Wearable Devices Acquired ECG Signals Detection Method Using 1D Convolutional Neural Network

According to reports, the number of people who die from sudden cardiac death in China every year is as high as 540,000, and the number of deaths due to arrhythmia accounts for about 90%. Although a single arrhythmia heartbeat may not seriously affect life, continuous arrhythmia can lead to fatal conditions. Therefore, it is very important to monitor the heart rhythm regularly to control and prevent arrhythmia. Because the ECG is too complex and large, sports bracelet on sale can only monitor the momentary ECG, we designed a wearable, portable device that can continuously monitor heart rhythm in real time. At the same time, a classification method with low complexity and high accuracy classification network is proposed to detect arrhythmia. The results show that compared with other existing algorithms, our proposed 1D CNN model has improved accuracy and reduced network complexity.