Knowledge base: Warsaw University of Technology

Settings and your account

Back

Modeling the action of pulsatile heart assist

Alicja Siewnicka

Abstract

The aim of this work was to derive a description that yields the blood flow rate estimation of the pneumatically controlled, pulsatile ventricular assist device. An adequate flow rate is crucial for the effectiveness of the therapy. Unfortunately, during clinical heart support, neither blood flow nor pressures can be measured from the device. In general, the flow rate depends on the device control and patient conditions. However, the patient’s hemodynamic parameters are not constantly monitored. For this reason, there was a need for development of a method of blood flow evaluation which would base on the standard measurements from the device control unit. For this purpose, an analysis of the device construction was carried out, the problems were identified and the directions of the necessary experimental work were defined. Three research experiments have been conducted for the different control units and various operating conditions of the cardiac assist system. Based on all signals available under the experimental conditions, preliminary tests were conducted to identify the most effective modeling method. A solution was proposed in the form of analytical model of the device based on the signal, which can contain information about the blood flow rate. It has been observed that this may be a signal of the air flow in the supply drain or a value of difference between supply pressure during assist and idle run. In the first case, a model with a high accuracy was defined but the additional measurements on the pneumatic side of the device were required. In the second approach a two-stage model was proposed. The first-stage model estimates the value of air supply pressure for the idle run. Its output signal is used in a second-stage model to estimate the blood flow in the normal operation of the device. In this way, a model based solely on the measurement signals available in the control unit was obtained. The achieved accuracy is comparable to all tested control and load conditions and is independent from the level of the device filling. An analysis of the utility of the developed model was carried out due to the purpose of device stroke volume estimation and the automatic detection of the extreme filling conditions. The obtained results have confirmed the utility of the developed model. The final conclusions from the conducted research were formulated and possible directions of the further research on the clinical use of the model were presented.
Record ID
WUT4b641791d940462aa5bb85b4c1f29aaa
Diploma type
Doctor of Philosophy
Author
Title in Polish
Modelowanie działania pulsacyjnego zespołu wspomagania pracy serca
Title in English
Modeling the action of pulsatile heart assist
Language
(pl) Polish
Certifying Unit
Faculty of Mechatronics (FM)
Discipline
mechanical engineering / (technology domain) / (technological sciences)
Status
Finished
Start date
01-10-2001
Defense Date
15-11-2017
Title date
29-11-2017
Supervisor
External reviewers
Jacek Kluska Jacek Kluska,, Author's external affiliation: Faculty of Electrical and Computer Engineering
Marek Darowski Marek Darowski,, Author's external affiliation: Instytut Biocybernetyki i Inżynierii Biomedycznej im. Macieja Nałęcza Polskiej Akademii Nauk
Pages
122
Keywords in English
modelling, flow estimation, ventricular assist device, VAD
Abstract in English
The aim of this work was to derive a description that yields the blood flow rate estimation of the pneumatically controlled, pulsatile ventricular assist device. An adequate flow rate is crucial for the effectiveness of the therapy. Unfortunately, during clinical heart support, neither blood flow nor pressures can be measured from the device. In general, the flow rate depends on the device control and patient conditions. However, the patient’s hemodynamic parameters are not constantly monitored. For this reason, there was a need for development of a method of blood flow evaluation which would base on the standard measurements from the device control unit. For this purpose, an analysis of the device construction was carried out, the problems were identified and the directions of the necessary experimental work were defined. Three research experiments have been conducted for the different control units and various operating conditions of the cardiac assist system. Based on all signals available under the experimental conditions, preliminary tests were conducted to identify the most effective modeling method. A solution was proposed in the form of analytical model of the device based on the signal, which can contain information about the blood flow rate. It has been observed that this may be a signal of the air flow in the supply drain or a value of difference between supply pressure during assist and idle run. In the first case, a model with a high accuracy was defined but the additional measurements on the pneumatic side of the device were required. In the second approach a two-stage model was proposed. The first-stage model estimates the value of air supply pressure for the idle run. Its output signal is used in a second-stage model to estimate the blood flow in the normal operation of the device. In this way, a model based solely on the measurement signals available in the control unit was obtained. The achieved accuracy is comparable to all tested control and load conditions and is independent from the level of the device filling. An analysis of the utility of the developed model was carried out due to the purpose of device stroke volume estimation and the automatic detection of the extreme filling conditions. The obtained results have confirmed the utility of the developed model. The final conclusions from the conducted research were formulated and possible directions of the further research on the clinical use of the model were presented.
Thesis file
  • File: 1
    doktorat_A_Siewnicka_MCHTR.pdf
    of 06-11-2017
    4 MB
    Request has been registered

Uniform Resource Identifier
https://repo.pw.edu.pl/info/phd/WUT4b641791d940462aa5bb85b4c1f29aaa/
URN
urn:pw-repo:WUT4b641791d940462aa5bb85b4c1f29aaa

Confirmation
Are you sure?
Report incorrect data on this page