All-optical optoacoustic micro-tomography in reflection mode (2024)

Journal List
Biomed Eng Lett
v.13(3); 2023 Aug
PMC10382435

As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice

Biomed Eng Lett. 2023 Aug; 13(3): 475–483.

Published online 2023 Apr 19. doi:10.1007/s13534-023-00278-8

PMCID: PMC10382435

PMID: 37519878

Tamar Harary, Yoav Hazan, and Amir Rosenthal

Author information Article notes Copyright and License information PMC Disclaimer

Associated Data

Data Availability Statement

Abstract

High-resolution optoacoustic imaging at depths beyond the optical diffusion limit is conventionally performed using a microscopy setup where a strongly focused ultrasound transducer samples the image object point-by-point. Although recent advancements in miniaturized ultrasound detectors enables one to achieve microscopic resolution with an unfocused detector in a tomographic configuration, such an approach requires illuminating the entire object, leading to an inefficient use of the optical power, and imposing a trans-illumination configuration that is limited to thin objects. We developed an optoacoustic micro-tomography system in an epi-illumination configuration, in which the illumination is scanned with the detector. The system is demonstrated in phantoms for imaging depths of up to 5mm and in vivo for imaging the vasculature of a mouse ear. Although image-formation in optoacoustic tomography generally requires static illumination, our numerical simulations and experimental measurements show that this requirement is relaxed in practice due to light diffusion, which hom*ogenizes the fluence in deep tissue layers.

Keywords: Imaging techniques, Medical and biological imaging, Optoacoustic tomography, Photoacoustic imaging

Introduction

One of the significant challenges in optical microscopy is the dominant effect of light diffusion, which generally limits the penetration depth to hundreds of micrometres [1]. Optoacoustic (photoacoustic) imaging offers a solution to this challenge by combining optical excitation with ultrasound detection, enabling imaging with optical contrast at penetration depths typical to ultrasound [2, 3]. Conventionally, high penetration is achieved in optoacoustic imaging by illuminating the tissue with a broad beam of pulse-laser light, leading to the generation of multiple acoustic sources within the tissue and detecting the acoustic signals at multiple positions on the surface. The image is then created by solving an acoustic-reconstruction problem, without loss of resolution due to optical diffusion [4].

Typically, acoustic-resolution optoacoustic systems are divided into two categories: optoacoustic tomography (OAT) [5] and optoacoustic microscopy (OAM) [6, 7]. In OAT, the acoustic detectors integrate the signals emanating from the entire region of interest, where image formation is performed via tomographic reconstruction algorithms. OAT is conventionally performed with multi-element arrays, leading to rapid data acquisition and image formation. Since the lateral resolution in OAT scales with the detector size, its typical resolution is generally worse than 100μm due to manufacturing limitations and the loss of sensitivity of piezoelectric transducers upon miniaturization. Higher resolutions may be achieved by OAM, which may be classified into two groups, based on the mechanism used for focusing. In optical-resolution OAM (OR-OAM), the optical beam is focused on the specimen, enabling imaging resolutions typical to optical microscopy, but at the cost of reduced penetration as it is limited by light diffusion [8]. In acoustic-resolution OAM (AR-OAM), the illumination is broad and the focusing is performed acoustically by the ultrasound detector, enabling penetration depths of several millimetres in optically diffusive media.

In contrast to OAT, the lateral resolution in AR-OAM is determined by the focal width, rather than the detector size, enabling the use of large-area detectors focused to a small spot, to achieve both high sensitivity and resolution. One of the main limitations of OAM is the trade-off between the focal length and depth of field of the focused detector, often leading to superficial imaging when the lateral resolution is optimized [9, 10]. For example, high-frequency focused transducers optimized for 20μm focal widths have depths of field of 100μm or lower [4]. Although synthetic-aperture algorithms may be used to extend the depth of field numerically [11, 12], their performance often deteriorates as the distance from the focus increases [11]. Thus, even when only superficial imaging is desired, AR-PAM still requires that object surface be relatively parallel to the detection surface to achieve optimal performance, thus complicating its use in imaging curved surfaces.

