Obtaining local SAR and blood perfusion data from temperature measurements: steady state and transient techniques compared.

A series of analyses and experiments was performed to determine the extent that SAR and blood perfusion information can be extracted from steady state temperature values and from transient temperature measurements following a step change in applied power. Multiple local temperature measurements were made in canine thighs heated by 2450 MHZ microwaves to evaluate two parameters: the local absorbed power in the tissue, and the local "effective blood perfusion." The theoretical bases for these calculations are presented in order to identify their underlying assumptions and to obtain a unified basis for comparison of the various calculation methods used by previous investigators. From energy balance considerations it can be shown that the local absorbed power can be obtained from either the rate of increase of temperature following a step increase in power, or from the rate of decrease in temperature immediately following a step decrease in power. These theoretical observations are verified experimentally by comparing the SAR results at fixed positions in canine thighs as calculated from both increasing and decreasing power steps. For decreasing power steps, the resulting decreasing temperature curves can also be used to calculate an effective blood perfusion rate if thermal conduction is included. Alternatively, this same effective blood perfusion rate can be calculated from steady state data. (These two approaches have been used by previous investigators to determine "blood perfusion" values. We have added the modifier "effective" to specifically denote the presence of thermal conduction effects in such perfusion calculations.) From our experimental results and theoretical calculations it appears that differences between the predictions of the two calculation methods arise from changing thermal conduction values during the cooling period of the thermal clearance method. The steady state calculation approach is easier to apply than the washout method, but it requires the additional knowledge of the local SAR value. It is important to realize that effective blood perfusion values calculated using thermal techniques are subject to large errors under conditions where thermal conduction is important, unless this conduction is explicitLy included in the calculation. Such effective blood perfusion values should not be quantitatively compared to values calculated from non-thermal techniques that are not affected by thermal conduction. Unless such conduction effects are known to be negligible, effective perfusion values are only qualitative indicators of the presence of changes in blood perfusion.