Experimental characterization of a vector Doppler system based on a clinical ultrasound scanner.

We have developed a vector Doppler system using a clinical ultrasound scanner with a research interface. In this system, vector Doppler estimation is performed by electronically dividing a linear array transducer into a transmit sub-aperture and two receive sub-apertures. The receive beams are electronically steered, and two velocity components are estimated from echoes received from the beam overlap region. The velocity vector is reconstructed from these two estimates. The goal of this study was to characterize this vector Doppler system in vitro using a string phantom with a pulsatile velocity waveform. We studied the effect of four parameters on the estimation error: beam steering angle, angle of the velocity vector, depth of the scatterer relative to the beam overlap region and the transmit focus depth. Our results show that changing these parameters have minimal effect on the velocity and angle estimates, and robust velocity vector estimates can be obtained under a variety of conditions. The mean velocity error was less than 0.06 x pulse repetition frequency. The velocity estimates are sensitive to the Doppler estimation method. Our results indicate that vector Doppler using a linear array transducer is feasible for a wide range of imaging parameters. Such a system would facilitate the investigation of complex blood flow and tissue motion in human subjects.