Benjamin Tyrrell

It has been demonstrated that ultrasound is excellent for thermometry because it is non-invasive, cost effective, and compact. To gather quantitative information about specific areas of tissue, like the specific temperature in degrees Celsius using ultrasound, multiband spectral analysis algorithms are used to process the collected data. Currently the spectral analysis algorithms are based on broadband ultrasound echoes from the target region of tissue without regard to the scattering model, a mathematical model frequently used to model tissue response to ultrasound. Recently, Professor Ebbini’s group has been investigating multiband signal processing of ultrasound imaging data matched to the underlying scattering models. This approach is likely to improve the quality of temperature estimation, but its advantages over the conventional broadband approach have not been investigated. In addition, the computational cost of the parallel multiband spectral estimation needs to be addressed in comparison with the conventional multiband approach. This projects goal is to modify and implement parallel spectral analysis algorithms on an FPGA, and to determine how the performance of the parallel algorithms implemented on an FPGA compares to implementation on Simulink. Parallel implementation of the spectral analysis algorithms improves the robustness of temperature estimation without compromising the real-time advantages of ultrasound thermometry.