This PDF file contains the front matter associated with SPIE Proceedings Volume 11960 including the Title Page, Copyright information, and Table of Contents.

Peripheral vascular disease (PVD) is a narrowing of the arteries in the extremities that might result in amputation when left untreated. The conventional ways to diagnose PVD are vascular imaging techniques such as ultrasound (US) Doppler, computed tomography (CT) angiography, and magnetic resonance angiography, which generally can only visualize large vessels or require contrast agents. Photoacoustic (PA) imaging could be a promising agent-free, highresolution modality to visualize vessels and be applied in various clinical studies. We have previously demonstrated an agent-free multi-structural photoacoustic (PA) and US peripheral angiography technique using contour scanning. To reduce the effect of the optical fluence, we compensated the PA images using the background PA signals that represent optical attenuation in the tissue. To evaluate its capability to provide functional information of the vasculature, we demonstrated an intentional blood flow occlusion with a pressure cuff and observed the changes in the quantification metrics. From the experiments, we confirmed the potential of the technique for providing a diagnostic evaluation of the peripheral vasculature.

Optoacoustic images are often afflicted with distortions and artifacts corresponding to system limitations, including limited-view tomographic data. We developed a convolutional neural network (CNN) approach for optoacoustic image quality enhancement combining training on both time-resolved signals and tomographic reconstructions. Reference human finger data for training the CNN were recorded using a full-ring array system with optimal tomographic coverage. The reconstructions were further refined with a dedicated algorithm that minimizes acoustic reflection artifacts induced by acoustically mismatch structures, such as bones. The combined methodology is shown to outperform other CNN-based methods solely operating on image-domain data.

To evaluate the utility of PA in imaging the GI tract in large animals, we conducted a feasibility study using the porcine gastric wall ex vivo. We used fresh pig stomachs within hours of sacrifice and successfully acquired multispectral images of the vasculature of the porcine gastric wall layer ex vivo. Imaging proceeded with samples of the entire stomach layer, including the mucosa, submucosa, muscularis externa, and serosa. It was possible to acquire the vascular signal up to 1.9 mm depth from the mucosal surface, which could cover the entire mucosa and submucosa.

Accurate visualization of biological events at scales in the order of seconds requires high frame rate acquisition of image data from living tissues. Yet, fast imaging performance commonly comes at the cost of limited field-of-view (FOV) and reduced image quality. Here we present single-sweep volumetric optoacoustic tomography (sSVOT) for whole-body imaging of mice from head to tail within 1.8 s. sSVOT capitalizes on a custom-made spherical matrix array transducer together with a multi-beam illumination approach, the latter playing a critical role in maximizing the effective FOV. We compare the performance metrics to the previously reported whole-body mouse imaging implementations.

One of the primary challenges in breast cancer diagnosis and treatment is intratumor heterogeneity (ITH), i.e., the coexistence of different genetically and epigenetically distinct malignant cells within the same tumor. Identification of ITH is critical for designing better treatments and hence to increase patient survival rates. Here, we report a non-invasive approach that integrates photoacoustic computed tomography (PACT) with multitargeted and multiplexed patchy polymeric photoacoustic contrast agents (MTMPPPCAs). The target specificity of MTMPPPCAs to distinguish estrogen and progesterone receptor-positive in breast tumors was demonstrated through both fluorescence and photoacoustic measurements and validated by tissue pathology analysis.

Vasoactivity is an important physiological indicator of cardiovascular health which is frequently measured using ex vivo vessels to determine functional mechanisms and evaluate pharmacological responses. Currently, there are no imaging methods available to assess vasoactivity in multiple vascular beds of living animals noninvasively. In this work, we have developed methods to use photoacoustic tomography to assess vasoactivity in vivo in systemic vasculature of living animals. A spherical-view photoacoustic tomography system was used to monitor acute vasodilation in the whole abdomen of a pregnant mouse in response to injection of G-1. After 3D image reconstruction, the diameter of the iliac artery and photoacoustic signal intensity of a placenta over time was measured. The artery and placenta had differential response to the vasodilator G-1. We validated the observed vasodilation of artery by monitoring the change in cross-sectional diameter of an individual artery using standard B-mode ultrasound imaging.

Pulsed laser diodes (PLD) are preferred as excitation sources in photoacoustic tomography due to their low cost, compact size, and high pulse repetition rate. When PLD is used in conjunction with multiple single-element ultrasound transducers (SUT), the imaging speed can be improved. However, during PAT image reconstruction, the exact radius of each SUT is required for accurate reconstruction. Herein, we propose a novel deep learning approach to alleviate the need for radius calibration. We developed a convolutional neural network (fully dense U-Net) with a convolutional long short-term memory (LSTM) block as the bridge to reconstruct the PAT images. In vivo imaging was used to verify the performance of the network. Our results and analysis demonstrate that the proposed network eliminates the need for radius calibration without sacrificing the reconstructed PAT image quality.