The standard distinction between OAT and OAM is technological rather than fundamental, as a tomographic configuration can, in principle, achieve any resolution, assuming the detector is sufficiently small, sensitive, and wideband. While in piezoelectric technology, there is a compromise between sensitivity and size, which limits the resolution of OAT, such a trade-off does not exist in optical technology. For example, optical resonators fabricated in a silicon-photonics platform can achieve exceptional sensitivities [27] and miniaturization levels, going below 1μm [13]. Recently, we have developed a miniaturized silicon photonics acoustic detector (SPADE) that also achieved high signal fidelity, enabling its use for in vivo OAT. SPADE was based on a π-phase-shifted Bragg grating (π-BG) resonator coated with an elastomer and achieved noise-equivalent pressures (NEPs) as low as 2.2 mPa Hz ^− 1/2, a bandwidth above 200MHz, enabling OAT with typical axial and lateral resolutions of approximately 15μm [14]. In comparison, polymer-based optical resonators, e.g., Fabry-Pérots [28] and micro-rings [29] conventionally achieve lateral resolutions on the scale of 100μm due to their larger element size. In addition, in contrast to fiber Bragg gratings in silica fibers [30–33], the acoustic response of SPADE is not affected by undesired acoustic phenomena such as guided acoustic waves or reverberations, which may hinder tomographic imaging. Nonetheless, the low NEP of 1.5 mPa Hz ^− 1/2 recently obtained for fiber Bragg gratings using heterodyne detection makes them a promising platform for endoscopic imaging when an OR-OAM configuration is used [30].

Similarly to conventional OAT, the implementation of optoacoustic micro-tomography (OMT) using SPADE was performed with static illumination, where the detector was scanned to capture the acoustic signals from different positions (Fig.1a), i.e., the projections in the tomographic problem. From a strictly theoretical perspective, static illumination was essential for OMT, since its underlying assumption is that all the projections correspond to the same optoacoustic source distribution. However, in practice, using static illumination in OMT required using high-power lasers with sufficient pulse energy to cover a large area and imposed a trans-illumination configuration, which limited imaging to thin objects. In contrast, OAM is generally performed in an epi-illumination configuration in which the illumination is scanned with a focused transducer, enabling the use of localized illumination and imaging of thick objects.

Open in a separate window

Fig. 1

OAT Implementations. Figure1-a: Illustration of the trans-illumination OMT geometry, developed in Ref. [14]. Figure1-b: Illustration of the epi-illumination geometry used in the current work. Figure1-c: Illustration of the 2D scanning pattern. While only the detector was scanned in the trans-illumination setup, in the epi-illumination format, the detector and illumination were scanned together

In this work, an epi-illumination OMT system is developed, which enables imaging thick objects with high resolution. In our system, the illumination fibers and SPADE are integrated into a single imaging head, which is scanned in two dimensions at a small distance from the imaged object to acquire the projection data (Fig.1b). In contrast to transmission-mode OMT, where the illumination had to cover the entire imaged object, the illumination area of our system was confined to a small area of approximately 4 mm², leading to two practical advantages: First, by reducing the illumination area, the laser energy could be reduced by a factor of five in comparison to Ref. [14]. Second, since the illumination was scanned, the system could be used to cover larger areas without loss of sensitivity.

Our design exploits the diffusive propagation of light in biological tissue to achieve a relatively depth-independent lateral resolution. In principle, if the illumination width is fixed, a deterioration in the lateral resolution is expected as the depth is increased beyond the beam width due to the lower angular span in which signals from the objects are detected. However, when light propagates diffusively in the imaged object, the illumination width increases with depth, effectively increasing the angular span.

Our OMT system was studied in numerical simulations, tissue-mimicking phantoms, and in vivo. In the numerical simulations, the effect of light diffusion on the reconstruction quality was studied, demonstrating its benefit in maintaining the lateral resolution at increasing depths. In the phantom measurements, dark absorbers were imaged inside a diffusive medium with similar optical properties to tissue, demonstrating axial and lateral resolutions of at least 25 µm and 33 µm at a depth of 2 mm and an ability to image with a lateral resolution of approximately 40 µm at depths up to 5 mm. In the in vivo measurement, the ear of a mouse was successfully imaged over an area significantly larger than that of the illumination.

