Hemodynamic Modeling of the Cardiovascular System
Under the Supervision of:
Dr. Ilker Tunay
Stereotaxis, St. Louis, MO
Abstract
A simulation model of the cardiovascular system has been developed that is easily adaptable for a variety of conditions and subjects.
The model is a crucial piece in a larger project whose ultimate goal is to develop a localization signal for a medical device in a patient’s heart that filters out interfering signals from respiration and the beating of the heart.
This report includes a full derivation of the model as well as the logic behind a Simulink implementation.
In addition, a variety of different scenarios were simulated for humans, including normal conditions, exercise, sleep, pulmonary embolism, systemic hypertension, myocardial infarction, and arrhythmia.
In addition, a canine model was developed, and simulations were conducted for normal conditions and arrhythmia. The cardiovascular model yields results within normal ranges for the normal condition simulations,
and the scenario modeling trends mirror those found in real subjects. This model should provide a solid basis for future development of the project.
Goal
- To create an effective simulation model of the cardiovascular system, which can be incorporated into future work
- The simulated results of this model must be similar to those found in real subjects and must be easily incorporated into future work.
- The model must also be applicable to a wide variety of subjects. (This is especially significant because initial testing will be performed on canines)
Modeling Approach
The pressure-volume approach divides the cardiovascular system into a series of elastic chambers, each with its own resistance. This pressure volume approach was chosen for the relatively small number of parameters required and because these parameters can easily be adapted for a wide range of subjects.
Matlab and Simulink are used to implement this approach because of the convenience of Simulink for simulating such physical processes. In addition, Matlab is widely available and will probably be used for future work in this project.
Model Derivation
Several sub models were needed to create the full cardiovascular model:
- Model of the electrical signals in the heart
- Circulation Model
- Respiration Model (created by Brent Goldman)
Scenario Simulations
Once the cardiovascular model was completed, a variety of scenarios were simulated to
determine the reliability and applicability of the model.
- Human Simulations
- Normal Conditions
- Exercise
- Sleep
- Pulmonary Embolism
- Systemic Hypertension
- Myocardial Infarction
- Arrhythmia
- Canine Simulations
- Normal Conditions
- Arrhythmia
Summary of Results
- The simulations of normal conditions for humans and canines resulted in output values within normal ranges.
- The human exercise simulation yielded the expected trends, such as high cardiac output and slight increase in stroke volume and arterial pressure.
- The human sleep simulation yielded the expected decrease in systemic arterial pressure and increase in the end diastolic volumes of the ventricles.
- The human pulmonary embolism simulation yielded the expected decrease of stroke volume in the ventricles and increase in pulmonary arterial pressure.
- The human systemic hypertension simulation yielded a decreased stroke volume of the left ventricle and the expected increase in systemic arterial pressure to above 140 / 90 mmHg.
- The human myocardial infarction simulation yielded the expected decrease in systemic arterial pressure, increase in ventricular volume, and decrease in stroke volume and cardiac output.
- The human and canine arrhythmia simulations yielded the expected decrease in stroke volumes.
Conclusion
The goal was to produce a model of the cardiovascular system that is easy to adapt to a wide range of conditions and subjects. This goal has clearly been demonstrated through the simulation of several different scenarios in humans and canines.