Correlation dimension and the largest Lyapunov exponent characterization of RR interval.
OBJECTIVE: To search for evidence of deterministic chaos and to characterise it quantitatively in RR intervals of different subjects. METHOD: An improved Grassberger-Procaccia (GP) algorithm was proposed for computing the correlation dimensions (CD) of RR intervals in ten healthy young men, ten healthy old men, and ten atrial fibrillation (AF) patients. Analysis was also performed by calculating the largest Lyapunov exponents. RESULT: It was found that the cardiac systems of healthy young men are some 7-8 dimensional complex dynamic systems, while those of healthy old men are 6-6.5 dimensional, and those of AF patients had dimensions of about 4-4.5. The young men had the largest Lyapunov exponents of about 0.019-0.23, the largest Lyapunov exponents of old men was 0.12-0.16, and that of AF patients was 0.083-0.123. CONCLUSION: The above results demonstrated that the RR intervals could be chaotic. The decrease in the CDs and the largest Lyapunov exponents as we move from the young to the old and then to the AF patients reflects the loss of complexity as the heart becomes less healthy. Therefore, correlation dimension and the largest Lyapunov exponent could be used as some novel non-invasive analysis tools for clinical cardiology.