Simulations

We simulated the illumination and detection geometries used in our epi-illumination OAM and demonstrated the effect of light diffusion on the quality of the reconstruction. The simulated imaging setup, presented in Fig.2a, was composed of two optical beams directed at angles of ± 40° and an acoustic point detector positioned between them, where the beams and detectors were scanned together over a line. The beams had an initial width of 1 mm and diverged with an angle of 0.02° when propagated in a scattering-free medium. The absorbing structures were cylinders with a diameter of 300 µm, positioned at depths ranging from 1 mm to 5 mm. Two types of media were studied: low scattering medium and high scattering medium, mimicking the optical properties of biological tissue [15], as specified in table table1.1. As the table shows, in the low-scattering case, the absorption and scattering coefficients of both the background and insertions were extremely low, making the entire medium essentially transparent. In contrast, in the high-scattering case, the absorption and scattering values in the background and insertions corresponded to those of tissue and blood vessels.

Open in a separate window

Fig. 2

Epi- imaging Simulation. Figure2-a: Schematics of medium and illumination cross-sectional view. Figure2-b: OA signals and reconstructed image from low scattering medium (top) and from and high scattering medium (bottom). Figure2-c: Normalized signals intensity versus depth from the low scattering medium (blue) and high scattering medium (orange)

Table 1

The absorption µ_a (mm^− 1) and scattering µ_s (mm^− 1) coefficients of the background medium and internal structures for the low and high scattering simulations

	Layer 1 - Water				Layer 2 - Tissue
	Background				Background		Insertions
	µ_a	µ_s	n	g	µ_a	µ_s	µ_a	µ_s	n	g
Low scattering	0.001	0.01	1.3	1	0.01	0.1	0.05	0.1	1.4	0.9
High scattering	0.001	0.01	1.3	1	0.05	10	1	23	1.4	0.9

Open in a separate window

The simulations were performed by combing three types of numerical tools: A light-propagation simulator based on the Monte Carlo software ValoMC [16, 17], an acoustic-propagation simulator based on k-wave software [18, 19], and a reconstruction model based on the back-projection (BP) algorithm [20]. We simulated a 1D scan of 10 mm lateral length in 450 µm steps, where for each scan position, the acoustic signals were acquired by first simulating the light propagation and then the acoustic propagation. Figure2b shows the simulated acoustic signals for the 1D scan and their corresponding reconstructions for both medium types. The amplitude of the reconstructed insertions as a function of depth is shown in Fig.2c, revealing a milder decay in amplitude for the high-scattering case.

The results clearly show that optical scattering leads to a significant improvement in the quality of the acquired acoustic signals and corresponding reconstructions. First, depth-dependent decay in the image intensity is lower in the optically diffusive medium owing to the effect of dark-field illumination, as also experimentally observed for handheld OA probes [21]. Second, the time series OA signals in the diffusive medium are fuller, leading to less pronounced streak artifacts in the reconstruction and higher resolution.

Methods

SPADE

The ultrasound detector used in this study, SPADE, is thoroughly described in Ref. [14]. Briefly, the detector is based on a silicon waveguide Bragg grating fabricated on top of an SOI chip using CMOS technology. A TE π-phase-shifted Bragg within a silicon core with a cross section of 225 x 500 nm² by creating a side wall corrugation of 40 nm with a period of 265 nm. Then, a 2 µm PDMS polymer layer was applied directly on the Si layer, serving as an over-cladding layer to increase sensitivity. Two polarization-maintaining fibers were bonded to couple light in and out of the silicon waveguide through grating couplers using the technique described in Ref. [14]. In the final step, a 300 nm layer of gold was deposited on the chip and the bonded fibers to reflect stray optoacoustic illumination from the chip, to minimize the thermal signal created by the optical absorption of the silicon substrate. Based on previous characterization measurements [14], the SPADE design used in this work had an effective length of approximately 30 µm, determined by the light localization in the device.

