A defibrillation rhythm recognition device

Abstract

The embodiment of the invention discloses a defibrillation rhythm recognition device based on a sequential pool convolution neural network and a recurrent neural network, which processes the originalelectrocardiogram signal through a sliding window. The band-pass filtering and waveform slope calculation of each sliding window ECG data are realized by convolution layer. Extracting the distributioncharacteristics of ECG data slope of each sliding window; the characteristics of slope distribution are input into recurrent neural network, and the defibrillation rhythm is identified by the dynamictime series of slope distribution. The heart rhythm recognition model aims at minimizing the recognition error; based on optimization of ECG signal preprocessing, feature extraction and classification, by making full use of the dynamic information of the slope distribution of ECG signal, the device realizes the end-to-end learning from the original ECG signal to the cardiac rhythm class marker, and can automatically adjust the parameters of band-pass filter and slope calculation, thus providing more accurate criteria for out-of-hospital cardiac defibrillation first aid, reducing the demand for ECG detection time and improving the first aid efficiency.

Description

technical field [0001] The present invention relates to the technical field of data identification, in particular to a defibrillable cardiac rhythm identification device based on sequential pooling convolutional neural network and recurrent neural network. Background technique [0002] According to the "China Cardiovascular Disease Report 2016" released by the National Cardiovascular Disease Center, the number of sudden cardiac death cases in my country exceeds 540,000 every year, which is equivalent to about 1,500 people dying of sudden cardiac death every day. Sudden cardiac death is mainly caused by Ventricular Fibrillation (VF) and Ventricular Tachycardia (VT). VF / VT attacks are often without warning. During the attack, the electrical activity of the ventricle is out of synchrony, and the pumping function of the heart is lost. If measures are not taken in time to convert the heart rhythm, the patient will die within a few minutes. [0003] Deploying Automatic External D...

AI Technical Summary

Problems solved by technology

Although the inventive scheme uses the order statistics of the waveform slope to better characterize the morphological difference of ShR/NshR, it needs to set the difference inte

Images