Prediction models for hyperglycemia effect duration of cardiac conduction systems using sinoatrial node field potential

Abstract

Sinoatrial node field potential is an important electrophysiological signal in cardiac conduction systems. The field potential is highly sensitive to high glucose. With the different effect duration of high glucose, the frequency characteristics of the field potential changes remarkably. In this paper, prediction models were built by using partial least squares (PLS), least squares support vector machine (LSSVM) and back propagation neural network (BPNN) respectively to predict the effect duration of hyperglycemia on sinoatrial node field potential in different glucose concentrations. Meanwhile, the prediction results of the three models were compared. The results showed that the predictive capability of the LSSVM was the highest and the model is very suitable for hyperglycemia effect duration prediction.