Excitation and SPADE holder setup

The optoacoustic excitation was performed using an Nd:YAG pulsed laser operating at the wavelength of 532 nm (Optogama, “WAVEGUARD”) with a pulse width of 1 ns, a repetition rate of 1 kHz, and pulse energy of 80µJ (“Laser1”; Fig.3b). The laser beam was guided through 2 beam splitters (“BS”; Fig.3b). Each of the 4 beams was focused with lens (“L”; Fig.3b) and coupled to 1 mm multi-mode fibers with an NA of 0.5. Accounting for coupling losses, the total pulse energy at the fiber output was approximately 60 µJ. The holder was manufactured by a 3D printer (Formlabs, FORM3) with Clear resin V4 (Formlabs, Transparent applications).

Open in a separate window

Fig. 3

System Setup. Figure3-a: Illustration of the optical readout part of the system based on MZI. Figure3-b: Illustration of the optical excitation part of the system. [ M- Mirror; B.S – Beam Splitter; L-Lens; Gray tube -Fiber]

Optical interrogation system

The optical interrogation system, used to convert the acoustically induced refractive-index modulation in the resonator into voltage signals, was based on a Mach-Zehnder interferometer (MZI), described in Fig.3a. The output of a tunable CW laser was split between a sensing arm with the SPADE chip and a reference arm with a piezoelectric fiber stretcher (OPTIPHASE, PZ3) and optical delay-line (OZ optics, ODL-100). The optical path difference between the two arms was minimized to reduce the effect of laser phase noise, and the arms’ outputs were recombined and delivered to a balanced photodetector. The MZI was stabilized to quadrature by the fiber stretcher, where the differential signal is zero, using a feedback circuit with a bandwidth of 3kHz [22]. The laser was tuned to the maximum transmission wavelength of the resonator, where its phase response changes linearly as a function of wavelength [23, 24], using the signal from a tap photodiode. When the wavelength of the resonator was acoustically modulated at frequencies above 3kHz, the induced phase shift was not compensated by feedback circuit, leading to a modulation in the output voltage signal, which was recorded by a digitizer (M3i.4860-Exp, SPECTRUM) with a sampling rate of 180MS\s.

Results

Phantom measurements

In the phantom measurements, two types of absorbing structures were used as the optoacoustic targets: nylon sutures with widths of 20 and 40μm (Sharpoint black monofilament, eSutures Inc. USA) and dark polymer spheres with a diameter of 100μm (Black paramagnetic polyethylene microspheres, Cospheric Inc., California, USA). The samples were embedded in a solid agar to restrict their motion, which was either transparent or translucent. In the case of the translucent phantoms, the agar mimicked the optical scattering properties of tissue, emulating the hom*ogenizing effect of light diffusion in tissue.

In the first phantom measurement, a single 100μm microsphere at a depth of 3mm was imaged, and scanning was performed over an area of 2 × 2 mm² with a 50μm step size. Figure4 shows the amplitude cross-sections of the 3D reconstructions of the microsphere obtained for the two types of agar media. The full lateral width at half-maximum (FWHM) of a single microsphere in the transparent medium equaled 149 and 161μm, and the axial reconstruction was 149μm, as shown in Fig.4a. In comparison, the lateral FWHM widths of a microsphere in scattering medium equalled 108 and 116μm, and the axial FWHM width equaled 116μm, as shown in Fig.4b. Similarly to the numerical simulations, the lower resolution obtained for the transparent phantoms may be attributed to the missing projections in detector positions where the illumination did not reach the absorber.

