Open Access
Add to BibSonomy
Abstract
We used high-resolution photoacoustic imaging (PAI) to guide sclerotherapy of vascular malformations in an in vivo animal model. A focus-adjustable PAI system was developed. It can adapt to the imaging needs of different depths by adjusting the focus. Blood samples drawn before and after sclerosis were examined with PAI, which could distinguish whether or not the blood had been exposed to a sclerosing agent. Superficial and deep vessels in the animal model were examined in vivo to prove the feasibility of guiding sclerotherapy. We found that PAI can distinguish sclerotic vessels from normal vessels within a certain depth range. Our findings suggest the potential of PAI to find accurate injection points and to localize thrombi, making it possible to reduce the dosage of sclerosing agents.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Vascular malformations, with a prevalence of approximately 1.5% in the population, are usually noted at birth, and often enlarge over time [1, 2]. Many of lesions are located on head, neck and limbs [3, 4]. Larger lesions may affect not only the appearance of patients but also the function of peripheral organs. Timely and effective treatment is necessary for lesions with these effects. Minimally invasive sclerotherapy is the major treatment for vascular malformations, which in principle destroys the cells and proteins in the abnormal blood vessels such that they form occlusive thrombi that cause fibrosis and regression of the vessels [5–7]. To reduce damage to surrounding normal structures, precise control of the position and amount of sclerosing agent injected is needed [8, 9]. Medical imaging techniques such as digital subtraction angiography (DSA), ultrasound imaging, and magnetic resonance angiography (MRA) are used for guiding sclerotherapy in the clinic [10–12]. The guidance procedure of existing methods is as follows: 1st, the vasculature in the lesion area is identified with medical imaging to find out the area where the vessel density is abnormally increased and the feeding vessel is located approximately. 2nd, a needle is then inserted into the lesion area with image guidance allowing identification of the feeding vessel or the central region of nidus. 3rd, sclerosing agent is slowly injected. 4th, after a region is fully injected with sclerosing agent, it is necessary to find another injection point, preventing excessive sclerosis in one point and too little sclerosis in other points [10–12]. However, each of these methods has its own shortcomings. DSA is the most commonly used clinical guidance method, but the contrast agents and radiation involved pose a hazard to the patient. Ultrasound or doppler ultrasound can display the contour and reflect the blood flow, but the imaging contrast is too poor and the resolution is not high enough to give a clear vascular morphology [13, 14]. What’s more, US is not suitable for small or medium-sized lesions for the existence of blind area. MRA involves injecting a mixture of contrast agents and sclerosing agent to observe the range of the sclerosing agent [15, 16]. The side-effects of contrast agents are also unavoidable. As MRA is based on the contrast between the blood flow and the surrounding static tissue, there is no way to provide a clear image when the blood flow rate is slow [17].
Photoacoustic imaging (PAI) offers opportunities for in vivo detection and monitoring of disease pathophysiology ranging from organelles to organs [18–24]. By making use of the photoacoustic (PA) effect, PAI technique combines the advantages of high resolution and good penetration [25–32]. The PA image is reconstructed by detecting the ultrasound signal generated after a short pulsed laser beam irradiation [33–38]. For physicians, finding the injection point and controlling the dosage are two crucial points in sclerotherapy. Finding the injection point needs to follow two rules: (1) keeping the needle away from normal vessels or tissues; (2) finding the central region of nidus or feeding vessel of lesion to insert the needle. Since clear images of vascular networks may be achieved by PAI based on the natural difference in light absorption between hemoglobin in blood and surrounding tissues, it can be helpful for finding an injection point for sclerotherapy without the use of contrast agents. Hemoglobin is denatured by the injection of sclerosing agent, which leads to a change in light absorption. Areas reached by the sclerosing agent and the positions of the resulting thrombi may be monitored by PAI. So the clinicians can stop the injection timely when observing the formation of thrombi. This helps to eliminate human factors affecting the dose. In this work, normal blood was injected with sclerosing agent, then sclerotic blood was separated into precipitate and supernatant. In order to simulate the different concentration distribution of precipitate in blood, precipitation was mixed at different mass fractions into the supernatant to form different samples. Then these sclerotic blood samples with different mass fractions of precipitation were imaged to compare differences between them and normal blood. In vivo vessels before and after sclerosing agent injection were imaged to verify the capability of PAI for showing the differences after sclerotherapy. The results showed three significant capabilities of PAI for guiding sclerotherapy, including displaying vascular structure, reflecting the areas reached by the sclerosing agent, and showing the thrombi positions. It indicates that the PAI holds tremendous potential and advantages for the guiding sclerotherapy of vascular malformations in clinical application.
