A multi-gate time-of-flight technique for estimation of temperature distribution in heated tissue: theory and computer simulation.

Non-invasive determination of temperature distribution in biological media is important in many heating-related studies, such as thermal treatment. In this paper, we present an in vitro ultrasound technique for estimation of temperature distribution in heated tissue. Our technique consists of two major steps: (1) using multiple time gates to track echo signals scattered from tissue regions at different depths; (2) estimating temperature distribution based on heating-induced changes of arrival times of echo signals scattered from the targeted tissue regions. We use the conventional cross-correlation approach to track echoes. For temperature estimation, we have developed an iterative method that takes into account the influences of thermal expansion and heating-induced change in the speed of sound on the time of flight. We have introduced a concept of thermal sensitivity of the time of flight and used it to derive a theoretical formula that relates the achievable accuracy on the estimation of tissue temperature to seven parameters. The seven parameters are tissue thermal sensitivity of the time of flight, signal-to-noise ratio, bandwidth and center frequency of the signal, degree of signal decorrelation induced by changes in tissue physical properties during tissue heating, and widths and spacing of the time gates. We tested our technique by computer simulation, using a random discrete scatterer model and temperature distribution data acquired in our laser heating experiments on prostate tissue of live dog. Simulation results showed that our technique could accurately estimate the temperature distribution in the heated tissue. Our technique is fast in terms of computation and could be used as a research tool for in vitro real-time monitoring of temperature distribution in tissue under hyperthermal heating.