Open in a separate window

Fig. 4

Sphere Source Performance. Figure4-a: Reconstruction cross section of 100μm sphere source in hom*ogeneous medium. The reconstruction XY lateral widths were 149 and 161μm, and the axial resolution was 149μm. Figure4-b: Reconstruction cross section at scattering medium. The reconstruction XY lateral widths were 108 and 116μm, and the axial resolution was 116μm

In the second phantom measurement, numerous microspheres were scattered inside the optically diffusive agar with depths up to 4mm. Scanning was performed over an area of 6 × 6 mm² with a 50μm step size. Figure5a-c shows the maximum intensity projection (MIP) of the 3D reconstruction performed on the three principal axes, where Fig.5d shows a typical histogram obtained in the measurement. The FWHM of the reconstruction in the lateral direction x as a function of depth is shown in Fig.5e, revealing a depth-independent lateral width of 116μm and the axial resolution width of 116μm up to depths of 3.5mm with a slight increase at 4mm. The magnitude of the reconstructed microspheres was also relatively constant up to depths of 3.5mm, as shown in Fig.5f, despite the optical scattering, which led to weaker acoustic signals from deeper microspheres (Fig.5d). This result may be explained by the broader illumination at lower depths, which led to the summation of more projections for the reconstruction of the deep spheres, partially countering the reduction in the acoustic signals. Nonetheless, at depths beyond 3.5mm, the reduction in the reconstruction magnitude as a function of depth due to optical scattering became more dominant.

Open in a separate window

Fig. 5

MIP of 100μm spheres at depths up to 4mm away from the surface. Figure5-a: MIP from 3D reconstruction obtained for the measurement at XY Plane. Figure5-b: MIP at ZY plane Fig.5-c: MIP at XZ plane. Scale bar: 1mm. Figure5-d: Typical raw OA signals. Figure5-e: Spheres width as a function of depth. Figure5-f: MIP as a function of depth

The third phantom measurement, a 20μm nylon suture was imaged to demonstrate the capability of the system for high-resolution OAT. The suture was embedded in scattering agar at a depth of 2mm, and the imaging head was scanned over an area of 3 × 3 mm² with a 20μm step size. The measured acoustic signals were used to tomographically form an image, whose MIPs are shown in Fig.6, demonstrating axial and lateral reconstruction widths of 25 and 33μm, respectively.

Open in a separate window

Fig. 6

Resolution Performance. 20μm Nylon suture MIP images and reconstruction cross sections. The reconstruction lateral (top) and axial (bottom) widths were 33μm and 25μm, respectively

In vivo measurements

The OMT system was tested in vivo for imaging the microvascular structures of a mouse ear. A CD-1 mouse model was anesthetized using isoflurane and placed under an IR heating lamp to maintain its body temperature. To enable the propagation of the acoustic signals from the mouse ear to the acoustic detector, the imaging head was positioned inside a water reservoir held by a thin polyethylene membrane in its bottom. The water reservoir was placed on the mouse ear, where an additional water drop was used to ensure continuous contact between the two. The imaging head was scanned 3mm above the mouse ear with a span of 14mm × 14mm and step size of 30μm, and the measured acoustic data was used to form a 3D reconstruction of the mouse ear using the BP algorithm [20].

Figure8a shows the optical image (left) compered to the maximum amplitude projection (right), whereas Fig.8b presents tomographic slices at four depths of the three-dimensional OMT reconstruction with 50μm steps. The depth information, along with the spatial resolution, is visible and emphasizes the microscopic resolution achieved in the reconstructions. A representative sample of the raw acoustic data acquired by the system over one of the line scans are shown in Fig.8c, showing multiple hyperbolic OA signals, typical with tomographic measurements.

Open in a separate window

Fig. 8

In-vivo Tomographic imaging. Figure8-a: Microscope images of mouse ear (left) and corresponding MIP of the optoacoustic image (right). Figure8-b: Montage of four different tomographic depth. The depth difference between each consecutive slice was 50μm. Figure8-c. Typical raw OA signals from a mouse ear. Scale bar: 1mm

Conclusion and discussion

In conclusion, a new OMT system was developed employing SPADE with wideband ultrasound detection in an epi-illumination in which the illumination is scanned with the detector. By using dark-field illumination and exploiting the optical diffusion of light in biological tissue, the variations in the illumination pattern at different detector positions is reduced, enabling the use of tomographic reconstruction algorithms that assume static illumination. The benefits of light diffusion and dark-field illumination are studied both in numerical simulations and phantom measurements, demonstrating higher resolutions, weaker artifacts, as well as slower depth-dependent signal decay.

