Open Access
Add to BibSonomy
Abstract
In this study, we evaluate the penetration capability of light in visible, near-infrared-I (NIR-I) and near-infrared–II (NIR-II) optical windows for photoacoustic macroscale imaging inside 9 biological tissues with three typical penetration depths. An acoustic resolution photoacoustic microscopy is designed to guarantee the consistent experiment conditions except excitation wavelength. Experimental results show that short NIR-II (1000-1150 nm) shows the best performance inside kidney, spleen and liver tissues at all depths, while NIR-I (700-1000 nm) works better for muscle, stomach, heart and brain tissues, especially in deep imaging. This study proposes the optimal selection of illumination wavelengths for photoacoustic macroscale imaging in rat organs, which enables the best signal-to-noise ratio (SNR) of the observed target.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Photoacoustic imaging (PAI) is mainly based on the optical absorption and Grüneisen parameter of the observed target, which absorbs photon energy and generates ultrasound waves [1–4]. By the virtue of rich optical contrast, deep penetration depth, high spatial resolution, PAI has been extensively applied in multiple interdisciplinary fundamental fields, such as cancer biology, nanomedicine, drug delivery and theranostics [5–10]. Besides, PAI shows great potential in clinical diagnosis of breast cancer, arthritis and image-guided surgery [11–14]. In general, PAI is classified into photoacoustic computed tomography (PACT) [15–17], acoustic resolution photoacoustic microscopy (AR-PAM) [18,19], and optical resolution photoacoustic microscopy (OR-PAM) [20–27]. OR-PAM is able to achieve a high lateral resolution of several microns, but suffers from a limited penetration depth of up to ∼1.5 mm [28]. Both AR-PAM and PACT feature centimeter scale penetration depth with a submillimeter spatial resolution, making it useful in more scenarios compared with OR-PAM, especially in clinics [3, 29].
In theory, attenuation is the dominant tissue property determining the light penetration depth, in which optical scattering plays a more important role in near-infrared (NIR) windows and optical absorption makes more contribution in the ultraviolet (UV 180-400 nm) and visible (VS 400-700 nm) windows [30]. UV photoacoustic can image individual cell nuclear without labeling and identify small clusters of cancer cells by observing nuclear and packing density [31]. VS light is widely used for imaging the vascular system in biological tissues by ORPAM to derive high resolution functional parameters such as oxygen saturation and blood flow [32–35]. Previous theoretical and experimental studies show that NIR-I (700-1000 nm) and NIR-II (1000-1700 nm) enable a better penetration capability compared with UV and VS since both optical absorption of hemoglobin and optical scattering of biological tissues decrease significantly with the increase of wavelength [36]. Therefore, to balance the spatial resolution and penetration capability, photoacoustic macroscale imaging coupled with NIR light is more applicable in biomedical studies. Besides the utilization of endogenous molecules, exogenous contrast agents, especially NIR imaging enhancers, have been extensively used in photoacoustic macroscale imaging to achieve deep penetration capability [37–39]. With the increase of excitation wavelength, the absorption of tissue constituents other than hemoglobin, such as water, lipid, becomes dominant. When the wavelength reaches mid-infrared (MIR 3000-5000 nm), the strong absorption of water and lipid enables high resolution, high contrast microscopic imaging of multiple chemical and structural information inside biological tissues [40]. Unfortunately, the strong optical absorption of water in MIR prevents it from macroscale imaging, especially for the targets located in deep tissues.
Although many previous studies have investigated the transport and interaction of light inside biological tissues based on the optical properties of chromophores such as absorption, scattering, anisotropy, and refractive index [41–43], biological organs are always formed by complicated combinations of various chromophores, making it more challenging to be simply simulated, calculated and theoretical analyzed using optical properties of chromophores. In addition, the principle of photoacoustic imaging is quite different from pure optical imaging. Besides optical absorption, the efficiency of energy conversion from electromagnetic energy to mechanical vibration is also determined by the mechanical properties of the observed target [44,45].
In this study, we evaluate photoacoustic macroscale imaging performance in organs using experimental data and interpret the data using existing knowledge of tissue components and optical properties. In detail, we investigated the photoacoustic spectrums of abdominal muscle, stomach, heart, kidney, liver, spleen, brain, fat, and skin within VS, NIR-I and NIR-II windows. We specially designed an AR-PAM platform with removal of all the experimental influence impacts except excitation wavelength at three typical penetration depths. Based on the experiments, all the selected tissues, especially the tissues with rich content of hemoglobin, show significant optical absorption in VS, indicating that VS is not a good choice for photoacoustic macroscale imaging. Both NIR-I and short NIR-II (1000-1150 nm) have a better performance in all selected penetration depths for all tissues compared with VS. Through investigation of penetration capability in depths, short NIR-II shows the best performance inside kidney, spleen and liver tissues at all depths, while NIR-I works better for muscle, stomach, heart and brain tissues in deep imaging.
