Baroreflex modeling in the genesis of stress reactivity using sigmoidal characteristic

According to a physiological hypothesis, children are separated into the two groups, (1)non-reactive and (2)high-reactive based on their different autonomic reactivity characteristic. In the non-reactive group, blood pressure(BP) and heart rate(HR) are regulated in a timely manner following external disturbances such as a stressful condition. However, this regulation process does not operate properly, or may even behave in an opposite direction for at least a period of the process in the high-reactive group. The purpose of this research is to analyze and compare the behavioral differences of the autonomic reactivity characteristic, represented by the short-term blood pressure regulation system (STBPRS), between these two groups of individuals. Similar to any regulation system, each component of this system has a specific role. For example, the autonomic nervous system (ANS) can be considered a controller while the heart and vasculature can be considered a plant under control. The arterial baroreceptor nerves - fiber endings in the arterial walls - play the role of sensor and feedback path. The STBPRS is called as baroreflex or baroreceptor reflex including the ANS and baroreceptors. We applied the Windkessel model and sigmoidal function as the model structures of the vasculature and baroreflex, respectively. To obtain the most similar simulated HR in comparison with measured HR, an optimization problem was defined. Due to the non-convex nature of the optimization problem, a genetic algorithm (GA) was applied to identify all of the corresponding unknown parameters for each component of the system. The obtained results of the system identification problem, verify the mentioned physiological hypothesis. Moreover, these results lead to a better understanding of the deficient baroreflex in high-reactive children. Furthermore, necessities of invasive blood pressure measurement in baroreflex studies is eliminated by using our proposed method.