The use of epi-illumination in OMT significantly improves the usability of this technique over the previously published trans-illumination configuration [14]. First, epi-illumination enables the imaging of thick objects, facilitating clinical applications such as skin imaging [25]. Second, while in trans-illumination the required laser power scaled with the area of the imaged region, in the epi-illumination this requirement is relaxed by scanning the illumination together with the acoustic detector and exploiting light diffusion to hom*ogenize the light distribution at higher depths.

The epi-illumination OMT system was used to image a 20μm suture, demonstrating that the system is capable of achieving axial and lateral resolutions of at least 25 and 33μm, respectively, which are comparable to values obtained by AR-OAM systems [9]. We note that these bounds on the resolution are mostly due to the size imaging target, and do not represent the resolution limit achievable by SPADE, which was approximately 15μm in all axes [14]. Specifically, the width of a reconstructed object,, is generally expected to be a Pythagorean addition between the object’s true width, , and the width of the PSF, : . Using µm and , one indeed obtains the measured reconstruction width 25μm. The higher value obtained for the lateral resolution may be explained by the limited tomographic view in the measurement, which is affected by the efficiency of the light diffusion in hom*ogenizing the illumination inside the medium.

The developed epi-illumination OMT system has two potential advantages over conventional AR-OAM systems, which employ a focused ultrasound transducer. First, the minute size of the sensor enables, in principle, miniaturization of the entire imaging head, enabling the use of scanning stages for low-weight objects. Second, the use of a tomographic image acquisition enables image reconstruction over a large depth of field, whereas AR-OAM is generally limited by the short depth of field of the focused transducer. Indeed, in our work imaging a 5mm thick phantom was demonstrated; to the best of our knowledge, no AR-OAM setup has successfully demonstrated imaging of such thick samples without vertical scanning, even when synthetic aperture techniques were used. The extended depth of field offered by OMT may enable imaging of surfaces with high curvatures or facilitated deeper imaging when combined with lasers of longer wavelengths.

The imaging speed of epi-illumination OMT may be significantly improved by replacing the single-element acoustic detector with an array. Specifically, SPADE arrays may be produced with a pitch down to 10μm, enabling fabrication of whole 1D arrays capable of capturing an entire B-scan from a single optoacoustic pulse excitation. Such arrays would require replacing the signal readout system with scalable techniques, e.g., using pulsed lasers [26, 33, 34]. Alternatively, one may fabricate a small number of SPADEs, and perform parallel readout by duplication of the readout system shown in Fig.3a. Such a system would still offer enhancement in imaging speed beyond OR-OAM, which is inherently limited to a single-detector configuration.

Acknowledgements

We thank the Micro-Nano Center at the Technion for the use of the clean room facilities for the SPADE fabrication.

Abbreviations

AR-PAM	Acoustic Resolution Photoacoustic Microscopy
BP	Back Projection
MZI	Mach Zehnder Interferometer
NEP	Noise Equivale Power
OAT	Optoacoustic Tomography
PA	Photo Acoustics
SPADE	Silicon-Photonics Acoustic Detector

Funding

This work has received funding from the Israel Science Foundation (1709/20 A.R.).

Code Availability

The data files generated during and/or analysed during the current study are available from the corresponding author on a reasonable request.

Declarations

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Conflict of Interest

The authors declare no conflicts of interest.