2. Methods and materials
2.1 Photoacoustic imaging system
A focus-adjustable PAI system was adapted to visualize vascular structures at different depths. The schematic of the experimental setup is shown in Fig. 1. An excitation laser (DTL-314QT, Laser-export, Moscow, Russia) operates at a wavelength of 532 nm with a pulse width of 8 ns and a repetition rate of 1 KHz. The laser is split into two beams. One laser beam is used to trigger the acquisition system with a photodiode. The other laser beam is collimated by a lens and pinhole, and then coupled into a multimode fiber (0.6-mm-core-diameter). The laser beam from the fiber is focused to the tissue by conical lens and mirrors. To maximize the sensitivity and signal-to-noise ratios, ultrasonic detection and optical excitation are confocal. The height of the probe can be adjusted over a range of approximately 8 mm. The transducer used here is a focused transducer with 20 MHz center frequency with focal length of 13.5 mm. The numerical aperture of the transducer is estimated to be 1.48 by the formula D/F (D is the diameter of transducer and F is the focal length). The lateral resolution is calculated to be 52 µm by the formula 1.02Fc/fD (F is the focal length, c is the sound velocity, f is the center frequency and D is the diameter of transducer). The PA signals from the tissue are detected by the transducer and then amplified by an amplifier (ZFL-500, Minicircuits, New York, USA) before they are collected by a dual-channel data acquisition card (NI5124, National Instruments Corp, Austin TX, USA). The computer then reconstructs the 2-D images from the data recorded with maximum amplitude projection along the z axis. The conical lens, mirrors and ultrasound transducer were integrated into a container that driven by a planar motor to scan. The container was fixed on elevator-platform. Focus could be translated in the axial axis by adjusting elevator-platform. Precision of the elevator-platform was ~5 µm. The step size of system could be adjusted from a few microns to dozens of microns. In the imaging process, the incident light energy density on the tissue surface was limited to 16 mJ/cm2 to conform to the American National Standards Institute safety limit (20 mJ/cm2) [39]. This system takes approximately 10 minutes to image a 9 × 9 mm2 area with a laser repetition rate of 1 KHz.
Fig. 1 Schematic of the photoacoustic imaging system. BS, beam splitter; OL, optical lens; OFC, optical fiber coupler; OF, optical fiber; CL, conical lens; UT, ultrasound transducer; PA signal, photoacoustic signal; DAS, data-acquisition system; PD, photodiode; NDF, neutral density filter.
2.2 Mechanism of detecting blood denaturation
Schematic diagram of the hemoglobin denaturation process is shown in Fig. 2. When the proteins are denatured and the original structure is destroyed, the hydrophobic groups in the molecule are exposed because of the loose structure, resulting in decreased solubility and precipitation, the light absorption of proteins in blood decreases accordingly [40]. Meanwhile, the denatured proteins will accumulate and form precipitates or thrombus. Through the differences in photoacoustic signals, areas reached by the sclerosing agent and the positions of thrombi may be monitored.
Fig. 2 Schematic diagram of hemoglobin denaturation (other cells and proteins in blood are omitted).
2.3 Guidance workflow of photoacoustic imaging
To illustrate the PAI procedure, a schematic diagram of PAI-guided sclerotherapy for vascular malformations is shown in Fig. 3. The first step is to obtain the PAI of the lesion. Then, the feeding vessels supplying oxygen or nutrients and the central region of the nidus where vessels are dense are found. The third and most significant step is to inject the agent. When formation of thrombi is observed, injection in the corresponding point can be stopped. Next, repeating the above process, the other points are injected. Compared with ultrasound, PAI is more suitable for guiding the treatment of small and medium-size lesions, with a higher resolution and contrast. With the existence of the blind area, US cannot accurately image these lesions. Compared with DSA or MRA, PAI has the ability of observing the formation of the thrombus in real time. DSA and MRA can only show the region that has been injected, but whether and where the thrombus has formed cannot be determined. The dosage of agent is based mainly on the experience of clinicians. Furthermore, DSA and MRA need contrast agents.
