Method for identifying atrial fibrillation and atrial premature beats from 10-second electrocardiogram

Abstract

The invention discloses a method for identifying atrial fibrillation and atrial premature beats from a 10-second electrocardiogram. The method comprises the following steps that S1, firstly, an electrocardiogram signal is pretreated to filter out baseline drift, power frequency interference and other noise, the electrocardiogram signal after the noise is filtered out is re-sampled to a certain fixed sampling rate, then electrocardiogram QRS wave feature points are extracted by adopting a method based on a wavelet transformation and logic regression algorithm, and then a T wave is searched forby adopting a search window dynamically determined based on an instantaneous heart rate. The invention relates to the technical field of medical examination. The method for identifying atrial fibrillation and atrial premature beats from the 10-second electrocardiogram has the advantages that by defining a set of feature parameters based on heart rate change and a P wave, an obtained feature combination can identify atrial fibrillation and other heart rhythm better, the feature parameters are less sensitive to noise interference, a set of feature parameters based on the P wave are defined, theaccuracy is further improved, and quality information of the electrocardiogram signal is utilized, so that the identification of PAC is more accurate.

Description

technical field [0001] The invention relates to the technical field of medical examination, in particular to a method for identifying atrial fibrillation and atrial premature beats from a 10-second electrocardiogram. Background technique [0002] At present, when identifying atrial fibrillation and atrial premature beats on the ECG, due to the difficulty in accurately identifying P waves, similar techniques usually rely on the RR interval of the ECG, where the root mean square of continuous difference (RMSSD), turning point ratio (TPR) The three parameters of Shannon entropy (SE) and Shannon entropy (SE) are the most important. It is stipulated that when these three parameter values ​​exceed their respective thresholds at the same time, the ECG is atrial fibrillation. These parameters are all derived from the RR interval of the ECG. Due to atrial premature beats (PAC), Premature ventricular contractions (PVCs), sinus arrhythmias, and second-degree atrioventricular block can ...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method for identifying atrial fibrillation and atrial premature beats from a 10-second electrocardiogram, which solves the problem that the prior art is easily affected by noise when detecting atrial premature beats, and there is a greater possibility of It is wrongly detected as atrial premature beat, which will greatly reduce the recognition accuracy of PAC

Images