2. Methods and materials
2.1 System configuration and experimental procedure
Figure 1(a) shows the configuration of a specially designed AR-PAM. To cover VS, NIR-I and NIR-II optical windows, we combined light beams from two separate OPO lasers using a removable reflection mirror (GCC-1021, Daheng Optics, China). The first laser (repetition rate: 20 Hz, pulse width: 7 ns) covers VS from 450 nm to 650 nm. The emission rang of the second laser (repetition rate: 20 Hz, pulse width: 7 ns) is from 700 nm to 1600 nm. Both beams are filtered using spatial filters. Both filters consist of two objectives and one pinhole with a diameter of 50 µm to form a clean Gaussian beam. Considering the variation of beam divergence in different wavelengths, we mounted the pinholes on motorized linear stages, tested and recorded the best position of the pinhole for each selected wavelength before the experiments. Post collimation, we roughly recorded the laser spots of 600 nm and 1100 nm from each laser as shown in insets in Fig. 1(b). The reflected laser beam is evenly irradiated to the biological tissue from the bottom of the water tank to avoid the optical absorption of the coupling medium, especially in NIR-II window. Besides, a ground glass located in front of the biological tissue diffuses the illumination beams to avoid the inhomogeneous excitation pattern on the tissue surface. Optical density of all wavelengths is adjusted to be 2.0 mJ before passing through the ground glass. We design a special sample holder as shown in Fig. 1(b). The height and diameter of the cylindrical holder are 21 mm and 10 mm, respectively. There are three through holes located at 5 mm, 10 mm, and 20 mm over the bottom surface of the holder to allocate targets and simulate the penetration depths. A 1 mm diameter tube fulfilled with Indian ink serves as the target. We understand that it is preferred to evaluate the photoacoustic performance at every depth and then provide a tabulated threshold for accurate guideline of optimal wavelength selection. Unfortunately, unlike the controllable situations in simulations and phantom experiments, it is much more complicated inside real biological tissues. Even the specimen resected from the same organ will not be the same. It is quite challenging to cover all the scenarios in experiments. Hence, we select three different penetration depths, which can meet the required penetration depth in small animal experiments using photoacoustic macroscale imaging [46,47].
Fig. 1. (a) Schematic configuration of the AR-PAM system. (b) Detailed design of the sample holder indicated by the dashed blue box in (a) and recorded spots of two laser beams post spatial filtering. (c) Measured photoacoustic spectrum of Indian ink in the optical window of 400∼1600 nm. G, ground glass; H, sample holder; L, excitation light; L1, lens; I, ink tube; M, mirror; Ph, pinhole; S, sample organ; T, transducer; WT, water tank.
A 7.5 MHz spherically focused ultrasound transducer (V320, Olympus) with an active area of 12.7 mm and a focal length of 24.5 mm is mounted on a motorized linear stage to perform a one-dimensional scan with a step size of 0.1 mm and a scanning distance of 10 mm. We used a pulse/receiver with an amplifying rate of ∼39 dB (5073PR, Olympus) to amplify the photoacoustic signals, and digitalized the signal using a data acquisition card (PCI-5122, National Instruments Corporation) at a sampling rate of 100 MS/s. Each depth-resolved signal, named as A-line, was generated with an averaging time of 5 to obtain a better signal-to-noise ratio (SNR), especially when the target located deep inside the tissue. Similar to B-mode ultrasound imaging, we directly back-projected the depth-resolved photoacoustic signal at each scanning position to form a two-dimensional cross-sectional image, referred to as B-scan.
2.2 Sample preparation
Tissue preparation: Female Sprague-Dawley rats, weighting 280-350 g, were used in this study. We carried out surgeries to resect fresh biological tissues from organs. The rats were anesthetized by injecting sodium pentobarbital at a dose of 30 µl/g. Before resection of the organs, all the large blood vessels cross the organs were ligated to prevent major bleeding and maintain biological organizational integrity. Besides organs, both muscle and fat were resected from the rat abdominal wall. Post the tissue resection, the rats were sacrificed using a standard procedure. All the operations have been approved by the ethic committee at the Southern University of Science and Technology.
