Abstract
One of the clinical examinations performed to evaluate the autonomic nervous system (ANS) activity is the tilt test, which consists in studying the cardiovascular response to the change of a patient's position from a supine to a head-up position. The analysis of heart rate variability signals during tilt tests has been shown to be useful for risk stratification and diagnosis on different pathologies. However, the interpretation of such signals is a difficult task. The application of physiological models to assist the interpretation of these data has already been proposed in the literature, but this requires, as a previous step, the identification of patient-specific model parameters. In this paper, a model-based approach is proposed to reproduce individual heart rate signals acquired during tilt tests. A new physiological model adapted to this problem and coupling the ANS, the cardiovascular system (CVS), and global ventricular mechanics is presented. Evolutionary algorithms are used for the identification of patient-specific parameters in order to reproduce heart rate signals obtained during tilt tests performed on eight healthy subjects and eight diabetic patients. The proposed approach is able to reproduce the main components of the observed heart rate signals and represents a first step toward a model-based interpretation of these signals.
Highlights
Heart rate variability (HRV) represents one of the most efficient indicators to characterize the modulation of the cardiovascular system (SCV) by the autonomic nervous system (ANS) [1]
Head-up tilt test has been applied to eight healthy subjects and eight type 2 diabetic patients
This formalism seems to be adapted to the description of the circulation and global mechanical activity of the ventricles, the marked nonlinearities involved in the genesis of the cardiac action potential has been modeled by a set of ordinary differential equations and coupled with the global Bond Graph model
Summary
Heart rate variability (HRV) represents one of the most efficient indicators to characterize the modulation of the cardiovascular system (SCV) by the autonomic nervous system (ANS) [1]. Time domain and frequency domain methods have been developed to assist the signal analysis and to estimate the levels of vagal (parasympathetic) and sympathetic activites [1] These classical indicators provide useful information and have been widely used in clinical practice, a model-based approach can be useful to complement this information and to ease its interpretation, as these mathematical models directly represent the interactions between the ANS and the CVS [2]. Such a model could assist in the prediction of the patient’s response to different physiological conditions or therapeutic strategies. A necessary step for such model-based interpretation methods is the creation of a patient-specific instance of the model, characterized by individualized parameters
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
