Applying ultrasound beamformers to photoacoustic imaging

While single-element ultrasound (US) transducer-based photoacoustic (PA) systems have provided stunning images, they suffer from the problem of speed: they are limited in scan speed due to inevitable motion artifacts, and limited in imaging speed due to their inability to capture multiple depth images at once. Array transducers offer advantages in electronic scanning and the ability to form an image from few excitations. However, whereas a single element transducer can produce an Aline simply by taking the envelope of the received voltage traces, array transducers depend on beamforming to provide images. In commercial array US systems, the beamforming operation has traditionally been done in hardware, offering speed in exchange for flexibility. PA array systems would benefit from being able to take advantage of commercial US systems to avoid excessive hardware design, but the beamforming presents a challenge beyond that of laser synchronization and US output suppression. Many US array systems offer the ability to adjust the speed of sound (c) to compensate for differences in the mechanical properties of different materials. Typical US beamformers are based on a second-order approximation of delays that are used to refocus the incoming US data, so the optimal choice of c is scaled by √2 rather than two as one might at first surmise. Using this second order approximation along with some image rescaling, we show that US beamformers typical in commercial systems can be adapted for use in PA imaging by adjustment of c, and some image coordinate remapping.