Photoacoustic endoscopy (PAE) allows close visualisation of internal tissues that are challenging to be assessed from outside. It usually employs optical fibres to deliver light to optically excite ultrasound from internal tissues and detect these ultrasound signals with a detector integrated with the fibre. Conventional designs of PAE usually comprise piezoelectric transducers for photoacoustic signals detection, however, due to their opaque nature, these transducers are required to be laterally offset, leading to reduced acoustic sensitivity and increased footprint dimensions. In this work, we fabricated a miniaturized transparent ultrasound sensor using a transparent polyvinylidene fluoride (PVDF) thin film coated with indium tin oxide (ITO) electrodes for PAE. The PVDF-ITO thin film was coated at the tip of a brass tube that had an outer diameter of 2 mm and an inner diameter of 1 mm. A multimode optical fibre was inserted into the brass tube to deliver nano-second-pulsed laser through the PVDF-ITO sensor for photoacoustic signals excitation. The sensor showed an optical transmission rate of ~ 65% to 75% in the wavelength range of 450 to 700 nm, and the photoacoustic signals generated from a carbon fibre had a centre frequency of 15 MHz and a -10 dB bandwidth of 30 MHz. Raster scanning of a focused laser beam on a carbon fibre phantom was performed to obtain optical-resolution photoacoustic endomicroscopy images using a high-speed digital micromirror device via wavefront shaping. Further pre-clinical experiments (tissue samples and in vivo) are required to evaluate the potential of the imaging system for guiding minimally invasive procedures.

Our dual wavelength atlas approach to spectral unmixing has the potential to differentiate two chromophores using only two optical wavelengths. This work investigates the optimization of our method by varying six parameters. An iterative process was used to determine the optimal set of parameters by changing each parameter one at a time and finding which value returned the highest combination of sensitivity and specificity. Once the optimal set of parameters were applied, our novel dual-wavelength atlas method resulted in 10% higher balanced accuracy. These results are promising for photoacoustic-guided surgery.

Optical resolution-photoacoustic microscopy (OR-PAM) is a microscopic system to provide optical absorption contrast in biological tissue. Currently, high-speed OR-PAM scanning faced the limitation of the millimeter-scale field of view (FOV). Without hardware updates, mosaic processing was used to generate a wide FOV PAM image by merging narrow-ranged PAM images. Using feature generation algorithms, feature points were selected by binary decision and were merged by homography estimation. In this study, the diverse feature generation algorithms were applied and compared to estimate their performances for mosaic PAM imaging. Based on the results, mosaic processing implemented with a wide-field OR-PAM system was applied.

Photoacoustic microscopy (PAM) is a multiscale microscopy technique with optical absorption contrast. With tight acoustic focusing, acoustic-resolution (AR-PAM) reaches the depth of several centimeters in biological tissue, but the lateral resolution is relatively poorer than optical-resolution PAM (OR-PAM). We proposed using CycleGAN to generate new OR-PAM images from the original AR-PAM images. We prepared two AR & OR-PAM datasets from leaf phantom and mouse-ear samples. After completing the CycleGAN process, we estimated the quality comparison between the original AR-PAM and the generated OR-PAM images. Finally, the results showed the ability to obtain high spatial resolution PAM images without hardware updates.

We present high-resolution and high-SNR in vivo vascular imaging of the mouse eye, brain, and ear with photoacoustic microscopy (PAM) integrated with a transparent ultrasound transducer. We observed microvessels in a chemically damaged mouse eye. Particularly, the performance of PAM as a comprehensive eye disease diagnosis tool was demonstrated by observing not only corneal neovascularization, but also iris blood vessels and hemorrhages in a cloudy state of the corneal. Second, we delineated in vivo mouse brain vascular imaging at high resolution with a depth encoding. Third, we monitored the ears of tumor-bearing mice to observe for angiogenesis over time.

Recently, 3D photoacoustic (PA) and ultrasound (US) technologies have been developed in various forms. There are roughly two main streams to obtain a 3D image; an 1D array US transducer with motorized scanner and a 2D matrix array US transducer. The former has limitations such as longer time to obtain a single 3D image, bulky design with the scanner that degrades portability, and anisotropic resolution. In contrast, the latter acquires and generates a 3D image at once, which enables real-time imaging, compact design, and isotropic resolution. In this study, we demonstrate dual-modal PA and US imaging using a 2D matrix array transducer scanner. The resolution was measured through a phantom experiment, and the methylene blue signal was verified through a rat SLN experiment.

