Computational evaluation of the thermal resolution on ultrasonic thermometry using an improved spectral analysis of echo-signal patterns modeled ad-hoc

This work provides a quantitative specific evaluation of the spectral analysis performance when applied to ultrasonic multiecho-signals, in order to extract information from them, for non-invasive temperature estimation purposes, inside a biological tissue. This specific evaluation was oriented to assess the thermal resolution provided by the application of resolution improvements in this tool. Simulated signals, based on a numerical model of an ultrasonic echo-trace acquired from biological media, were used to perform tests considering several parameters under a controlled situation. One of the most important parameters to be controlled, in calculations for this simulated approach, was the minimum temperature rise considered for evaluation: 0.05°C. Results about harmonics behavior in the here obtained High-Resolution Power Spectral Densities (HR-PSD) are shown and compared with the theoretically expected shift calculations. Commitments between ultrasonic parameters, biological tissues characteristics and signal processing requirements, are discussed for a real experiment, and some conclusions about results are presented.