Limits of uncertainty in measured values of embolus-to-blood ratios in dual-frequency TCD recordings due to nonidentical sample volume shapes

Transcranial Doppler (TCD) ultrasound (US) is widely used to detect cerebral embolisation. A drawback of the technique is that it is difficult to distinguish between small gaseous emboli and large solid emboli. One potential solution is to make measurements from blood and emboli at two different frequencies, because the frequency-dependence of ultrasonic backscatter will be different for the two types of emboli. Unfortunately, it is not possible to produce two identical sample volumes at two different frequencies, and this will introduce an uncertainty into any such measurements. Experimental measurements have been made of US fields distorted by temporal bone at 2.0 MHz and 2.5 MHz, and a mathematical model has been developed to examine the uncertainty introduced by the different field shapes at these frequencies. The results suggest that the uncertainties are sufficiently large (on the order of 2 to 4 dB) that a significant percentage of emboli are likely to be misclassified by the dual-frequency method. (E-mail: dhe@le.ac.uk)