A Metric and Method for Calculating Energy-Preserving Spatiotemporal Cavitation Dose: Cavitational Radiated Energy Density (CRED).

Abstract

Passive Acoustic Mapping (PAM) is rapidly emerging as a ubiquitous tool for real-time localization and monitoring of therapeutic ultrasound treatments involving cavitation in the context of safety or efficacy. The ability of PAM to spatially quantify and resolve cavitation activity offers a unique opportunity to correlate the energy of cavitation phenomena with locally observed bioeffects. We aim to develop methods of measuring and reporting spatio-temporally varying cavitation energies that are energy-preserving, device-independent, and adequately normalized to the volume of tissue being affected by the reported cavitation activity. We introduce a new metric: cavitational radiated energy density (CRED), defined as the total energy of radiated acoustic emissions over a volume, divided by that volume. We show that image blurring caused by the instrumentation-dependent point spread function introduces significant image artefacts and CRED errors when using existing conventional and adaptive beamformers. We therefore introduce a novel PAM algorithm that utilizes the Lucy-Richardson deconvolution (LRD) technique to compensate for the point-spread function. In single and multi-source simulations and experiments, LRD provides cavitation maps with reduced tail artefacts and improved energy estimates with <5% increase in computation time relative to a conventional beamformer. We demonstrate the quantitative significance of CRED and the potential of the LRD-PAM algorithm to enable energy-preserving, quantitative, real-time monitoring of cavitation-based therapies independently of the instrumentation employed.