1522: Pre-Clinical Studies of the Measurement of Volumetric Blood Flow

Abstract

Verify a previously reported non–invasive, angle independent, 3-D Doppler ultrasound method for measurement of volumetric blood flow in vivo with comparison to an accepted, but invasive, standard method. A GE Logiq 9 with a 4D10L probe acquired 3D Doppler of the femoral and carotid arteries in a canine model. Color flow volume spanned a 6 by 4 cm axial-lateral field of view with a 29° elevational sweep, encompassing the vessel in <7 s. Constant depth planes were used to integrate color flow pixels and power Doppler data was used to correct partial volume effects. An artificial stenosis varied the ambient volume flow by restricting the vessel using a pneumatic cuff. An ultrasonic blood flowmeter with flow probe around the vessel served as the invasive gold standard. Unrestricted, bidirectional, flow was measured as high as 400 mL/min with flow restrictions as low as 20 mL/min. All flow rate estimates (N=23 from 4 animals) were plotted against the gold standard yielding a general line fit of y=0.926x-0.87 (R2=0.95). The percent-error distribution was computed and fitted to a gaussian function with a mean=-7.04% and sigma=9.52%. A secondary curve fit with the slope = 1 and the intercept = 0 yielded R2=0.93. Direct comparison of volume flow rate estimates using volumetric color Doppler and an independent standard showed good accuracy under in vivo pulsatile blood flow conditions. Subsequent comparisons under clinical conditions will demonstrate possible applications.