Complex blind source separation for acoustic radiation force impulse imaging in the peripheral vasculature, in vivo

Complex blind source separation (CBSS) is presented in an application to adaptive clutter filtering for acoustic radiation force impulse (ARFI) imaging in the peripheral vasculature, in vivo. In addition to the conventional challenges of wall filtering for Doppler imaging in peripheral vessels, ARFI imaging introduces the complication of isolating ARFI-induced blood motion from ARFI-induced vessel wall and surrounding elastic tissue motion. Raw RF data was collected using ARFI M-mode imaging sequences from a pulsatile flow vessel phantom, the jugular vein of a male volunteer performing the Valsalva maneuver, and the carotid artery of a female volunteer. The imaginary component of the RF data was computed via the Hilbert transform, and CBSS clutter filtering was performed on the isolated blood signal. Following CBSS filtering, phase-shift estimation reveals radiation force-induced axial blood velocities that peak at 10-25 cm/s. These results demonstrate that CBSS sufficiently rejects elastic tissue and vessel wall signal components to allow for measurement of ARFI-induced axial blood velocities in both phantom and clinical investigation.