Evaluating age-related loss of nonlinearity degree in short-term heartbeat series by optimum modeling dimension

Detecting the alterations of nonlinearity in short-term heartbeat dynamics has got the realistic meaning. The paper first studies some short nonlinear series by predicting the series with a nonlinear autoregressive model. Then inspired by the fact that the optimum modeling dimension of the model can characterize the intrinsic dimensional feature of the dynamics, the paper presents a new idea that the optimum modeling dimension can be taken to evaluate the nonlinearity degree of the original dynamics. Finally, we apply this proposition to the heartbeat dynamics. We study the short-term heart interbeat interval series of a group of healthy subjects and have a comparison between the results of the young subjects and those of the old ones. We find that the optimum modeling dimensions of the old subjects are smaller than those of the young ones. It shows the nonlinearity degree decreases with aging.