Imaging target preparation: In order to remove the experimental impact generated by the photoacoustic spectrum of the imaging target, we chose India ink that owns a relatively flat absorption spectrum over VS, NIR-I and -II windows. To match the center frequency of the transducer, we infused the diluted India ink with a ratio of 1:20 in a 1 mm plastic tube with a wall thickness of 0.2 mm. Two ends of the tube were sealed using an agarose phantom. Before we did quantitative analysis, an experimental photoacoustic spectrum of India ink was used to calibrate the original photoacoustic signals.
2.3 Data processing
We used MATLAB 2016b (MathWorks, Inc.) to process the data. To derive the relative optical absorption of ink tubes and organs, we carried out Hilbert transform of all the A-lines. In addition, the data used for both image reconstruction and quantitative analysis were calibrated using the measured photoacoustic spectrum of the India ink. We carried out 5 independent experiments of each organ by two individuals, and shown the data in the form of Mean ${\pm} $ SEM (standard error of mean).
3. Results and discussions
Figure 1(c) shows the photoacoustic spectrum of the Indian ink within the optical window of 400∼1600 nm. The signal amplitude is comparable in VS and NIR-I, and slightly decreases in NIR-II. There is a peak signal around 1500 nm, which is mainly contributed by water in the diluted ink solution. We applied this measured photoacoustic spectrum to calibrate the original photoacoustic signals generated by the ink tubes with different excitation wavelengths at penetration depths of the biological tissues. Therefore, the calibrated photoacoustic signals of ink directly reflect the energy of penetrating light coupled with the inherent photoacoustic convention of the ink.
Black curves in Figs. 2(a)–2(c) show the measured photoacoustic spectrums of muscle, stomach and heart, respectively. The main absorbing chromospheres in biological tissues include hemoglobin, myoglobin, melanin, fat and water. The abdominal muscle belongs to skeletal muscle that contains a large amount of myoglobin and a small amount of hemoglobin. Hence, as shown in Fig. 2(a), the photoacoustic spectrum has a peak value in VS window (400-650 nm) due to the strong optical absorption of hemoglobin, and a low absorption over NIR-I and short NIR-II windows because of the decreased optical absorption of hemoglobin. When the excitation wavelength is longer than 1250 nm in NIR-II, the photoacoustic signal of the muscle tissue increases significantly contributed by water. Red curves present the photoacoustic spectrum of the ink tube embedded at 5 mm beneath the tissue surface with VS, NIR-I and NIR-II light excitation. When the wavelength is below 650 nm, contributed by the ultra-high optical absorption and scattering properties of the abdominal muscle, the photoacoustic signal of the ink is much lower than that of the ink excited by the NIR-I laser pulses. Within the wavelengths from 700 nm to 1050 nm, the photoacoustic amplitude remains almost the same with a slight decrease at the wavelength around 950 nm. When the excitation wavelength is longer than 1150 nm, the photoacoustic amplitude decreases suddenly indicating that the photon energy inside the tissue has been significantly attenuated. This phenomenon is majorly contributed by the increased optical absorption of water in the tissue.
Fig. 2. The photoacoustic spectrums of ink and tissue in (a) muscle, (b) stomach and (c) heart. Black and red curves represent the photoacoustic spectrums of the tissue and ink, respectively. (d)-(f) B-scan images of ink and organ at wavelength of 550, 1050 and 1350 nm. Scale bar: 1 mm.
To remove the influence of digested food in the stomach, rats were only treated with water at 24 hours prior to the surgery. The wall of the rat stomach composes three layers of tissues including mucosal layer, serosa layer, and smooth muscle. Although myoglobin does not exist inside smooth muscle, the mucosal layer of the stomach wall has rich blood vessels and there are a bunch of blood vessels surrounding the outer surface of the stomach. Figure 2(b) shows the photoacoustic spectrums of stomach (black), and the ink (red) embedded inside it. Contributed by the larger amount of hemoglobin inside the stomach, the photoacoustic signal is stronger than abdominal muscle within VS window, leading to a weaker photoacoustic amplitude of the embedded ink. Similar to abdominal muscle, the stomach has a lower optical absorption within the NIR-I and short NIR-II windows, resulting in a strong photoacoustic amplitude of the embedded ink tube. When the excitation wavelength is above 1250 nm, the water inside stomach tissues absorb most of the excitation light because of increased absorption coefficient and thus the photoacoustic amplitude of the ink decreases significantly.
To avoid the massive loss of blood, the blood vessels were ligated before the resection of the heart. The ventricle and atrium of the heart is fully filled with blood and the myocardium contains more myoglobin. Therefore, the heart has the strongest absorption in VS windows compared with the stomach and abdominal muscle, leading to an extremely low penetration depth of the visible light. As shown in Fig. 2(c), the heart has the similar photoacoustic spectrum with the stomach and abdominal muscle tissues in NIR-I and NIR-II windows.