We updated a clinical photoacoustic (PA)/ultrasound(US) handheld scanner and an imaging system to enhance clinical usability and enable wide-field volumetric PA/US imaging. The scanner was optimized for the weight and dimensions of 600 g and 70 × 62 × 110 mm3, respectively, which is 60 and 63 % of the previous scanner, respectively. The imaging system enabled online maximum amplitude projection (MAP) imaging, which provided images of n × 25 × 38 mm2, where n is the number of scans. We believe that the system and scanner can be useful for a variety of clinical applications.

Characterisation of the elasticity of biological cells is growing as a way to study cell biology. Cell mechanics are related to cell behaviour and potential applications offer great opportunities. Current methods to study cell mechanics are often limited as they may require contact or greater resolution. From the state of the art, the use of high frequency ultrasound (phonons) is one of the most promising since it offers label-free, high resolution and can be integrated on optical microscopes.

Heart and blood vessels called the cardiovascular system closely interact with each other to control cardiac output and maintain vital activities. To observe heartbeat and vascular changes, high-resolution high-speed photoacoustic microscopy (PAM) and/or photoplethysmography (PPG) that are applied technologies of optics have been utilized in preclinical and clinical studies. Herein, we have embedded PPG sensing functionality in a high-speed PAM system to simultaneously perform microvascular imaging and heart rate measurement. In human fingers, we observed capillaries’ movements in blood vessel images from PAM, and moreover found that this phenomenon is due to pulsation by directly comparing between the capillaries’ movements and pulsation-dominant PPG signals. Further, the cardiac cycle could be extracted by quantifying the capillaries’ movements in consecutive blood vessel images, and this periodicity from PAM agreed with that from that of PPG sensing. From these results, the high-speed PAM with PPG could be potentially used as clinical tools for monitoring the changes in multiple cardiovascular information in response to internal and/or external circumstances.

Histological imaging is essential for fields such as oncology, and biological research. However, capturing histological images requires tissue to be prepared in thin sections, then stained with dyes such as hematoxylin and eosin (H&E) to capture tissue morphology. This requires intensive processing, which fundamentally alters tissue structure and chemistry. Presented here is the second generation of photoacoustic remote sensing (SG-PARS) histology microscopes, which aim to circumvent current limitations, providing histological visualizations label-free directly within bulk unprocessed tissues. As photons interact with tissues, they may be scattered or absorbed, where absorption will cause the emission of photons (radiative relaxation) or the generation of heat and pressure (non-radiative relaxation). The SG-PARS features a refined architecture providing simultaneous sensitivity to optical scattering, non-radiative relaxation, and radiative relaxation. Leveraging these contrasts, the SG-PARS may provide visualizations in unstained specimens which directly emulate H&E staining. Combined with deep-learning based (cycleGAN) colorization, the SG-PARS rapidly generates emulated H&E images nearly equivalent to traditional H&E preparations. In addition, the SG-PARS may provide novel chromophore specific properties proposed as the total absorption (the combined radiative and non-radiative absorption magnitude) and the quantum efficiency ratio (QER) (the proportional radiative and non-radiative absorption response). These characteristics may enable visualizations with chromophore specificity beyond that provided by traditional H&E staining. In addition, the SG-PARS features extensive architecture innovations, such as a new 2.7 MHz excitation with a hybrid optomechanical scanning architecture, and a novel visible wavelength detection with a circulator-based pathway and an avalanche photodetector. Applied directly in unprocessed human tissues, the SG-PARS provides marked improvements in contrast, sensitivity, resolution, and scanning speed representing a vital milestone in the development of a microscope ready for clinical adoption.

Photoacoustic tomography with genetically encoded near-infrared probes enables visualization of specific cell populations in vivo at high resolution in deep tissues. Here, we report a near-infrared Forster resonance energy transfer biosensor based on a pair of the near-infrared fluorescent proteins, miRFP670-iRFP720, which enables dynamic photoacoustic imaging of active biological processes. Specifically, we detected apoptosis in single cells at a resolution of ~3 µm in a mouse ear and mouse brain tumors (>3 mm in depth) at a spatial resolution of ~150 µm. These results open the way for high-resolution photoacoustic imaging of dynamic biological processes in deep tissues.

We present the continued development a clinical ultrasound (US) imaging device with a photoacoustic (PA) mode. The combined USPA imaging platform is designed around a compact US component capable of B-mode, M-mode, color Doppler, and pulsed wave Doppler US imaging modes with a 128 element US probe. The PA mode can support a 256 element PA probe with real-time 2D imaging up to 20 Hz. The PA signals are amplified by a 40 dB pre-amplifier while US mode signals bypass the circuit, resulting in high quality PA images. We demonstrate the USPA platform’s capabilities using tissue-mimicking phantoms.

