|Speciale, modelbygger-variant, 3. modul, 2010, id:406|
|Vejleder:||Johnny Ottesen og Mette S. Olufsen|
|Findes på RUb:||Ja|
In this thesis we examined the phenomena caused by initial gravitational effects and short term neural control when the human body is subjected to a head-up-tilt (hut) test. This was done by constructing a simple mathematical model, which describes the systemic circulation of the cardiovascular system. The human body is divided into two parts; the lower and the upper body, which are further divided into venous and arterial compartments. The left heart isdescribed as a pump with time-varying elastance, thus making the model pulsatile. The system is modeled as an electrical analogue in which pressure is analogous to voltage. Flow to current, compliance to capacitance, and resistance is the same in both settings. The circuit can be represented by five non-linear ordinary differential equations. To predict dynamics during hut, the neural control is included by defining time-varying peripheral resistance, venous compliance, and heart contractility. These regulating mechanisms were modeled by set-point fuctions and sigmoidal functions, which resemble the behavior seen in experiments. We also developed an idealized cardiovascular model of the human body, where teh body is considered an elastic cylinder. We found the hydrostatic invariance point, and related this to the hut. The model was validated using data from a healthy 25 year old subject who underwent hut, and parameter estimation techniques were employed to estimate a set of parameters that minimized the least squares error between the model and the data. By identifying correlated parameter pairs and fixing insensitive parameters we were able to obtain reliable estimates for five model parameters during hut, and nine parameters in steady state. In steady state, we used three data sequences to obtain more reliable estimates. The constructed model was able to predict dynamics observed in the data both during steady state and during hut. There are fluctuations in the data which cannot be explained by the model, but we believe these originate from the respiration of the patient, something that is not included in the model. All model variables produce physiologicaly reasonable results during the hut: cardiac output drops the same amount as seen in the lterature, lower venous pressure is increased in agreement with other studies, and upper arteral pressure resembles the data obtained. These results show that the simple model can be used to predict responses from larger groups of subjects. Subsequent analysis of parameter estimates have potential to be used to predict differences within and between these groups of subjects. However, before using the model in a larger study, we suggest to include the following model improvements: 1) Include a simple model for the respiration of the patient. 2) The control of the lower venous compliance should be a strictly mechanical feature, and not part of the baroreflex system. 3) The upper peripheral resistance should also be controlled.