The B-scans in Figs. 2(d)–2(f) present the typical images of corresponding organ surfaces and ink tubes located at 5 mm beneath the organ surfaces with excitation wavelengths of 550 nm, 1050 nm and 1350 nm, respectively. The ink tubes have the best contrast with NIR-I and short NIR-II illumination because both hemoglobin and water prevent the optical penetration depth in VS and long NIR-II (1150-1700 nm) windows. The images are well consistent with the spectrum measurements. Inside some B-scans, we clearly observe artifacts surrounding the imaged tissue surfaces. These artifacts were mainly contributed by the reflection of the photoacoustic signal from tissue-tissue and tissue-water interfaces inside the organs.
The abdominal muscles, stomach, and heart are made up of muscles. The tissue absorption is low at spectral regions in NIR-I and short NIR-II windows, resulting in a better penetration capability compared with VS window. In addition, different tissues have distinct contents of myoglobin and hemoglobin, which results in specific photoacoustic spectrums.
Figure 3 shows the photoacoustic spectrums (black curves) of kidney, spleen and liver, respectively, and the photoacoustic spectrums (red curves) of the India ink embedded inside the tissues. Kidney, spleen and liver contain a large amount of blood. Hence, the surrounding blood vessels are ligated before the resection of organs to avoid bleeding. The B-scan images of three organs and inserted ink tubes are shown in Figs. 3(d)–3(f). Suffering from ultrahigh optical absorption of hemoglobin, the visible light has been extensively attenuated before they reach the ink tubes, which leads to ultralow photoacoustic amplitudes of ink inside these three organs. Benefiting from the low optical absorption coefficient of hemoglobin, NIR-I and short NIR-II enable less light attenuation compared with VS. Different from the scenarios in abdominal muscle, stomach and heart, the excitation wavelengths from 1000 nm to 1100 nm in NIR-II allows a deeper penetration depth compared with NIR-I wavelengths. Through absorbing more photons, the photoacoustic amplitude of ink in this window is the highest at 1050 nm. Similarly, when the excitation wavelengths in NIR-II window is longer than 1250 nm, the increased absorption coefficient of water prevents the penetration capability of light, making it difficult for photoacoustic imaging using wavelengths longer than 1250 nm even they have less scattering compared with shorter wavelengths.
Fig. 3. The photoacoustic spectrums of ink and tissue in (a) kidney, (b) spleen and (c) liver. Black and red curves represent the photoacoustic spectrums of the tissue and ink, respectively. (d)-(f) B-scan images of ink and organ at wavelength of 550, 1050 and 1350 nm. Scale bar: 1 mm.
Fat, a subgroup of lipids, is one of the most important biological tissues in the animal body, and forms many important organs. The black and red curves in Fig. 4(a) show the photoacoustic spectrum of subcutaneous fat and embedded ink with multispectral excitations, respectively. Although there is less blood vessels inside the fat compared with many other important organs, the optical absorption of hemoglobin in the VS window is still obvious and limits the penetration capability of visible light. Based on our further investigation, the adipocytes inside the fat have very strong optical absorption within VS window, leading to a significant light attenuation. When we used NIR lights, the photoacoustic amplitude of ink becomes stronger due to the decrease of the optical absorption of adipocytes and hemoglobin. Compared with other organs, the light absorption of fat at 900 nm is stronger than 1000 nm contributed by glycerol and fatty acids. Interestingly, fat owns a weak absorption at 1250 nm due to the low absorption of lipid, and allows a deeper penetration depth compared with other organs. When the excitation wavelength is longer than 1500 nm that is the peak absorption of water, the photoacoustic amplitude of fat decreases with the increase of the excitation light. This may be caused by decreased water absorption of light.
Fig. 4. The photoacoustic spectrums of ink and tissue in (a) Fat, (b) Brain and (c) Skin. Black and red curves represent the photoacoustic spectrums of tissues and ink, respectively. (d)-(f) B-scan images of ink and organ at wavelength of 550, 1050 and 1350 nm. Scale bar: 1 mm.
Brain, the most important organ, contains a lot of neuronal cells, nerve fibers, blood vessels and lipids. The optical absorption of brain tissue is mainly affected by both hemoglobin and lipids. From Fig. 4(b), similar to all previous organs, the penetration depth using VS light is limited due to the high attenuation of hemoglobin. NIR-I and short NIR-II works better compared with VS lights. However, the optical absorption of hemoglobin and water significantly decrease photoacoustic amplitude of ink tubes at 1000 nm. We observed the same phenomena in other organs that contained a large amount of blood, such as kidney, spleen and heart. Similarly, when the wavelength is above 1250 nm, water dominates the optical attenuation, leading to a decrease of light penetration.