Photoacoustic (PA) endoscopy promises to be useful in a variety of clinical contexts including intravascular imaging, gastrointestinal tracts imaging and surgical guidance. Recent advancements of optical wavefront shaping allow the development of ultrathin endoscopy probes based on multimode optical fibres, which can provide higher spatial resolution than previously reported fibre bundle-based endoscopes. In this work, we developed a forward-viewing PA endomicroscopy imaging system and further improved its performance with a deep image prior (DIP) neural network. Laser was focused and scanned through a multimode fibre via wavefront shaping, in which a real-valued intensity transmission matrix approach was used for fibre characterisation, and a digital micromirror device (DMD) was used for light modulation. The excited ultrasound waves at the distal fibre tip were detected by an ultrasound transducer. High fidelity images of ex vivo mouse red blood cells were acquired. A DIP neural network was then used to improve the spatial resolution with unsupervised learning. Convolutional filters were used to learn features of low-level images as priors and reconstruct high-resolution images accordingly. The performance of the DIP approach was evaluated using a structural similarity index measure (SSIM) at a level of 0.85 with 25% effective pixels, which outperformed the bicubic method. The use of DIP allows reducing scanning positions by several times, and thus improves the speed of pixel-wise PA microscopy imaging. With further miniaturisation of the ultrasound detector, we anticipate that this system could be used for real-time guidance of minimally invasive surgeries by providing micro-structural, molecular, and functional information of tissue.

Severe distortion of ultrasound waves traversing the skull impedes visualization of cerebral structures in humans. Accurate modelling of ultrasound propagation effects is challenging due to highly heterogeneous acoustic properties of the skull bone. Here we demonstrate that acoustic distortions induced by the skull are preserved for optoacoustic waves generated at neighboring point sources. This memory effect is exploited for building a model describing generation and detection of a signal originating from light-absorbing particle at given position. Model-based inversion is shown to accurately recover the absorption distribution with comparable spatial resolution to that obtained without the presence of the skull.

The aim of this paper is to overcome the acoustic diffraction limit to achieve finer lateral resolution in and out of focus regions. For this purpose, a well-known structured-illumination method is used to increase the acoustic spatial bandwidth to improve the lateral resolution of the system. The illuminated optical beam is shaped by a digital micromirror device (DMD) to form sinusoidal patterns force to shift the object spatial spectrum. The shifted spectrum passes high spatial frequency components through the finite bandwidth of the system. Here, the phase compounding method is modified in comparison with [1] to increase the lateral resolution improvement from 2 to 5 times. To achieve this goal, adding the spatial frequency component of the second order to the fundamental component by considering phase shifts of 0,45,90 and 180 degree and orientations of 0,45,90, and 135 degree. The experimental results validate the performance of the system.

Optoacoustic tomography (OAT) is rapidly becoming an important tool in biological and clinical research. Current OAT system implementations commonly feature bulky and expensive hardware impeding wide-scale adoption. We introduce LightSpeed: a compact, high-speed, low-cost and configurable 3D optoacoustic imager based on an optical-link software-defined acquisition platform and fiber-coupled laser diode illumination. We demonstrate real-time imaging capability of the LightSpeed system by fast handheld scanning of human vasculature. LightSpeed attains image quality comparable to conventional data acquisition electronics at a fraction of hardware costs and significantly faster frame rates. It is then poised to result in new generation compact and high-performance handheld OAT scanners facilitating clinical translation.

Delay-and-sum (DAS) is a widely utilized beamforming algorithm in PAI because of its simplicity and real-time capabilities. However, it leads to strong sidelobes and low resolution. A beamformer called delay-multiplyand-sum (DMAS) was proposed, which improved the image quality compared to DAS in PAI. We introduce a weighting factor for DMAS called LAG-DMAS-SCF based on the phase correlation of signals in photoacoustic imaging (PAI). In LAG-DMAS-SCF, the pair-wised signals from DMAS are reorganized in terms of lag and the sign coherence factor (SCF) is applied based on the new signal to the output of DMAS. We have tested the performance of this technique using simulated and experimental data. We have used an LED-based PAI system for the experimental setup. Results show that the proposed algorithm improves the image quality compared to DAS and DMAS by suppressing sidelobes and reducing noise. In phantom imaging, the proposed algorithm showed improvement of around 45 dB and 30dB in Signal-to-noise ratio (SNR), and improvement of around 30% and 20% in full-width-half-maxima (FWHM) compared to DAS and DMAS, respectively. Hence, this algorithm can be an appropriate choice for improving image quality in the LED-based PAI system.

Peripheral arterial disease (PAD) is a condition affecting a large population globally. It is important to identify and treat PAD at an early stage. In this work, we demonstrate the potential of an LED-based photoacoustic and ultrasound imaging system (AcousticX) in PAD diagnosis and treatment monitoring. We used AcousticX to image micro-vasculature networks in palm and foot of a human volunteer. Also, we performed human volunteer imaging experiments to evaluate the efficacy of 3D LED-based photoacoustic and ultrasound imaging in assessing the constriction of microvasculature in response to cold exposure and to study the impact of occlusion in peripheral microvasculature.

