output terminal of the third softmax layer is used as the output terminal of the second binary classification model;

[0187] The second five classification models include: the seventh DSCEMP convolutional layer, the eighth DSCEM...

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 ultra-lightweight convolutional neural network (CNN)-based electrocardiograph (ECG) signal classification method. The ECG signal classification method comprises the following steps: S1, acquiring an ECG data set, and making the ECG data set into a training set and a validation set; S2, performing training and validation on an ECG signal classification model by employingthe training set and the validation set to obtain an optimal ECG signal classification model; S3, collecting ECG heartbeat signals in real time and preprocessing the ECG heartbeat signals to obtain multiple segments of ECG data; and S4, inputting the multiple segments of ECG data into the optimal ECG signal classification model in sequence to obtain a classification result of the ECG data. The ECGsignal classification method solves the problems of the huge amount of parameters of a conventional full-connecting layer and the huge amount of calculation of a conventional convolution layer in a CNN used by an end-to-end detection algorithm, and realizes that the magnitude order of algorithm calculation complexity is reduced to a level suitable for storing resources and limited equipment thereof.

Description

technical field [0001] The invention relates to the field of ECG signal classification, in particular to an ECG signal classification method based on an ultra-lightweight convolutional neural network. Background technique [0002] An electrocardiogram (ECG) records the electrical signals passing through the heart and is used clinically to further diagnose conditions related to cardiac arrhythmias. The traditional arrhythmia detection method uses the hospital's bulky electrocardiometer to collect short-term ECG signals of the patient, and the cardiologist visually diagnoses them. However, since arrhythmias occur intermittently, especially in the early stages of the problem, it is difficult to detect arrhythmias from ECG signals in a short time window and miss the best treatment opportunity for cardiac patients. Therefore, long-term ECG monitoring with real-time arrhythmia detection capability is necessary for early detection of potential problems. [0003] Long-term and pra...

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): A61B5/0452A61B5/04
Inventor 周军肖剑彪
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA