Normalizing fractional moving blood volume estimates with power Doppler US: defining a stable intravascular point with the cumulative power distribution function.

Abstract

To normalize the power Doppler ultrasound (US) signal to the expected signal from 100% blood in the calculation of a fractional moving blood volume estimate. To locate the signal from flowing blood with a consistent backscatter coefficient, the authors estimated the knee of the cumulative Doppler power distribution function. They used a flow-tube phantom to test the use of this knee to locate a radial position that would fall into a region of high shear stress and minimal rouleaux formation. They also studied how well the method normalized fractional moving blood volume estimates of the right renal cortex in a volunteer when simulating different body habitus and in a group of six healthy volunteers to estimate variability. Over five flow velocities and over undersaturated to severely oversaturated receiver gains, the calculated flow-tube area was a mean 89% +/- 7 (+/- standard deviation) of a standard. In humans, the technique normalized the fractional moving blood volume estimates over an 8-dB receiver gain variation; the mean +/- standard deviation of fractional moving blood volume estimates for the six volunteers was 37.6% +/- 3.6. Vascularity estimates with power Doppler US are feasible with a normalization scheme based on the cumulative Doppler power distribution function.