Lymphaticovenous anastomosis (LVA) is a surgical method of redirecting excess lymphatic fluid into the venous system to improve the lymphatic flow and prevent the recurrence of edema. In this work, using phantom and human volunteer experiments, we explore the potential of a portable LED-based dual-wavelength photoacoustic imaging system in 3D visualization of human lymphatic vessels and veins when using ICG as a contrast agent. Our results demonstrate that LED-based photoacoustic imaging system used in this study can visualize and separate lymphatic vessels and veins with high spatial resolution, and thus holds potential in guiding LVA procedure.

Recently, piezoelectric transparent ultrasound transducers (TUTs) have gained the attention of the photoacoustic imaging community. The ability to illuminate living subjects through the TUT and minimal acoustic coupling requirements allows for miniaturization, reduces cost, and increases the ease of multimodal integrations. TUTs, therefore, have great potential for photoacoustic imaging; however, challenges remain on matching their acoustic sensitivity to conventional opaque piezoelectric transducers. Various noise sources arising from TUT fabrication and electronics limit their sensitivity. This paper discusses various TUT-specific noise considerations that are detrimental to signal to noise ratio. We also assess the effect of different design considerations by comparing different TUTs. The results can help optimize TUTs design for improving performance, thus paving the way for powerful TUT based photoacoustic applications.

An integrated functional ultrasound (fUS) and photoacoustic (PA) imaging modality is favorable for obtaining multiparametric functional information of deep vasculature. fUS maps blood volume changes of microvasculature using power Doppler frames generated from multiple plane-wave angles with high spatiotemporal resolution, while PA complements vascular oxygen saturation images. In this work, we demonstrated that fUS, ultrasound, and PA imaging can be performed in real-time with a single ultrasound transducer probe. For this, we attached a custom-designed fiber optic illuminator to the fUS probe to optimally deliver light for deep tissue PA imaging. Validation studies on blood flow phantoms and in vivo finger images with a compact imaging head show reduced motion artifacts and ease out the acoustic coupling challenges, making it suitable for multimodal US, fUS and PA imaging of microvasculature.

This study introduces a novel and powerful algorithm for optoacoustic tomography (OAT) in heterogeneous optical scattering media surrounded by optical fibers and ultrasonic transducers in a ring arrangement. The proposed algorithm uses the extended formulation of full-waveform inversion (FWI) based on the alternating direction method of multipliers (ADMM) and fluence compensation (FC). The speed of sound (SOS) map is required in this algorithm as prior information to perform classification and find different regions in the SOS map with equal optical absorption coefficients. The SOS map can be extracted by another FWI procedure applied to the acoustic data generated and recorded by transducers. The fluence map was calculated by Zemax inside the medium at near-infrared (NIR) optical wavelength. Finally, the proposed algorithm iteratively updates the optical absorption coefficients until the data computed with the estimated coefficients match the recorded photoacoustic signals at ultrasonic transducers. This approach shows extremely better results than classical time-reversal algorithms.

Full-waveform inversion (FWI) is an imaging technique that exploits the full information content of waves, including all the phases and their relative amplitude, to get a much higher resolution than tomographic methods, which only exploit a small fraction of the data. Over the last decade, FWI has reached maturity and emerged as the leading seismic imaging technique, along with the continuous development of high-performance computing. This trend also contributes to extending FWI towards new fields such as medical imaging. In this paper, we have developed a 2D extended formulation of FWI based on the alternating direction method of multipliers (ADMM) to estimate the speed of sound (SOS) and acoustic attenuation inside a heterogeneous medium from a circular array of 128 ultrasonic transducers. Our Extended 2D FWI has shown promising results when it is compared with travel time tomography and the classical formulation of FWI.

Various beamforming reconstruction algorithms have been proposed in the literature for the reconstruction of the initial pressure rise such as Dealy and sum (DAS), Delay multiply and sum (DMAS), p-DAS, time reversal (TR), Fourier domain reconstruction etc. The universal back projection (UBP) is most popular amongst all due to its simplicity and fast implementation. Also, it can be applied to planar, spherical and cylindrical geometry. The reconstruction from conventional beamforming algorithm and UBP suffers from artifacts when insufficient data is used for the reconstruction. Due to the limited data, the under-sampling artifacts appears in the reconstructed image, which can degrade the image quality and leads to misinterpretation of the physiological condition. In this study, a compressed sensing-based algorithm based on ADMM is proposed for the linear array transducer which can mitigate the effect of the under-sampling artifacts and provide better results with the conventional beamforming reconstruction algorithms and UBP.

