Ultrasonic estimation of flow volume has conventionally been made by multiplying the cross-sectional area of a vessel by
maximum flow velocity through that area as determined by the pulsed Doppler method, and then by applying a correction
coefficient. This method is inaccurate, however, because it is often difficult to determine precise vessel diameter, and because
it is not always appropriate to use a coefficient that has been obtained from animal experiments. Here we used a phantom to
examine the validity of a new flow volume measurement with color flow signals. The method first entails formation of a velocity
profile on a line that passes through the center of a vessel and that is perpendicular to its long axis. Once the cross section of
the vessel has been deemed circular, the profile is developed into a two-dimensional distribution. In doing this, each half of the
linear profile is treated differently to avoid any error that might be caused by an eccentric peak. Flow volume at a given point
in time can thus be calculated by multiplying each velocity by the number of pixels in that velocity and by summing the products.
Flow volume per cardiac cycle can then be calculated by integrating flow volumes at amount for the cycle. Results thus obtained
correlated closely with those of the actual flow volume through the phantom (r=1.00). Because it is rapid and does not require
measurement of a laminar diameter to assume a velocity profile, and does not rely on application of a correction coefficient, the
present method appears to be more useful than the pulsed Doppler method in determining volume of blood flow.