On the reproducibility of the average gray-level for noninvasive temperature estimation in the hyperthermia range

The thermal treatment of deep-sited tissues requires temperature feedback by minimally invasive or even by completely noninvasive methods. Ultrasound has been pointed as a viable technology for noninvasive temperature assessment. More precisely, parameters from B-Mode images were pointed to vary monotonically with temperature and able to estimate temperature with appropriate resolution. In this paper, a bovine muscle sample is subjected to several heating and cooling periods in order to analyze how the average gray-level from B-Mode images is reproducible with temperature. This is an important aspect concerning the real-time application on a clinical setting. Results point that for formaldehyde fixed tissue the average gray-level was reproducible with temperature. In non-fixed (unstable) media reproducibility was achieved by considering relative (temperature and average gray-level) changes to the starting of a particular heating/cooling phase.