Analysis of Heart Rate Variability to Predict Patient Age in a Healthy Population

OBJECTIVES To estimate age of healthy subjects by means of the heart rate variability (HRV) parameters thus assessing the potentiality of HRV indexes as a biomarker of age. METHODS Long-term indexes of HRV in time domain, frequency domain and non-linear parameters were computed on 24-hour recordings in a dataset of 63 healthy subjects (age range 20-76 years old). Then, as interbeat dynamics markedly change with age, showing a reduced HRV in older subjects, we tried to capture age-related influence on HRV by principal component analysis and to predict the subject age by means of a feedforward neural network. RESULTS The network provides good prediction of patient age, even if a slight overestimation in the younger subjects and a slight underestimation in the older ones were observed. In addition, the important contribution of non-linear indexes to prediction is underlined. CONCLUSIONS HRV as a predictor of age may lead to the definition of a new biomarker of aging.

[1]  I. Antelmi,et al.  Influence of age, gender, body mass index, and functional capacity on heart rate variability in a cohort of subjects without heart disease. , 2004, The American journal of cardiology.

[2]  S M Pincus,et al.  Approximate entropy as a measure of system complexity. , 1991, Proceedings of the National Academy of Sciences of the United States of America.

[3]  S. Sakata,et al.  Aging and spectral characteristics of the nonharmonic component of 24-h heart rate variability. , 1999, The American journal of physiology.

[4]  G. Piccirillo,et al.  Power spectral analysis of heart rate in subjects over a hundred years old. , 1998, International journal of cardiology.

[5]  C. Peng,et al.  Cardiac interbeat interval dynamics from childhood to senescence : comparison of conventional and new measures based on fractals and chaos theory. , 1999, Circulation.

[6]  J. Fleiss,et al.  Power law behavior of RR-interval variability in healthy middle-aged persons, patients with recent acute myocardial infarction, and patients with heart transplants. , 1996, Circulation.

[7]  Rich Caruana,et al.  Overfitting in Neural Nets: Backpropagation, Conjugate Gradient, and Early Stopping , 2000, NIPS.

[8]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[9]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[10]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[11]  A Giuliani,et al.  Estimating a cardiac age by means of heart rate variability. , 1997, American journal of physiology. Heart and circulatory physiology.

[12]  H. Stanley,et al.  Quantification of scaling exponents and crossover phenomena in nonstationary heartbeat time series. , 1995, Chaos.

[13]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..