Thermoacoustic imaging (TAI) utilizes the advantages of excellent penetration depth and contrast of microwave energy and the high spatial resolution of ultrasound imaging. We evaluate the use of TAI for the detection of hemorrhage in the neonatal brain through fontanelle. We use a 3D human neonatal brain model, an antenna, and a linear array transducer in simulation to characterize the thermoacoustic signal and corresponding reconstructed images. All the characterizations are conducted using Computer Simulation Technology (CST) Studio Suite based on finite integration in technique (FIT). The absorbed electric field by the target and the time varying heating function data are reconstructed with a spatial resolution of 100 μm. To evaluate the impact of the applied microwave beam on the generated acoustic pressure wave, different pulse widths ranging from 0.01μs - 5μs at different frequencies from 1-3 GHz are tested. We also explore the impact of the type of antenna, by evaluating a horn antenna, a waveguide and a helical antenna.

In transcranial brain imaging, the presence of the skull represents a challenge in the generation of clear signals. For the human skull, and animals with a thick skull, optical and ultrasound propagations are significantly degraded due to the bone properties. First, the optical and acoustic transmission and attenuation after crossing the skull bone are separately identified. Optical measurements are performed in the range of 690 to 940 nm in order to determine the transmission at different regions of the skull; the total hemoglobin absorption coefficient is considered for the imaging target. Then, the impact of the skull thickness on the ultrasound propagation through the bone is evaluated using 2.25 MHz transducers. In the second part, photoacoustic imaging on the frontal, parietal and occipital bones are evaluated. In addition to the skull thickness variation, the angle of illumination and the depth of the imaged target are studied. Based on the results we determine regions of the human head where transcranial photoacoustic imaging at 2.25 MHz is feasible, taking advantage of the good resolution due to the high frequency.

Cerebral oxygenation monitoring at early stage of life is a measure of diagnosing hypoxia-induced brain injuries that may otherwise result into severe neurological disorders. Two primary clinical methods (percutaneous insertion of a catheter and insertion of a probe through cranium) used to monitor brain oxygenation are invasive. Therefore, an accurate and non-invasive brain oxygen saturation measurement in neonates is of great significance. In addition, the presence of the optical/acoustic window in neonates, known as fontanelle, provides an opportunity to extract cerebral oxygen saturation from deeper structures within the brain tissue. High sensitivity to oxy and deoxyhemoglobin of photoacoustic (PA) technique enables the non-invasive oxygenation measurement based on spectroscopic analysis at two or more wavelengths. Here, we have utilized a handheld Transfontanelle PA probe to measure brain oxygen saturation in sheep brain in-vivo through surgically induced fontanelle. The rationale of using sheep brains for this study is the similarity in size to a human neonatal brain. The inspired oxygenation level by the animal ranging from 12 to 100% was externally controlled to introduce both hypoxia and hyperoxia. The performance of the PA probe was evaluated by comparing with the gold standard arterial blood gas analyzer results. The extracted oxygenation saturation from Transfontanelle PA strongly correlates with measured value of the blood gas analyzer as compared to pulse oximeter reading. Our results indicate that monitoring of cerebral blood oxygenation can be performed in neonates noninvasively and accurately through the fontanelle.

Photoacoustic microscopy (PAM) is a high-resolution imaging modality capable of visualizing fine microvasculature in a biological tissue. Clinical translation of PAM system is still an issue due to the use of expensive and high energy laser sources for imaging. Although low energy laser sources can facilitate clinical transition of PAM systems as they are rugged, portable, affordable, and safe to use, the photoacoustic (PA) signal they generate is very weak, resulting in very low signal-to-noise ratio (SNR) PA signals and in turn low quality PA images. In this study, we have developed an enhancement autoencoder (EAE) utilizing fully convolutional neural network, that can improve the quality of the PA signals received, consequently improving the SNR of the reconstructed images. We acquired PAM data from rat brain tissue with both high energy (target data) and low energy (input data) of the laser for training purposes and tested our trained model on PAM data obtained from new rat brains. Our effort is to reconstruct the vascular structure as well as an accurate reading for the blood concentration. The latter has been neglected in the previous studies. The performance of our EAE is evaluated in terms of SNR, structural similarity index (SSIM), root mean square error (RMSE) and correlation.

Optical resolution photoacoustic microscopy (OR-PAM) is a high-resolution imaging configuration of photoacoustic Imaging. However, due to limited active detection area of ultrasound transducer, the field of view of an OR-PAM is limited to only few millimeter. Researchers have tried to improve the field of view of OR-PAM by implementing the mechanical scanning of the ultrasound transducer, but this has slowed down the speed of imaging significantly. We have developed a very fast, ultra-wide laser scanning photoacoustic system that provides a very large active detection area. The system utilizes a novel spiral scanning mechanism, using traditional 2D galvo mirrors to scan an area of ~4cm in less than 5 seconds. To characterize and demonstrate the imaging capability of the system, we image rat brain in vivo.