Skin, the largest organ of mammalian, which can be divided into epidermis and dermis, and contains melanophores, lipids, nerve fibers and blood vessels. The thickness of epidermal tissue is about 4 mm in rats. Figure 4(c) shows the photoacoustic spectrums of skin (black curve) and ink tube (red curve). Compared with NIR I and short NIR II, the energy of VS light through skin and muscle is much lower due to the high absorption of melanophores and hemoglobin. In addition, we found that the photoacoustic signal of ink decreases significantly because of the increased absorption coefficient of water, when the wavelength is over 1250 nm. B-scan images in Figs. 4(d)–4(f) are consistent with the spectrum measurements.
The photoacoustic amplitudes of biological tissues and ink tubes are different since different tissues have different optical performance. Tables 1 and 2 summarize the measured photoacoustic amplitudes of ink tubes embedded in 9 different biological tissues and excited with different wavelengths in VS, NIR-I and NIR-II windows.
Table 1. The measured photoacoustic amplitudes of ink embedded in 9 tissues at 400-1050 nm.
Table 2. The measured photoacoustic amplitudes of ink embedded in 9 tissues at 1100-1600 nm.
Deep imaging always remains the most attractive topic in both fundamental and clinical studies using optical imaging modalities. For whole-body imaging of small animals, the maximum penetration of 20 mm is sufficient [48]. To evaluate the optical penetration inside different biological tissues (muscle, stomach, heart, brain, kidney, spleen and liver) at different wavelengths, we performed the imaging experiments of 7 tissues except fat and skin with three typical embedded depths of ink tubes since the thicknesses of fat and skin are normally less than 5 mm in small animals. Based on early studies, we are clear that NIR-I and short NIR-II enables deeper penetration in all selected tissues. Hence, we only investigated the optical penetration of NIR-I and short NIR-II lights with a tuning step of 50 nm.
Figures 5(a)–5(d) show the photoacoustic spectrums of ink tubes in muscle, stomach, heart and brain with penetration depths of 5 mm (black), 10 mm (red), and 20 mm (blue), respectively. We magnified the blue curves to show the detailed spectrums at Figs. 5(e)–5(h). As we expected, when the penetration depth increases, the photoacoustic amplitude of ink tube becomes weaker for all four organs. However, from the magnified blue curves, we find the photoacoustic spectrum changes with the increase of penetration depths. As indicated by the blue arrows, the best illumination wavelength changes from short NIR-II light to NIR-I light when we performed photoacoustic imaging of ink tubes embedded 20 mm inside muscle, stomach, heart and brain. As shown in Fig. 6, inside kidney, spleen, and liver, the photoacoustic spectrums are almost the same in all depths, in which short NIR-II reveals a better penetration capability compared with NIR-I. This discrepancy of photoacoustic spectrums in different depths might be caused by hemoglobin. Optical absorption of hemoglobin at NIR-I is higher than that of other chromospheres including myoglobin, fat and water, leading to a better penetration performance of short NIR-II inside vascular- or blood- rich organs such as kidney, spleen, and liver. Heart, which contains a lot of blood, is supposed to show the same performance with blood-rich organs. However, when we embedded the ink tube into the heart, it is inevitable to avoid the severe loss of blood. The measured photoacoustic amplitudes of the ink tube at different depths with NIR-I and short NIR-II excitation are summarized in Table 3. Table 4 roughly provides a suggestion for wavelength selection based on our experimental data.
Fig. 5. (a) - (d) The photoacoustic spectrums of ink located at at the depth of 5 mm (black), 10 mm (red) and 20 mm (blue) in muscle, stomach, heart and brain, respectively. (e) – (h) The zoomed figures of blue curves in (a) – (d). The peak values of blue curves are marked by blue arrows.
Fig. 6. (a) - (c) The photoacoustic spectrums of ink tubes located at the depths of 5 mm (black), 10 mm (red) and 20 mm (blue) in kidney, spleen and liver. (d) - (f) The magnified figures of the blue curves in (a) - (c). The peak values of blue curves are marked by blue arrows.
Table 3. The measured data of the ultrasonic signal of ink as generated in seven different tissues at 700-1200 nm.
Table 4. The suggested selection of wavelength for different organs/tissues.
