Nonlinear analyses of data in cardiovascular physiology and epidemiology

Abstract

The nonlinear dynamics of heart action are such as to generate and to support irregular nonrandom variations in the main output parameters of pulse and pressure, in response to external stimulation such as stressors (Pearson 1972; Gregson 2009; Herbert 1995; Mezentseva et al. 2002; Rao and Yeragani 2001; Shiferaw and Karma 2006). These dynamic complexities have serious implications for the analysis and treatment of hypertension, as the probabilities of misdiagnosis of state, dysfunction, and potential morbidity cannot with certainty be based on the linear or Gaussian statistics usually employed in epidemiological and related studies. A Bayesian analysis of decision probabilities is used to point up ambiguities in reported results. There is gradual recognition of this situation, but so far little established linkage between nonlinear dynamical modeling and clinical practice, except perhaps in some recent biomedical physics modeling.