Precise assessment of nerve injury optimizes outcomes after surgical nerve repair. Photoacoustic imaging is a promising technique for this intraoperative assessment, but is challenged by optical scattering which reduces light penetration into nerve tissue. This work investigates custom light delivery methods to enhance optical penetration into nerve tissue using Monte Carlo simulations. Light sources were positioned in four configurations surrounding the nerve using sparse activation patterns that were evenly distributed or clustered. Results indicate that a custom light delivery system with combined radial and lateral trajectory illumination maximize optical penetration into nerve tissue for intraoperative photoacoustic assessment of nerve injury.

We report on the development of a novel piezo-MEMS-based optofluidic platform to detect the concentration of various species dissolved in a fluid. This platform employs piezoelectric micromachined ultrasound transducers (PMUTs) to work as a photoacoustic receiver, receiving ultrasound from fluid targets present in microfluidic channels while illuminated with a nanosecond pulsed laser. We fabricate both the PMUTs and the microfluidic channels and subsequently use them for the experiment. We also show the capability of PMUTs as a general photoacoustic receiver and demonstrate its signal-to-noise characteristics (~31) and its wide fractional bandwidth (~73%).

Timely, accurate assessment of atherosclerotic plaque composition in the carotid arteries to enable prevention of ischaemic stroke is currently an important topic in medical imaging. Photoacoustic imaging, which lacks ionizing radiation and foreign contrast agents, is attracting much interest in the field of non-invasive carotid plaque imaging. We present results of the characterization of our new system, based on a concave ultrasound transducer and optical fibre illumination, as well as multi-wavelength photoacoustic images of a PVCP-based phantom, supplemented with ultrasound plane wave images. The multi-wavelength photoacoustic imaging capability of the probe has potential for eventual carotid plaque imaging and quantification.

In the present work we show that the interaction between short lasers pulses and tailor-made nanostrucured materials can produce broadband high-frequency ultrasound. In order to obtain efficient light-to-pressure conversion we used nanostructured carbon nanotubes permeated with high thermal expansion polydimethylsiloxane and nanometric TiO₂ particles with adsorbed dye infused polystyrene. It is possible to achieve ultrasound pulses with central frequencies of 80 MHz and bandwidths of 125 MHz (at -6 dB) using a 6 ns pulsed laser. The aforementioned materials are used in order to establish proof-of-concept of the non-destructive ultrasound pulses interaction with biological membranes, exemplified by the unloading of Green Fluorescence Protein from giant unilamellar vesicles, and of non-invasive material inspection, illustrated with the measurement of the thickness of thin paint layers based on ultrasound frequency attenuation.

The feasibility of real-time tracking of microparticles intravenously injected into living organisms can significantly facilitate the development of new biomedical applications, including blood flow characterization, drug delivery, and many others. However, existing imaging modalities generally lack the sensitivity to detect the weak signals generated by individual particles flowing through vascular networks deep within biological tissues. Also, the temporal resolution is usually insufficient to track the particles in an entire three-dimensional region. Herein, we capitalize on the unique advantages of a state-of-the-art high-frame-rate optoacoustic tomographic imaging system to visualize and track monodisperse core-shell microparticles with a diameter of ~4 μm in the mouse brain vasculature. The feasibility of localizing individual solid particles smaller than red blood cells opens new opportunities for mapping the blood flow velocity, enhancing the resolution and visibility of optoacoustic images, and developing new biosensing assays.

Photoacoustic tomography defines new challenges for ultrasound detection compared to ultrasonography. To address these challenges, a sensitive, small, scalable, and broadband optomechanical ultrasound sensor (OMUS) has been developed. The OMUS is an on-chip optical ultrasound sensor, using optical interferometric ultrasound detection. It consists of an acoustic membrane on top of an optical ring resonator that modulates the optical ring resonance with high efficiency enabled by an innovative optomechanical waveguide. Raster scanning photoacoustic tomography has been demonstrated with a single-element OMUS. Based on performance and form factor, the OMUS combined with passive optical multiplexing may enable new applications in photoacoustic imaging

A 3D photoacoustic microscopy (PAM) system is presented and characterized, using optical MEMS and fibre tip transducers as active elements to provide all-optical positioning and read-out. The excitation beam position is controlled using an electrostatically actuated 2-axis MEMS scanner. This allows for fast 3D scanning without motion-induced artefacts caused by stage scanning, and selective imaging of regions of interest. A 20 MHz fibre tip transducer is used for acoustic detection, which allows a variety of sample holders to be used including common approaches such as multi-well plates and petri-dishes.