4. Conclusion
In this study, we evaluate the penetration capability of VS, NIR-I, and NIR-II light for photoacoustic macroscale imaging inside 9 biological tissues. Based on the experimental results, the excitation wavelength can thus be properly selected in order to achieve the best SNR and imaging depth. It turns out that wavelengths in short NIR-II window are ideal for macroscale photoacoustic bio-imaging inside all 9 tissues at the superficial penetration depth. The ideal wavelength to observe peaks of photoacoustic signal amplitude from ink tube as embedded in kidney, spleen and liver appears at short NIR-II window for all depths. Apart from blood-rich organs/tissues, NIR-I works better for muscle, stomach, heart and brain in deep tissues compared with NIR-II. However, the penetration capability of VS and long NIR-II window is limited by the ultra-high optical absorption of hemoglobin and water, respectively. This study can provide a rough suggestion for the selection of excitation wavelengths according to the type of tissues and the aiming imaging depth using photoacoustic macroscale bio-imaging using intrinsic molecules. In addition, different from intrinsic chromophores with inherent absorption spectrum, the absorption spectrums of exogenous molecules are controllable. Hence, this study will offer a suggestion for the absorption spectrum optimization with consideration of background tissues.
Funding
National Natural Science Foundation of China (61775028, 81571722); Southern University of Science and Technology (Startup grant).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. A. A. Oraevsky, S. L. Jacques, R. O. Esenaliev, and F. K. Tittel, “Laser based optoacoustic imaging in biological tissues,” Proc. SPIE 2134A, 16 (1994). [CrossRef]
2. L. V. Wang, X. Zhao, H. Sun, and G. Ku, “Microwave-induced acoustic imaging of biological tissues,” Rev. Sci. Instrum. 70(9), 3744–3748 (1999). [CrossRef]
3. H. Zhang, K. Maslov, G. Stoica, and L. V. Wang, “Functional photoacoustic microscopy for high-resolution and noninvasive in vivo imaging,” Nat. Biotechnol. 24(7), 848–851 (2006). [CrossRef]
4. L. V. Wang and S. Hu, “Photoacoustic tomography: in vivo imaging from organelles to organs,” Science 335(6075), 1458–1462 (2012). [CrossRef]
5. M. Mehrmohammadi, S. Joon Yoon, D. Yeager, and S. Emelianov, “Photoacoustic imaging for cancer detection and staging,” Curr. Mol. Imaging 2(1), 89–105 (2013). [CrossRef]
6. S. Valluru and J. Willmann, “Clinical photoacoustic imaging of cancer,” Ultrasonography 35(4), 267–280 (2016). [CrossRef]
7. S. Changbeom, K. Haemin, M. Hyungwon, L. Hohyeon, C. Jin, and K. Hyuncheol, “Photoacoustic-based nanomedicine for cancer diagnosis and therapy,” J. Controlled Release 203, 118–125 (2015). [CrossRef]
8. X. Yang, E. Stein, S. Ashkenazi, and L. V. Wang, “Nanoparticles for photoacoustic imaging,” WIREs Nanomed. Nanobiotechnol. 1(4), 360–368 (2009). [CrossRef]
9. H. Shangguan, L. Casperson, A. Shearin, K. Gregory, and S. Prahl, “Photoacoustic drug delivery: the effect of laser parameters on the spatial distribution of delivered drug,” Proc. SPIE 2391, 394 (1995). [CrossRef]
10. C. Moore and J. Jokerst, “Strategies for Image-Guided Therapy, Surgery, and Drug Delivery Using Photoacoustic Imaging,” Theranostics 9(6), 1550–1571 (2019). [CrossRef]
11. R. Kruger, R. Lam, D. Reinecke, S. Rio, and R. Doyle, “Photoacoustic angiography of the breast,” Med. Phys. 37(11), 6096–6100 (2010). [CrossRef]
12. S. Ermilov, T. Khamapirad, A. Conjusteau, M. Leonard, R. Lacewell, K. Mehta, T. Miller, and A. Oraevsky, “Laser optoacoustic imaging system for detection of breast cancer,” J. Biomed. Opt. 14(2), 024007 (2009). [CrossRef]
