Earth's field nuclear magnetic resonance measurement of blood pressure metrics in a compliant vessel model
A dissertation submitted to the Faculty of Engineering and the Built Environment, University of Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science in Engineering 2019 The following experimental design and setup have been created by the author and supe...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10539/28375 |
| Summary: | A dissertation submitted to the Faculty of Engineering and the Built Environment, University of Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science in Engineering 2019 The following experimental design and setup have been created by the author and supervisor. Current methodologies for non-invasive blood pressure measurement are flawed due to the subjective nature of measurement. Experimentation has been performed to determine to what degree relevant cardiovascular metrics can be noninvasively extracted from idealised software and hardware models of a human upper limb. The physical model is constructed to replicate the dimensions of major arteries and veins of the upper limb and is filled with water to approximate the vessels in the upper arm. Static water tests are performed to calibrate the model in order to obtain appropriate parameter values. Pulse and collect experiments have been performed on volumes of water flowing through cardiovascular analogues that are polarised upstream by 300 mT annular Halbach arrays. Testing is performed using the TerraNova-MRI Earth’s Field Nuclear Magnetic Resonance apparatus from Magritek. Signal models obtained from literature have been modified to determine the effect of vessel compliance on flow measurements. A direct positive correlation (p= 0.88) between signal level received and the increasing pressure and volume of water within the model is observed. The average standard error between measurements is 0.04 µV, or 10% of the magnitude of the smallest measured signal. The relative error between theoretical and measured signals is 30%. The relative error is unreliable as the theoretical model does not include the effects of noise. The compliance of the model’s vessel analogues is estimated to be 1.48 ml/mmHg compared to normal healthy young human compliance values of ~ 2 ml/mmHg. An empirical relationship between the pressure and signal captured by the apparatus is found. The problem is ill-posed with many possible ... |
|---|