Ethics approval

IL-168-12-19.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Ntziachristos V. Going deeper than microscopy: the optical imaging frontier in biology. Nat Methods. 2010;7:603. doi:10.1038/nmeth.1483. [PubMed] [CrossRef] [Google Scholar]

2. Taruttis A. Ntziachristos. Advances in real-time multispectral optoacoustic imaging and its applications. Nat Photon. 2015;9:219–27. doi:10.1038/nphoton.2015.29. [CrossRef] [Google Scholar]

3. Wang LV. Photoacoustic tomography: in vivo imaging from organelles to organ. Science. 2012;335:1458–62. doi:10.1126/science.1216210. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

4. Rosenthal A, Ntziachristos V, Razansky D. Acoustic inversion in optoacoustic tomography: A Review.Curr. Med. Imaging 2013; Rev.9,318–336. [PMC free article] [PubMed]

5. Omar M, Aguirre J, Ntziachristos V. Optoacoustic mesoscopy for biomedicine. Nat Biomed Eng. 2019;3:354–70. doi:10.1038/s41551-019-0377-4. [PubMed] [CrossRef] [Google Scholar]

6. Baik JW, Kim JY, Cho S, Choi S, Kim J, Kim C. Super wide-field photoacoustic microscopy of animals and humans in vivo. IEEE Trans Med Imaging. 2020;39:975–84. doi:10.1109/TMI.2019.2938518. [PubMed] [CrossRef] [Google Scholar]

7. Liu W. Yao. Photoacoustic microscopy: principles and biomedical applications. Biomed Eng Lett. 2018;8:203–13. doi:10.1007/s13534-018-0067-2. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

8. Hu S, Wang LV. Optical-resolution Photoacoustic Microscopy: auscultation of biological systems at the cellular level. Biophys J. 2013;105(4):841–7. doi:10.1016/j.bpj.2013.07.017. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

9. Park S, Lee C, Kim J, Kim C. Acoustic resolution photoacoustic microscopy. Biomed Eng Lett. 2014;4:213–22. doi:10.1007/s13534-014-0153-z. [CrossRef] [Google Scholar]

10. Zhang HF, Maslov K, Stoica G, Wang LV. Functional photoacoustic microscopy for high-resolution and noninvasive in vivo imaging. Nat Biotechnol. 2006;24:848–51. doi:10.1038/nbt1220. [PubMed] [CrossRef] [Google Scholar]

11. Jeon S, Park J, Managuli R, Kim CA. Novel 2-D synthetic aperture focusing technique for acoustic-resolution photoacoustic microscopy. IEEE Trans Med Imaging. 2019;38:250–60. doi:10.1109/TMI.2018.2861400. [PubMed] [CrossRef] [Google Scholar]

12. Liao CK, Li ML, Li PC. Optoacoustic imaging with synthetic aperture focusing and coherence weighting. Opt Lett. 2004;29(21):2506–8. doi:10.1364/OL.29.002506. [PubMed] [CrossRef] [Google Scholar]

13. Shnaiderman R, Mustafa Q, Ülgen O, Wissmeyer G, Estrada H, Razansky D, Chmyrov A. V. Ntziachristos. Silicon-photonics point sensor for high-resolution optoacoustic imaging.Advanced Optical Materials, 2021.

14. Hazan Y, Levi A, Nagli AM, Rosenthal A. Silicon-photonics acoustic detector for optoacoustic micro-tomography. Nat Commun. 2022;13:1488. doi:10.1038/s41467-022-29179-7. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

15. Jacques SL. Optical properties of biological tissues: a review. Phys. Med. Biol. 2013;58. [PubMed]

16. S. A. Prahl, M. Keijzer, S. L. Jacques, and A. J. Welch. A Monte Carlo model of light propagation in tissue. 1989; Proc. SPIE 10305.

17. A. A Leion, A. Pulkkinen, and T. Tarvainen. ValoMC: a Monte Carlo software and MATLAB toolbox for simulating light transport in biological tissue. 2019.

18. E. Bradley, T. Treeby B. Cox.k-Wave: MATLAB toolbox for the simulation and reconstruction of photoacoustic wave fields. J. Biomed. Opt. 2010;15 (2)021314. [PubMed]

19. k-Wave A MATLAB toolbox for the time domain simulation of acoustic wave fields User Manual

20. M. Xu, and L. V. Wang. Universal back-projection algorithm for photoacoustic computed tomography. Phys. Rev. 2005; E 71, 016706. [PubMed]

21. Z. Or, AR. Levi, Y. Hazan, A. Rosenthal. Hand-Held optoacoustic system for the localization of mid-depth blood vessels. Photonics, 2022; 9(12):907. 10.3390/photonics9120907

22. S. Tsesses, D. Aronovich, A. Grinberg, E. Hahamovich, and A. Rosenthal. Modeling the sensitivity dependence of silicon-photonics-based ultrasound detectors. Opt. Lett. 2017; 42, 5262–5265. [PubMed]

23. A. Rosenthal, S. Kellnberger, G. Sergiadis and V. Ntziachristos. Wideband fiber- interferometer stabilization with variable phase. IEEE Photonics Technology Letters. 2012; vol. 24(17) 1499–1501.

