Phase aberration correction on a 3D ultrasound scanner using RF speckle from moving targets
In 1992, Zhao et al. described a variation of the speckle-brightness algorithm for phase correction on moving targets [1, 2]. By subtracting two beam-summed consecutive interrogations of a region of interest, they were able to remove effects of reverberation and strong targets from the region of interest. We have modified this approach of image subtraction for use on radio-frequency (RF) signals from the Duke/Volumetrics Medical Imaging real-time 3D ultrasound scanner. We have devised a system to use the 16:1 parallel receive processing capabilities of the 3D scanner to simultaneously acquire RF data from 16 separate receive subapertures. We then modify the correlation model matrix to allow correlations only between simultaneous RF lines. By using overlapping sub-apertures and the least-mean-squares approach, we compute an aberration profile with phase closure for the entire aperture. To test this method, we have performed experiments using electronic and physical aberrators. Using an electronic aberrator, root-mean-square strength (RMS) = 80 ns and autocorrelation length = 3.4 mm, and using echoes from a flowing corn-starch/water mixture, we saw marked improvement in color Doppler visualization as well as an 11% increase in brightness, a 47% increase in contrast to speckle ratio (CSR), and a 17% increase in lesion contrast. Using a physical aberrator (plastic casting of a skull bone), RMS strength = 31 ns autocorrelation length = 3.3 mm, we saw a 5% increase in brightness, an 11% increase in CSR, and a 26% increase in lesion contrast. We believe that this technique will be useful in transcranial ultrasound with contrast agents, since it will allow us to use the strong echoes from the microbubbles for correction and since it will mediate the effects of reverberations due to the skull bone. We also include images from the first real-time 3D transcranial color Doppler ultrasound with contrast agents.
