Rapid volumetric photoacoustic tomographic imaging with a Fabry-Perot ultrasound sensor

Rapid volumetric photoacoustic tomographic imaging with a Fabry-Perot ultrasound sensor depicts peripheral arteries and microvascular vasomotor responses to thermal stimuli
Peripheral arterial disease (PAD) is a common, important condition, affecting approximately 27 million individuals across the USA and Europe.

Most patients with PAD have generalised atherosclerosis throughout the cardiovascular system, including significant coronary artery disease in one-third, meaning that cardiovascular deaths are increased six-fold. Furthermore, severe PAD itself causes significant local morbidity via tissue necrosis and ultimately amputation or even death. Healthcare costs are substantial (>$4.5 billion annually in the USA alone), similar to cerebrovascular disease and cardiac failure. Therefore, prompt identification and treatment of PAD are important to patients, healthcare services, and wider society.

PAD involves both large vessels (LVs) and small vessels (SVs) beyond named arterial branches. Whereas LVs are well depicted by existing techniques, primarily ultrasonography (US) and computed tomographic or magnetic resonance angiography (CTA/MRA), limited spatial resolution precludes meaningful assessment of SV. This is problematic, because SV-PAD is closely associated with diabetes mellitus, which is increasing in incidence globally. Accordingly, there is clear need to develop newer techniques to accurately assess SV-PAD non-invasively.

One possible solution is photoacoustic (optoacoustic) imaging (PAI), which exploits the photoacoustic (PA) effect, whereby laser illumination of tissue provokes broadband ultrasound emission. These ultrasound waves can be reconstructed into high-resolution images, based on the optical absorption properties of the tissue. PAI is particularly suitable for vascular imaging because haemoglobin has strong optical absorption, maximising ultrasound emission and image contrast.

However, imaging at clinically relevant depths can be problematic because normal tissues have high optical and acoustic attenuation, mandating highly sensitive ultrasound detectors. Furthermore, it is challenging to simultaneously illuminate tissue and detect ultrasound without the ultrasound detectors obscuring the laser light. We have developed a PA system that uses a Fabry-Perot (FP) interferometer as the ultrasound sensor.

This is optically transparent to the excitation laser, avoiding such obscuration. Furthermore, the FP sensor outperforms conventionally used piezoelectric ultrasound sensors in sensitivity, bandwidth, and element size, yielding improving 3D image resolution and contrast for sub-centimetre scale vascular imaging.

Several previous PAI studies have depicted human vasculature successfully, although most only imaged superficial skin vessels, or while investigating dermatological disease, although clear proof-of-concept of deeper vessel imaging has been reported, of both larger named arterial vessels and within organs such as the breast. A further recent article described high-resolution two-dimensional peripheral foot vessel imaging using a handheld probe and a concave ultrasound detector array. Further ex vivo animal model work has shown that volumetric PAI can successfully depict subcutaneous veins during endovenous laser therapy.

However, PAI has not yet been shown capable of detecting vascular changes induced by either normal physiology or disease in humans, nor has it been subject to appropriately powered, prospective clinical studies with interpreter blinding and pre-specified endpoints aimed at clinical validation, as befits development of a new imaging technology.

Since SVs are readily able to vasoconstrict and vasodilate in response to cold and heat respectively, they are an excellent experimental test bed for evaluation of PAI-based vascular imaging, serving as a safe, laboratory-controlled paradigm for the constriction of the SVs seen in atherosclerotic disease. We therefore wished to determine if our FP sensor-based PAI system can successfully depict (1) peripheral leg arteries and (2) microvascular circulatory changes in response to thermal stimuli as a reliable, controllable means of safely inducing SV vasoconstriction and vasodilation.