Ultrasonic non-contrast perfusion imaging with adaptive demodulation

Abstract

Ultrasonic non-contrast perfusion imaging remains challenging due to spectral broadening of the tissue clutter signal caused by patient and sonographer hand motion. Simply, the velocity of the slowest moving blood has similar spectral support to the moving tissue. To address this problem, we developed an adaptive demodulation (AD) scheme to suppress the bandwidth of tissue prior to high-pass filtering. The method works by directly estimating the modulation imposed by tissue motion, and then removing that motion from the signal. Our initial implementation used single plane wave power Doppler imaging sequence combined with a conventional high-pass IIR tissue filter. However, other recent advancements in beamforming and tissue filtering have been proposed for improved slow-flow imaging, including coherent flow power Doppler (CFPD) and singular value decomposition (SVD). Here, we aim to evaluate AD separately as well as in comparison and in conjunction with improvements in beamforming and filtering using simulations and an in vivo muscle contraction experiment. We show that simulated blood-to-background SNR is highest when using AD + CFPD and a 100 ms ensemble, which resulted in a 9.88 dB increase in SNR compared to CFPD by itself. Additionally, AD + SVD resulted in a 54.6% increase in mean power within the in vivo muscle after contraction compared to a 6.84% increase with AD and a conventional IIR filter.