An automatic detection system for atrial fibrillation based on deep learning

Abstract

The invention discloses an automatic detection system for atrial fibrillation of electrocardiographic signals based on deep learning, which includes an electrocardiographic signal processing module, a training module and an atrial fibrillation detection module; Detect to obtain the RR interval sequence; preprocess the RR interval sequence, and sample the preprocessed RR interval sequence from the initial moment of its corresponding ECG signal based on the sliding window to obtain the RR interval sample; where , there is a time correlation between the RR interval values ​​in each RR interval sample; the atrial fibrillation detection module is used to use the atrial fibrillation signal recognition model based on the CLDNN network to determine whether each extracted RR interval sample is an atrial fibrillation signal The automatic detection of atrial fibrillation on the ECG signal can be carried out, which can not only extract the inherent characteristics of the ECG signal, but also consider its time correlation characteristics, and the detection accuracy is high; in addition, the system is an end-to-end system, more convenient and faster.

Description

technical field [0001] The invention belongs to the technical field of electrocardiogram analysis, and more specifically relates to an automatic detection system for atrial fibrillation of electrocardiogram signals based on deep learning. Background technique [0002] Atrial fibrillation is the most common tachyarrhythmia symptom in clinical practice, which is characterized by irregular excitation of the atrium. There are about 33.5 million people suffering from atrial fibrillation in the world, among which the number of atrial fibrillation patients in my country exceeds 10 million, which is one of the countries with the largest number of atrial fibrillation patients in the world. In addition to arrhythmia, atrial fibrillation often leads to other complications, which can form left atrial thrombus and lead to arterial embolism. 90% of the resulting embolism is cerebral artery embolism (ischemic stroke), and 10% is peripheral arterial embolism or mesenteric artery embolism, ...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the above defects or improvement needs of the prior art, the present invention provides an automatic detection system for atrial fibrillation of ECG signals based on deep learning. Technical problems with low detection accuracy

Images