Our purpose was to validate in vitro a kinetic flow model based on microbubble signal decay curve. Using a 3.5 MHz transducer and phase-inversion (1.8 MHz central transmit frequency), a renal dialysis cartridge oriented vertically was imaged in the transverse plane as 1:1000 dilution of AF0150 was infused at 50, 100, 200, 300 and 400 mL/min. Ten gray-scale images were acquired at each infusion rate using 2.5, 5 and 10 frames/s at 100%, 40%, 15% or 1% of maximum transmit power. Video-intensity measured on each 10 images was fit to a kinetic model using Sigma Plot that yielded microbubble concentration, velocity and destruction per frame. These were correlated with the experimental conditions. At 100% power, video-intensity on the first frame (microbubble concentration at equilibrium) was similar for all flow and frame rates. The model fit the experimental data for all flows at 10 frames/s and for flows lower than 400 and 100 mL/min at 5 frames/s and 2.5 frames/s, respectively. The calculated flow was similar to the experimental flow rates, regardless of technique (r(2) = 0.98). Microbubble fraction destroyed per frame was similar for all flow and frame rates and increased linearly with transmit power (r(2) > 0.98). These results suggest that using appropriate power and frame rate for a given flow rate, estimates of fractional blood volume, flow and destruction fraction can be calculated from the decay curve using 10 frames that can be acquired in 1 to 4 s.