A method and device for identifying and evaluating ECG signals based on nmf

Abstract

The invention discloses an NMF-based ECG signal identification and evaluation method and device. The method includes: receiving the ECG signal as a sample point, and preprocessing the ECG signal; according to the structure of the ECG signal, Perform two-way granular scanning of the QRS band, and integrate it with the band before and after the QRS to reconstruct the preprocessed ECG signal; perform non-negative matrix decomposition on the reconstructed ECG signal to obtain coefficients matrix; using the subcomponent analysis method to select the subcomponent of the coefficient matrix as the final eigenvalue matrix of the ECG signal; using the eigenvalue matrix to identify the ECG signal. Using the eigenvalue matrix, the homologous similarity matrix and the heterologous similarity matrix are calculated by Mahalanobis distance, and the misrecognition rate, rejection rate and equal error rate are calculated from the homologous similarity matrix and heterologous similarity matrix. The recognition effect of electrical signal identification is measured and evaluated.

Description

technical field [0001] The disclosure belongs to the technical field of electrocardiographic signal identification, and relates to a method and device for identification and evaluation of electrocardiographic signals based on NMF. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] With the advancement and development of science and technology, due to the uniqueness of ECG signals, high security, and strong collectability, ECG signals can be used to identify identities. However, there are the following difficulties when using ECG signals for identification: [0004] First, in the acquisition process, it will be affected by external factors and cause a certain degree of distortion; [0005] Second, the ECG signal is also easily affected by the psychological and physiological activities of the human body; [0006] Therefore, the collected EC...

AI Technical Summary

Problems solved by technology

[0004] First, in the acquisition process, it will be affected by external factors and cause a certain degree of distortion;

[0005] Second, the ECG signal is also easily affected by the psychological and physiological activities of the human body;

[0008] In wavelet decomposition, it is difficult to select wavelet basis and the decomposition result is not unique;

[0009] In the empirical mode decomposition, since the ECG signal is a non-stationary signal, when the signal has large fluctuations, pulse interference, or large noise influence, it is easy to produce modal aliasing, which is not conducive to the analysis of the ECG signal. identify

Images