| dc.description.abstract | Ultrasound elastography (USE) is a well-established technique used to non-invasively map tissue stiffness. More recently, a novel branch of USE, called poroelastography, has been proposed, which aims at estimating the poroelastic response of a tissue. The hypothesis of poroelastography is that underlying pathology, such as cancer, alters the fluid transport mechanisms in the tissue, which alters its poroelastic response. Poroelastography techniques estimate the temporal behavior of axial strain and effective Poisson’s ratio (EPR) of a poroelastic media under compression. While these methods have been successfully proven to estimate the poroelastic response of poroelastic media and tissues, there are some major limitations which need to be addressed. There is a lack of tissue mimicking poroelastic phantoms with tunable poroelastic properties. Also, while US poroelastography techniques aim to estimate the temporal behavior of EPR, the estimation of the EPR using USE is challenging due to known image quality limitations of lateral strain elastography techniques. Also, the relationship between the spatio-temporal behavior of interstitial fluid pressure (IFP), axial strain and EPR is currently not known. IFP is an important parameter, which is known to help in the diagnosis and characterization of soft tissue cancers. In this dissertation, I attempt to solve some of these current issues in the field of ultrasound poroelastography imaging, taking the field one step forward.
In this work I investigated the use of polyacrylamide gel for creating new class of phantoms for poroelastography. Results of the study indicate that by using polyacrylamide gel, tissue mimicking poroelastic phantoms with controlled fluid flow can be generated. This new class of phantom material can be used to conceptualize and validate techniques in poroelastography and for temporal ultrasound elastography imaging in general. For reliable estimation of EPR, I proposed a new US poroelastography technique which uses two ultrasound transducers. By using a simulation module, image quality from this new technique was statistically compared from previously used methods. The feasibility to experimentally estimate EPR using two-transducers was also demonstrated. The results indicate that the two-transducer poroelastography technique has superior image quality than the existing methods. In this study, I also studied the spatio-temporal behavior of axial strain, EPR and IFP with change in interstitial permeability. A 2D poroelastic finite element model was used, followed by an ultrasound simulation algorithm. The temporal behavior was estimated by finding the time constant of the temporal curves. Results of the analysis indicate that increased IFP creates a new contrast mechanism in both the axial strain and EPR elastograms. Also, the spatio-temporal patterns of IFP are closely related to that of the EPR and, hence, a reliable estimation of EPR may aid in a non-invasive assessment of underlying IFP. | en |
