Liver disease was recently revealed to be the main cause of death among 35-49 years old people in the UK, with overall mortality rate having quadrupled in the last half century. Causes are related to hepatitis B/C, obesity and alcohol consumption, which can lead to inflammation and fibrosis. One way of assessing liver health is MR-Elastography (MRE), which is a technique that can characterise tissue biomechanical properties by imaging propagating shear waves. Wave propagation is influenced by local tissue structure, but also by tissue deformation. For example, the liver is deformed during respiratory motion, which leads to changes in the apparent stifness. Hence it is clinically relevant to investigate the bias induced by large deformation on wave behaviour. A theory was developed in this regard, which links deformation to its effect on wave propagation. The theory has been incorporated into a new MRE reconstruction technique, with the purpose of correcting the bias in apparent stifness introduced by deformation. To validate this theory, an MRE experiment on poly-vinyl alcohol (PVA) phantoms under a range of large deformations was performed. It was shown that, by combining knowledge of the deformation and material law (here inferred from complementary rheological experiments), the stifness biasing effects of large deformation on small wave propagation can be undone. Similar to the phantom work, MRE experiments were performed on deformed ex vivo
liver samples, which displayed apparent stifness biases. In order to apply the proposed theory, a viscoelastic liver constitutive law was needed. This was obtained from rheological experiments, where tissue samples were examined under a range of combined large deformations (compression and shear strain) and at multiple frequencies. The complex nature of liver tissue structure posed additional challenges when incorporating the knowledge on deformation and tissue law into the MRE reconstruction. This led to a comprehensive analysis which revealed the importance of accurate assessment of deformation metrics for the recovery of intrinsic tissue properties. Lastly, healthy volunteers and patients suffering of liver disease were scanned using MRE. Two states were acquired, at end expiration and end inspiration. For both groups, differences in stifness measurements were observed between the two respiratory states. A comparison of the two groups is discussed, in the context of determining the behaviour of healthy, inflamed and fibrotic liver tissue under deformation.
|Date of Award||1 Mar 2020|
|Supervisor||David Nordsletten (Supervisor) & Ralph Sinkus (Supervisor)|