| dc.description.abstract | Head and neck cancer treatments involve surgery, radiation, or chemotherapy which leaves survivors with disrupted lymphatic structures and soft tissue damage that can lead to secondary lymphedema and fibrosis (LEF). Ultrasound elasticity imaging could provide early detection of fibrosis that may not be noticed in a physical examination of the skin, enable comparison of stiffness across different patients, and inform the need for therapy before fibrosis progression. However, skin elasticity imaging has been difficult due to the skin's thin nature and complicated surrounding structures and has been difficult to validate across different patients. Acoustic radiation force (ARF)-based elasticity imaging displaces tissue at a specified region and determines information about the mechanical properties by monitoring the displacement using ultrasound. It is also less operator dependent and allows for imaging deeper muscle regions that may be affected by LEF that could not as easily be quantified using a surface technique or palpitation. Shear wave elasticity imaging (SWEI) measures the speed of shear wave displacements arriving at lateral locations from an applied ARF, but is biased based on the skin thickness and does not have an analytical model to estimate shear modulus. To avoid measuring the velocity of a complex shear wave in skin, we developed an ARF-based elasticity technique that measures displacement at the region of excitation and uses a simulated look-up table of responses from varying elasticities to quantify elasticity based on the time-to-peak displacement for a certain ARF push. In order to measure displacement at the region of excitation, we applied an advanced Bayesian displacement estimator. We validated the on-axis quantitative elasticity technique in simulations and homogeneous elasticity phantoms. We have also optimized the Bayesian displacement estimator to compute displacement estimates two orders of magnitude faster by solving for analytical derivatives. Finally, we have shown preliminary results of the on-axis elasticity method in layered, skin-mimicking phantoms and compared the results to SWEI. We demonstrate the ability to use a simulated look-up table on real experimental data and get quantitative elasticity estimates on-axis to the ARF. | |
