Single Lead Dynamic ECG Signal's QRST Cancellation Based On ESN

Atrial fibrillation is one of the most common aging diseases,with morbidity increasing year by year.At present,it is difficult to effectively track the changes in the state of atrial fibrillation,resulting in low accuracy of the diagnostic of the atrial fibrillation and poor curative effect.With the development of remote ECG monitoring system as well as the miniaturization and convenience of medical equipment,single-lead ECG monitoring system has realized more applications,the ECG monitored by the remote ECG monitoring system is dynamic ECG.Therefore,it is of great significance to study the QRST cancellation method for single-lead dynamic ECG signals.The main contents of this paper are as follows:The Echo State Network is applied to cancel QRST for the single-lead dynamic ECG signals.The traditional ESN-based QRST cancellation algorithm(ESNTL)is improved by adding an input vector which indicates the R wave position,so that the network can find the position of QRST more accurately.This method can improve the performance of the QRST cancellation effectively.The differential evolution algorithm and binary particle swarm optimization algorithm are employed to optimize the parameters of the ESN.The differential evolution algorithm is used to optimize the reserve pool parameters,and the binary particle swarm optimization algorithm is used to optimize the output layer connection of ESN and improves the generalization ability of the network.The proposed method is applied to the ECG monitoring service platform developed by our research group.The testing results with the collected ECG signal show the feasibility of the proposed method.