Estimation of subsurface temperature profiles from infrared measurements during ultrasound ablation

Abstract

Measurement of in situ spatiotemporal temperature profiles would be useful for developing and validating thermal ablation methods and therapy monitoring approaches. Here, finite difference and analytic solutions to Pennes’ bio-heat transfer equation were used to determine spatial correlations between temperature profiles on parallel planes. Time delays and scale factors for correlated profiles were applied to infrared surface-temperature measurements to estimate subsurface temperatures. To test this method, exvivo bovine liver tissue was sonicated by linear image-ablate arrays with 1–6 pulses of 5.0 MHz unfocused (7.5 s, 64.4–92.0 W/cm2 in situ I SPTP) or focused (1 s, 562.7–799.6 W/cm2 in situ I SPTP, focus depth 10 mm) ultrasound. Temperature was measured on the liver surface by an infrared camera at 1 fps and extrapolated to the imaging/ablation plane, 3 mm below the surface. Echo decorrelation maps were computed from pulse-echo signals captured at 118 fps during 5.0 s rest periods beginning 1.1 s after each sonication pulse. Tissue samples were frozen at −80 °C, sectioned, vitally stained, imaged, and segmented for analysis. Estimated thermal dose profiles showed correspondence with segmented tissue histology, while thresholded temperature profiles corresponded with measured echo decorrelation. These results suggest utility of this method for thermal ablation research.