As two important acoustic imaging modalities, photoacoustic microscopy (PAM) and ultrasound microscopy (UM) provide complementary functional and structural information on biological tissues. At penetration depths beyond the optical diffraction limit, the spatial resolution of both PAM and UM is determined acoustically by the receiving (focused) ultrasound transducer. To obtain good dual-modal imaging resolution, the transducer should have a high center frequency with a wide bandwidth. More importantly, a high acoustic numerical aperture (NA) should be maintained for receiving the PA and ultrasound signals, which is however difficult to achieve as the light blockage by the transducer is more likely to occur at higher NAs. In this paper, we report the dual-modal PAM and UM based on an optically-transparent focused PVDF transducer with a high NA of 0.64. The transducer has an acoustic center frequency and bandwidth of 36 MHz and 44 MHz, respectively. The acoustic focal diameter and zone are 37.8 μm and 210 μm, respectively. With a central transparent window, the excitation laser pulses can directly pass through the transducer to illuminate the target. This allows a high acoustic NA to be obtained without light blockage. For demonstration, co-registered 3D and 2D PAM/UM imaging experiments have been conducted on a twisted wire target in both water and chicken breast tissue, and in-vivo on a mouse tail, respectively. The imaging results show that high acoustic resolution and sensitivity can be achieved to resolve the target at different depths with a simple and compact dual-modal imaging setup

The transparent ultrasound transducer (TUT) has recently emerged as an attractive platform for development of multiscale photoacoustic imaging (PAI) systems. TUT allows an easy co-alignment of optical illumination and acoustic detection paths on the tissue surface, averting the complex beam arrangements employed by the current PAI systems that use opaque conventional ultrasound transducers. However, TUTs suffer from narrow bandwidth and low sensitivity due to the lack of suitable transparent acoustic matching and backing layers. To address the above challenges, in this study we explored a novel optically translucent acoustic matching layer made out of glass beads suspended in transparent epoxy to improve both the transducer bandwidth/sensitivity and light fluence on the tissue surface. Our experiments with 13 MHz center frequency TUTs coated with varying glass bead concentration demonstrated that both pulse-echo bandwidth and sensitivity increases with glass bead concentration. Approximately 3.3 fold improvement in bandwidth and a 2.5 times higher pulse-echo sensitivity was measured with TUT fabricated with an acoustic matching layer of 40%GB. Optical measurements were conducted to confirm that the translucent glass bead layer can act as a light diffuser to help achieve uniform light distribution on the tissue surface.

Photoacoustic tomography (PAT) is a hybrid imaging technique, which is capable of providing abundant optical contrast and high spatial resolution at deep penetration depth. One fundamental challenge in current PAT system design is caused by the opaque structure of the ultrasound transducer arrays, which prevents effective light energy delivery onto the imaging target. To address this issue, we report a new optically-transparent PVDF (polyvinylidene fluoride) transducer array for PAT. It consists of a 1D array of 16 elements, each of which has a transparent window to allow the excitation laser pulses to directly pass through. As a result, the optical excitation and ultrasound detection zones are automatically aligned to improve the illumination condition and PA excitation efficiency. Its imaging performance, such as contrast, lateral resolution, and penetration depth has been characterized with optical phantoms. Preliminary ex-vivo PAT experiments on chicken breast tissues have been conducted to demonstrate its imaging capability on real biological samples. The experimental results show that the optically-transparent PVDF transducer array could provide a new solution for the miniaturization and clinical translation of handheld PAT systems.

Laparoscopic surgery is commonly used in the abdominal creating minimal trauma to the patient. With the current standard videoscope, visualizing vessels under the tissue is challenging. Photoacoustic (PA) imaging is laser-generated ultrasound imaging that offers vascular mapping. However, most in vivo PA imaging is limited by the lack of depth penetration due to the insufficient illumination in deep tissue caused by the light scattering and the risk of over-irradiation near laser contact tissue. Having a method to bring the signal sensing device closer to the target area is critical for the PA-guided intervention. A miniaturized PA laparoscopic should increase the maneuverability inside the abdominal for better visualization of the operational region. A PA imaging probe mainly consists of optical fibers and an ultrasound (US) transducer. While the US device miniaturization has been investigated, previous works used customized angled-tips fiber for side-illumination to miniaturize the light delivery, requiring multiple fibers or having limited illumination region. Diffusing fiber can illuminate a wider region with fewer fibers than angled-tip side-illumination fibers. This could further minimize the diameter of the entire PA laparoscopic by several millimeters. We propose a PA laparoscopic system imaging with only two diffusing side-illumination fibers while maintaining high image contrast. The phantom study shows the system can image 0.30 mm resolution with a 30 dB signal-to-noise ratio in an optical scattering environment. The result demonstrated the feasibility of delivering sufficient energy with diffusing fiber to miniaturize the dimensions of the PA laparoscopic device.