Wall filtering challenges in two-dimensional vector velocity estimation

To formulate a clinically viable methodology for tracking blood flow in two dimensions using ultrasound, one must address some challenges posed by the physical limitations of the imaging modality. At the frequencies of interest in medical diagnostic imaging, the echos from blood are much smaller (often 40-60 dB) than those from the surrounding vessel walls. A blood velocity estimation technique must isolate the blood echo from that of the wall, and that is most often accomplished through the use of wall filters. The conventional application of wall filters so two-dimensional velocity estimation is aggravated. However, by the fact that the azimuthal spatial frequency content of conventional two-dimensional ultrasound scans is typically 5-10 times smaller than the axial spatial frequency content. In this paper, we pose the problem of isolating the blood signal within the composite blood/wall/noise signal received by the imaging system and estimate the effectiveness of wall filters to eliminate stationary echos from both the axial and azimuthal components of motion