Blind source separation-based adaptive filtering of physiological and ARFI-induced tissue, blood, and cyst fluid motion, in-vivo

In the context of complex tissue motion patterns including physiological and acoustic radiation force impulse (ARFI) -induced tissue, blood, and cyst fluid motion, the same adaptive blind source separation (BSS)-based filtering method is effective for four diverse applications: 1) clutter filtering prior to blood velocity estimation, 2) extracting axial velocity components from noisy velocity measurements given large flow angles, 3) reducing noise in measured ARFI displacement profiles, and 4) complex phase clutter filtering to measure ARFI-induced cyst streaming. RF data was collected using ARFI imaging beam sequences from the carotid arteries of healthy volunteers and from female subjects (recruited under IRB guidance), each presenting one breast lesion. Biopsy determined that one lesion was a fluid-filled cyst and the other was a malignant mass. In the examined arteries, velocity profiles are parabolic in shape, with nearly zero axial velocity beyond the vessel lumen following BSS clutter filtering and BSS extraction of small axial velocity components. The method maintains peak displacement (1.81 to 21.5μ m) and time to 63% recovery (1.38 to 2.20 ms) parameters after BSS noise reduction in ARFI displacement profiles. Colorflow images generated after BSS complex phase clutter rejection reveal ARFI-induced streaming in the fluid-filled cyst but no streaming in the malignant lesion. Successful filtering for four diverse purposes illustrates the breadth of applications for BSS adaptive filtering in the clinical imaging environment.