PAT: Photoacoustic Tomography; PACT: Photoacoustic Computed Tomography; PAM: Photoacoustic Microscopy; PATER: Photoacoustic Topography through an Ergodic Relay; ER: Ergodic Relay;  FOV: Field-of-view

Photoacoustic imaging or photoacoustic tomography (PAT), also known as optoacoustic tomography, is an emerging hybrid imaging modality that combines the advantages of both optical and ultrasonic imaging. It inherits the high optical contrast from optical imaging and offers high spatial resolution in deep tissue based on its acoustic detection [1,2]. Photoacoustic signals are produced by illuminating a tissue of interest with short laser pulses (typically in ~ns). Some molecules within the tissue absorb the incident photons regardless scattered or not and convert the light energy into acoustic waves, also called PA waves. Ultrasonic transducers measure the Photoacoustic (PA) waves, and digital reconstruction yields images that map optical absorption inside the tissue. As biological tissues scatter incident photons heavily, conventional optical imaging cannot provide high-resolution imaging in deep tissue. By converting photons into phonons and detecting the acoustic waves for image formation, PAT offers superior spatial resolution at depths inside the tissue. Moreover, by tuning the excitation wavelengths to match the optical absorption signatures of the molecules to be imaged, PAT has imaged hemoglobin [3-6], melanin [7,8], DNA/RNA [9,10], lipid [11-13], myoglobin [14], organic dyes [15,16], nanoparticles [17-23], and genetically encoded proteins [24-29] with high contrast.

PAT has two major incarnations–photoacoustic computed tomography (PACT) and photoacoustic microscopy (PAM) based on their signal detection and image formation methods [30-32]. In PAM, the generated PA signals are detected by raster scanning of a single-element ultrasonic transducer for serial acquisition [33,34]. While PACT employs multi-element transducer arrays to detect the PA waves in parallel [35,36]. In the former case, point-by-point mechanical scanning limits the imaging throughput. In the latter case, multi-element transducer arrays with multi-channel amplifiers and digitizers significantly increase the system’s complexity, cost, and size. PAT urgently demands a new implementation using a much-simplified setup with high throughput. Recently, photoacoustic topography through an ergodic relay (PATER) has been developed to offer high-throughput snapshot wide-field imaging with a single element detector [37,38]. PATER demonstrated a snapshot wide-field imaging with a frame rate up to 2 kHz. Taking advantage of such high imaging speed, PATER directly visualized blood oxygen level changes in the mouse brain and achieved fast biometric authentication by matching vascular patterns [39].

The key component of PATER is the ergodic relay (ER), an acoustic waveguide that encodes acoustic waves from the input points to an output point with distinct amplitudes and time of flight. Previous studies have shown that a right-angle optical prism can serve as an ER for PATER [37,38]. PATER requires two acquisition steps: calibration and wide-field imaging, as shown in Figure 1. During calibration, a focused laser beam is used to scan across the entire field-of-view (FOV) of the object (Figure 1a). The laser pulse width (~5 ns) is much narrower than the central period of the ultrasonic transducer (~100 ns), and the focused beam diameter (~10 µm) is much smaller than the central acoustic wavelength (~600 µm), then each PA wave excited by the focused laser pulse can be treated as a spatiotemporal delta function of the whole system. Thus, the system impulse response at each pixel can be calibrated by raster scanning the laser beam over the FOV, and the system matrix K  can be obtained. During wide-field imaging, a broad beam uniformly illuminates the object (Figure 1b). The PA signals from all pixels of the entire illumination volume, denoted as s , are acquired in parallel using only a single-element ultrasonic detector, permitting snapshot wide-field imaging. The wide-field measurement can be expressed as

s= KP                                    (1)

where P is the wide-field images, which can be reconstructed by solving the inverse problem of Equation 1. The reconstructed wide-field images map the optical absorption changes of the object. With a high-repetition-rate laser for broad illumination, PATER provides snapshot imaging and captures fast dynamics with a sub-millisecond temporal resolution over a large FOV.

High-speed monitoring of blood oxygen changes in the brain

Inside our body, the hemoglobin within red blood cells binds with oxygen and carries it through the bloodstream to support cellular activities. It is crucial to monitor the blood oxygen level as it is a vital sign. To demonstrate the snapshot wide-field imaging, PATER first imaged the changes in blood oxygen saturation (sO₂) in a mouse brain in vivo [39]. As a demonstration, sO₂ inside the mouse brain was artificially altered via oxygen challenges, where oxygen concentration of inhalation gas supplied to the mouse was manipulated. Initially, the mouse inhaled a mixture of 95% oxygen and 5% nitrogen. After calibration, the mouse brain vasculature was first imaged with a broad laser illumination at 620 nm through the intact skull (Figure 2a). During the oxygen challenge, a mixture of 5% oxygen and 95% nitrogen was supplied to the mouse for 3 minutes. Then the inhalation gas mixture was then changed back to the initial concentration. The time course of the PA signals during the oxygen challenges was shown in Figure 2b. Upon two different wavelengths (532 nm and 620 nm) illumination, the sO₂ map of the cortical vessels was also imaged before (normoxia) and after (hypoxia) the oxygen challenges, as shown in Figure 2c. Due to the simplified system configuration and high-throughput, PATER offers an alternative solution to pulse oximeters, which provide only the averaged sO₂ level of the arteries. In contrast, PATER can yield real-time images of the sO₂ distribution of a blood vessel network.

Fast matching of vascular patterns in vivo

Biometric authentication that employs unique biological characteristics of individuals to verify their identities is widely used in our daily life. Under-the-skin biological characteristics, such as vascular patterns, are more secure and reliable in identifying an individual than surface characteristics such as fingerprints. Because they are not directly exposed and contain in vivo physiological features—such as blood flow and oxygenation—that cannot be readily replicated by others. Biometric authentication based on under-the-skin biological characteristics hold great promise; however, they require high-speed processing and high accuracy for reliable usage. To demonstrate the capability of fast and reliable biometric authentication, PATER achieved fast differentiation of vascular patterns.

During the demonstration, one mouse was mounted in a stereotaxic frame, and the cortical vasculature was imaged in calibration mode (Figure 3a). During the wide-field imaging, the mouse was first detached from the ER, then reattached to the same position. The same procedure was repeated for a second mouse. The wide-field images of the first mouse’s vasculature were reconstructed using the calibration data of both itself and the second mouse. As a result, the wide-field image reconstructed from the calibration data of itself (matched) recovered the original vasculature, while the wide-field image reconstructed from the calibration data of the second mouse (mismatched) revealed no visible features (Figure 3b). The correlation coefficients between the calibration images and the wide-field reconstruction images were calculated, as shown in Figure 3c. The wide-field images reconstructed with the matched calibration data show much higher correlation than those reconstructed from the mismatched calibration data. The vasculature is also recovered after being repositioned to the ER.

The small size, low cost, and simplified system setup of PATER make it a feasible solution for portable applications, such as a wearable device to monitor vital signs in patients. Moreover, PATER can both identify vessel patterns and quantify functional dynamics with high throughput, and it promises comprehensive, secure, and robust biometric authentication.