13. J. Jo, C. Tian, G. Xu, J. Sarazin, E. Schiopu, G. Gandikotac, and X. Wang, “Photoacoustic tomography for human musculoskeletal imaging and inflammatory arthritis detection,” Photoacoustics 12, 82–89 (2018). [CrossRef]
14. L. Xi, G. Zhou, N. Gao, L. Yang, D. Gonzalo, S. Hughes, and H. Jiang, “Photoacoustic and Fluorescence Image-Guided Surgery Using a Multifunctional Targeted Nanoprobe,” Ann. Surg. Oncol. 21(5), 1602–1609 (2014). [CrossRef]
15. M. Xu and L. V. Wang, “Universal back-projection algorithm for photoacoustic computed tomography,” Phys. Rev. E 71(1), 016706 (2005). [CrossRef]
16. D. Razansky, M. Distel, C. Vinegoni, R. Ma, N. Perrimon, R. Köster, and V. Ntziachristos, “Multispectral opto-acoustic tomography of deep-seated fluorescent proteins in vivo,” Nat. Photonics 3(7), 412–417 (2009). [CrossRef]
17. N. Huang, H. Guo, W. Qi, Z. Zhang, J. Rong, Z. Yuan, W. Ge, H. Jiang, and L. Xi, “Whole-body multispetral photoacoustic imaging of adult zebrafish,” Biomed. Opt. Express 7(9), 3543–3550 (2016). [CrossRef]
18. K. Song and L. V. Wang, “Deep reflection-mode photoacoustic imaging of biological tissue,” J. Biomed. Opt. 12(6), 060503 (2007). [CrossRef]
19. D. Cai, Z. Li, and S. Chen, “In vivo deconvolution acoustic-resolution photoacoustic microscopy in three dimensions,” Biomed. Opt. Express 7(2), 369–380 (2016). [CrossRef]
20. C. Zhang, K. Maslov, and L. V. Wang, “Subwavelength-resolution label-free photoacoustic microscopy of optical absorption in vivo,” Opt. Lett. 35(19), 3195–3197 (2010). [CrossRef]
21. J. Yao, C. Huang, L. Wang, J. Yang, L. Gao, I. Kinstantin, Z. Jou, and L. V. Wang, “Wide-field fast-scanning photoacoustic microscopy based on a water-immersible MEMS scanning mirror,” J. Biomed. Opt. 17(8), 080505 (2012). [CrossRef]
22. Q. Chen, T. Jin, W. Qi, X. Mo, and L. Xi, “Label-free photoacoustic imaging of the cardio-cerebrovascular development in the embryonic zebra fish,” Biomed. Opt. Express 8(4), 2359–2367 (2017). [CrossRef]
23. T. Jin, H. Guo, H. Jiang, B. Ke, and L. Xi, “Portable optical resolution photoacoustic microscopy (pORPAM) for human oral imaging,” Opt. Lett. 42(21), 4434–4437 (2017). [CrossRef]
24. W. Qin, W. Qi, T. Jin, H. Guo, and L. Xi, “In vivo oral imaging with integrated portable photoacoustic microscopy and optical coherence tomography,” Appl. Phys. Lett. 111(26), 263704 (2017). [CrossRef]
25. W. Qin, Q. Chen, and L. Xi, “A handheld microscope integrating photoacoustic microscopy and optical coherence tomography,” Biomed. Opt. Express 9(5), 2205–2213 (2018). [CrossRef]
26. C. Song, T. Jin, R. Yan, W. Qi, T. Huang, H. Ding, S. Tan, N. Nguyen, and L. Xi, “Opto-acousto-fluidic microscopy for three-dimensional label-free detection of droplets and cells in microchannels,” Lab Chip 18(9), 1292–1297 (2018). [CrossRef]
27. Q. Chen, H. Xie, and L. Xi, “Wearable optical resolution photoacoustic microscopy,” J. Biophotonics 12(8), e201900066 (2019). [CrossRef]
28. K. Maslov, H. Zhang, S. Hu, and L. Wang, “Optical-resolution photoacoustic microscopyfor in vivo imaging of single capillaries,” Opt. Lett. 33(9), 929–931 (2008). [CrossRef]
29. C. Kim, T. Erpelding, L. Jankovic, M. Pashley, and L. Wang, “Deeply penetrating in vivo photoacoustic imaging using a clinical ultrasound array system,” Biomed. Opt. Express 1(1), 278–284 (2010). [CrossRef]
30. J. Yao and L. Wang, “Sensitivity of photoacoustic microscopy,” Photoacoustics 2(2), 87–101 (2014). [CrossRef]
