Heart condition analysis from ECG signals based on nonlinear theory

In the article ECG signals both normal and arrhythmia are studied with different nonlinear methods including chaotic attractor, RPS, correlation dimension, etc, to revealing the connection between ECG signals and heart rhythm conditions from nonlinear view. The results indicate that when heart rhythm state changes, the projection of ECG signals attractors in RPS changes remarkably, which corresponds with clinic diagnosis. Relating to results from correlation dimension combined with SVD, some certain pattern can be discovered. Meanwhile, the nonlinear dynamic process of the changing heart rhythm state is discussed to have a new sight into some kind of heart rhythm from nonlinear theory, in contrast to the clinic physiological characters. The results can be reference for intelligent diagnosis of heart disease.