A method and system for classifying and identifying electrocardiogram signals based on feature extraction

By extracting the higher-order features of the electrocardiogram signal and using neural networks for classification and identification, the problem of low diagnostic recognition rate of the existing technology center is solved, and high-accuracy classification of the electrocardiogram signal and early cardiovascular disease screening is achieved.

CN115153583BActive Publication Date: 2024-11-01杭州数智莱达科技有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210887771.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-26
Publication Date
2024-11-01
Estimated Expiration
2042-07-26

Smart Images

  • Figure CN115153583B_ABST
    Figure CN115153583B_ABST
Patent Text Reader

Abstract

The present invention proposes an electrocardiogram signal classification and recognition method and system based on feature extraction. The method is carried out based on a training mode and a test mode. The continuous electrocardiogram signal time series in the training mode is analyzed and processed. On the basis of extracting the principal component data vector of the time-varying data series of the P wave, the QRS wave and the T wave, a neural network is used to model and approximate the information and variation law of the nonlinear high-order derivative model of the position under the time-varying data, so that the extracted features can better reflect the essential time-varying characteristics of the electrocardiogram signal, thereby having better fault tolerance to changes in complex conditions. In the test mode, the same analysis and processing steps are used for the continuous electrocardiogram time series. By extracting different electrocardiogram time domain nonlinear indicators in the two modes, the nonlinear characteristics of the electrocardiogram system are reflected, so that the electrocardiogram patterns under different conditions are conveniently compared, and then used for electrocardiogram signal classification. The recognition process is simple and rapid with high accuracy.
Need to check novelty before this filing date? Find Prior Art