31. J. Shi, T. Wong, Y. He, L. Li, R. Zhang, C. Yung, J. Hwang, K. Maslov, and L. Wang, “High-resolution, high-contrast mid-infrared imaging of fresh biological samples with ultraviolet-localized photoacoustic microscopy,” Nat. Photonics 13(9), 609–615 (2019). [CrossRef]
32. S. Hu, K. Maslov, and L. V. Wang, “Second-generation optical-resolution photoacoustic microscopy with improved sensitivity and speed,” Opt. Lett. 36(7), 1134–1136 (2011). [CrossRef]
33. J. Yao, L. Wang, J. Yang, K. Maslov, T. Wong, L. Li, C. Huang, J. Zou, and L. V. Wang, “High-speed label-free functional photoacoustic microscopy of mouse brain in action,” Nat. Methods 12(5), 407–410 (2015). [CrossRef]
34. L. Lin, J. Yao, L. Li, and L. V. Wang, “In vivo photoacoustic tomography of myoglobin oxygen saturation,” J. Biomed. Opt. 21(6), 061002 (2015). [CrossRef]
35. C. Wood, K. Harutyunyan, D. Sampaio, M. Konopleva, and R. Bouchard, “Photoacoustic-based oxygen saturation assessment of murine femoral bone marrow in a preclinical model of leukemia,” Photoacoustics 14, 31–36 (2019). [CrossRef]
36. P. Beard, “Biomedical photoacoustic imaging,” Interface Focus 1(4), 602–631 (2011). [CrossRef]
37. Y. Sun, F. Ding, Z. Chen, R. Zhang, C. Li, Y. Xu, Y. Zhang, R. Ni, X. Li, G. Yang, Y. Sun, and P. Stang, “Melanin-dot-mediated delivery of metallacycle for NIR-II/photoacoustic dual-modal imaging-guided chemo-photothermal synergistic therapy,” Proc. Natl. Acad. Sci. U. S. A. 116(34), 16729–16735 (2019). [CrossRef]
38. K. Li and B. Liu, “Polymer-encapsulated organic nanoparticles for fluorescence and photoacoustic imaging,” Chem. Soc. Rev. 43(18), 6570–6597 (2014). [CrossRef]
39. Z. Sheng, B. Guo, D. Hu, S. Xu, W. Wu, W. H. Liew, K. Yao, J. Jiang, C. Liu, H. Zheng, and B. Liu, “Bright Aggregation-Induced-Emission Dots for Targeted Synergetic NIR-II Fluorescence and NIR-I Photoacoustic Imaging of Orthotopic Brain Tumors,” Adv. Mater. 30(29), 1800766 (2018). [CrossRef]
40. T. Wong, R. Zhang, P. Hai, C. Zhang, M. Pleitez, R. Aft, D. Novack, and L. Wang, “Fast label-free multilayered histology-like imaging of human breast cancer by photoacoustic microscopy,” Sci. Adv. 3(5), e1602168 (2017). [CrossRef]
41. J. Hui, R. Li, E. Phillips, C. Goergen, M. Sturek, and J. Cheng, “Bond-selective photoacoustic imaging by converting molecular vibration into acoustic waves,” Photoacoustics 4(1), 11–21 (2016). [CrossRef]
42. S. Jacques, “Optical properties of biological tissues: a review,” Phys. Med. Biol. 58(11), R37–R61 (2013). [CrossRef]
43. P. Kumar and M. Pramanik, “Photoacoustic imaging in the second near-infrared window: a review,” J. Biomed. Opt. 24(4), 1–20 (2019). [CrossRef]
44. L. Wang and J. Yao, “A practical guide to photoacoustic tomography in the life sciences,” Nat. Methods 13(8), 627–638 (2016). [CrossRef]
45. L. Wang and L. Gao, “Photoacoustic Microscopy and Computed Tomography: From Bench to Bedside,” Annu. Rev. Biomed. Eng. 16(1), 155–185 (2014). [CrossRef]
46. L. Li, L. Zhu, C. Ma, L. Lin, J. Yao, L. Wang, K. Maslov, R. Zhang, W. Chen, J. Shi, and L. Wang, “Single-impulse panoramic photoacoustic computed tomography of small-animal whole-body dynamics at high spatiotemporal resolution,” Nat. Biomed. Eng. 1(5), 0071 (2017). [CrossRef]
47. C. Zhan, Y. Huang, G. Lin, S. Huang, F. Zeng, and S. Wu, “A Gold Nanocage/Cluster Hybrid Structure for Whole-Body Multispectral Optoacoustic Tomography Imaging, EGFR Inhibitor Delivery, and Photothermal Therapy,” Small 15(33), 1900309 (2019). [CrossRef]
48. X. Wang, Y. Pang, G. Ku, X. Xie, G. Stoica, and L. Wang, “Noninvasive laser-induced photoacoustic tomography for structural and functional in vivo imaging of the brain,” Nat. Biotechnol. 21(7), 803–806 (2003). [CrossRef]
