Investigating the change in point spread function and resolution of random apodization passive cavitation images

Abstract

Passive cavitation imaging is an important technique in measuring cavitation dynamics, localizing concomitant bioeffects, and guiding various ultrasound therapies. The canonical algorithm for passive cavitation imaging operates via a delay, sum, and integrate method. With standard diagnostic arrays, this method yields poor axial resolution. We hypothesized that using a random apodization, minimum projection compounding technique could overcome this limitation. The random apodization approach randomly selected only half of the elements to be included in the delay, sum, and integrate algorithm to form a PCI. This process was repeated 30 times using the same data set but a different random subset of half the elements to create 30 images. The final image was obtained via a minimum intensity projection across the 30 images. The improvement of image quality can be tracked by comparing the point spread function (PSF), and the distance at which two cavitation sources can still be resolved, relative to when using the standard algorithm. The PSF reduced by 86% ± 9% when the random apodization was applied. Lateral and axial resolution exhibited qualitative changes associated with a lower PSF, but no quantitative change in resolution, based on the Rayleigh criteria, was observed.