Medical Ultrasound Aberration Correction via Acoustic Droplet Vaporization and Time-Reversal Acoustics.

Abstract

Time-reversal acoustics (TRA) is an alternative to standard focusing approaches in medical ultrasound. TRA records backscattered ultrasound and time-reverses it (Ψ(t) -> Ψ(−t)). Due to the time-reversal invariance of the lossless wave equation, a transmitted time-reversed signal will focus back to the scattering source. It has been proposed that acoustic droplet vaporization (ADV) can be used to generate point-scatterers for TRA focusing. ADV is a process where micron-sized liquid droplets are phase-transitioned into gas bubbles via an acoustic wave. The feasibility of performing medical ultrasound aberration correction using TRA and ADV is explored in three different contexts. The first is transcranial transmit aberration correction. It is demonstrated that stable gas-bubbles can be produced transcranially. Additionally, it is demonstrated that time-reversal focusing can be used to correct for transcranial aberrations with a gas bubble. Aberration correction was performed using a synthetic aperture approach. Under the conditions described, time-reversal aberration correction resulted in a linear gain of 1.9 ± 0.3. This demonstration is particularly relevant for therapeutic applications. The second context is aberration correction on receive. A synthetic aperture algorithm is implemented with the decomposition of the time-reversal operator algorithm. It is shown that this combination can produce aberration corrected images with a clinically relevant ultrasound unit. Aberration is induced electronically to mimic a near-field aberrator. Various dependences on imaging parameters and reconstruction are explored. This demonstration is particularly relevant for diagnostic applications. The third context is the development of a theoretical shot noise model to statistically describe the output of a multiple-scattering time-reversal experiment using arbitrary inputs and windowing. It is found that the largest noise contribution depends on the windowing and can occur at times outside the main lobe. A common set of parameters is applied to the general result and it is seen that as the duration of the input function increases, the signal-to-noise ratio (SNR) decreases (independent of signal bandwidth). It is also seen that longer persisting impulse responses result in increased main lobe amplitudes and SNR.