24. L. Riobó, Y. Hazan, F. Veiras, M. Garea, P. Sorichetti, and A. Rosenthal. Noise reduction in resonator-based ultrasound sensors by using a CW laser and phase detection. Opt. Lett. 44. 2019; 2677–2680.

25. J. Aguirre, M. Schwarz, N. Garzorz, M. Omar, A. Buehler, K. Eyerich and V. Ntziachristos. Precision assessment of label-free psoriasis biomarkers with ultra- broadband optoacoustic mesoscopy. Nat Biomed Eng. 2017; 0068

26. Y. Hazan, and A. Rosenthal. Simultaneous multi-channel ultrasound detection via phase modulated pulse interferometry. Opt. Express. 2019; OE 27, 28844–28854. [PubMed]

27. W.J. Westerveld, M.Ul. Hasan, R. Shnaiderman, V. Ntziachristos, X. Rottenberg, S. Severi, V. Rochus. Sensitive, small, broadband and scalable optomechanical ultrasound sensor in silicon photonics. Nature Photonics. 2021; 1–5. doi:10.1038/s41566-021-00776-0.

28. E. Zhang, J. Laufer, P. Beard. Backward-mode multiwavelength photoacoustic scanner using a planar Fabry-Perot polymer film ultrasound sensor for high-resolution three-dimensional imaging of biological tissues. Appl. Opt. 2008; AO 47, 561–577. [PubMed]

29. Q. Rong, Y. Lee, Y. Tang, T. Vu, C. Taboada, W. Zheng, J. Xia, D.A. Czaplewski, H.F. Zhang, C. Sun, J. Yao. High-frequency 3D photoacoustic computed tomography using an optical micro ring resonator. BME Frontiers. 2022; 2022, ID:9891510. [PMC free article] [PubMed]

30. Liang Y, Fu W, Li Q, Chen X, Sun H, Wang L, Jin L, Huang W, Guan BO. Optical-resolution functional gastrointestinal photoacoustic endoscopy based on optical heterodyne detection of ultrasound. Nat Commun. 2022 Dec 9;13(1):7604. [PMC free article] [PubMed]

31. I. A. Veres, P. Burgholzer, T. Berer, A. Rosenthal, G. Wissmeyer, and V. Ntziachristos. Characterization of of the spatio-temporal response of optical fiber sensors to incident spherical waves. The Journal of the Acoustical Society of America, 2014; 135, 1853. 10.1121/1.4868391 [PubMed]

32. A. Rosenthal, M. Ángel, A. Caballero, S. Kellnberger, D. Razansky, and V. Ntziachristos. Spatial characterization of the response of a silica optical fiber to wideband ultrasound.Opt. Lett., 2012; 37, 3174–3176. 10.1364/OL.37.003174 [PubMed]

33. J. Pan, Q. Li, Y. Feng, R. Zhong, Z. Fu, S. Yang, W. Sun, B. Zhang, Q. Sui, J. Chen,Y. Shen and Z. Li. Parallel interrogation of the chalcogenide-based micro-ring sensor array for photoacoustic tomography. Research square, 2022; 10.21203/rs.3.rs-1965703/v1 [PMC free article] [PubMed]

34. Y. Hazan, M. Nagli, A. Levi, A. Rosenthal. Miniaturized ultrasound detector arrays in silicon photonics using pulse transmission amplitude monitoring. Optics Letters. 2022;47(21). [PubMed]

Articles from Biomedical Engineering Letters are provided here courtesy of Springer