Deep Learning for Real-time ECG R-peak Prediction

Abstract

In this work, we propose a novel algorithm to achieve real-time R-peak prediction during ECG signal recording. More specifically, from the current frame of ECG signal, we aim to predict how far into the future the next R-peak will occur, taking into account information about the variability in the intervals between beats seen in previous frames of the ECG signal. Currently there has been little work in this area. However, the real-time prediction of the next beat has important research significance, including timing of artificial heart pumps and integration into the cardiopulmonary support heart (CPS-heart), as well as narrowing the search range of R-peaks to assist R-peak detection. This paper proposes use of an integrated network using one-dimensional convolution network (1D CNN) with long short-term memory (LSTM) network. The deep learning model we have proposed is shown to effectively predict R-peaks with results of preliminary studies achieving a prediction accuracy of 90.61%.