3. Results
3.1 Lateral resolutions at three depths (1 mm, 3 mm, 5 mm)
The resolutions were measured at three depths to determine the lateral resolution of the system and the change of lateral resolution with depth. A schematic diagram of the focal depth adjustment is shown in Fig. 4(a). Three carbon filaments (~15 µm in diameter) were placed at different depths (1 mm, 3 mm, and 5 mm) in agar. The step size of the system was set as 15 µm. The B scan of three carbon filaments was performed when the focal depth was fixed at three different depths. The results are shown in Figs. 4(b)-4(d). The lateral resolutions at three different depths (1 mm, 3 mm, and 5 mm) were approximately 58 µm, 116 µm, and 175 µm, respectively, when the focus was fixed at 1 mm. When the focus was deepened to 3 mm, the lateral resolutions were 114 µm, 63 µm, and 116 µm, respectively. When the focus was fixed at a depth of 5 mm, the lateral resolutions at the three depths were 128 µm, 116 µm, and 72 µm, respectively.
Fig. 4 (a) Schematic of focal depth adjustment. (b) Lateral resolutions of system at different depths (1 mm, 3 mm, and 5 mm) when focal depth is 1 mm. (c) Lateral resolutions of system at different depths (1 mm, 3 mm, and 5 mm) when focal depth is 3 mm. (d) Lateral resolutions of system at different depths (1 mm, 3 mm, and 5 mm) when focal depth is 5 mm.
3.2 Light absorption spectra and photoacoustic signals of blood before and after sclerosing agent injection
Ethanol was used as the sclerosing agent in the experiments. To verify the spectral differences of blood before and after ethanol injection, a UV-VIS spectrophotometer (UV-2600 and ISR-2600Plus, SHIMADZU, Kyoto, Japan) was used to detect the absorption spectra of normal ex vivo blood and the one mixed with equal amounts of alcohol from 450 to 800 nm (both of them were freeze-dried). As shown in Fig. 5(a), normal blood had a strong absorption peak at 532 nm since the existence of hemoglobin and then the absorption peak decreased after ethanol injection due to the denaturation of proteins (including hemoglobin). However, the denatured proteins tend to accumulate and form precipitates and thrombus, which leads to a higher concentration of proteins at some positions, where the PA signals may be greater than that of normal blood. Due to the inhomogeneous distribution of precipitate in vessels, the PA signals of vessels will show significant inhomogeneity. Therefore, in order to know the signal differences caused by different precipitate concentrations, the PA signal values of the sclerotic blood with different concentrations were measured. As shown in Fig. 5(b), (1) is normal blood, (3) is blood mixed with ethanol, (2) and (4) are the supernatant and precipitate of the blood-ethanol mixture, respectively. Precipitation was mixed at different mass fractions into the supernatant and PA images of these samples with different mass fractions of precipitate were acquired. The step size was set as 40 µm. When the mass fraction of precipitate reaches a certain value, the PA signal will be greater than that of normal blood. For a more accurate understanding of their differences, the PA signals of blood with different mass fractions of precipitate in Fig. 5(b) were calculated in Fig. 5(c) and the contrast is sharp. Therefore, the locations where precipitate accumulates, and thrombi may result, may be identified by PAI. What needs to be noted here is the denaturation decreases the intensity of PA while becoming thrombi enhances it. They conflicts each other in the sense. But precisely because of the opposite effect, we can be sure precipitation (or thrombus) has formed when the PA intensity increases in an injected point. Otherwise, it cannot be sure whether the signal enhancement is caused by thrombi or denaturation. Hence, PAI based on the change of the blood absorption spectra resulting from the combination of blood and sclerosing agent provides the possibility for localizing thrombi.
Fig. 5 (a) Absorption spectrum of normal blood and sclerotic blood (both of them were freeze- dried). (b) Photographs and photoacoustic images of blood samples with different mass fractions of precipitation. (1) normal blood; (2) supernatant of sclerotic blood; (3) sclerotic blood (mixed normal blood with the same amount of ethanol); (4) precipitate of sclerotic blood (freeze-dried). 1% means the mass ratio of (4) to (2) is 1% and so on. (c) Comparison of photoacoustic signal intensities between normal blood and sclerotic blood with different mass fractions of precipitation in (b).
3.3 In vivo imaging of superficial vessels before and after ethanol injection in rabbit
To further demonstrate the capability of PAI guiding the sclerotherapy of vascular malformations, comparison of the rabbit ear before and after vein sclerosis was performed. The experimental rabbit was a New Zealand white rabbit (body weight: 2 kg), whose ear hair was removed with a hair remover (Payven Depilatory China) before experiment. Meanwhile, during the experiment, sodium pentobarbital (30 mg/kg) was administered to keep the rabbit motionless. All experimental animal procedures were in accordance with the guidelines of the South China Normal University. In animal experiments, the step size was also set as 40 µm. In vivo PA images of rabbit ear veins before and after ethanol injection are shown in Figs. 6(a) and 6(b). Vessels 1 and 2 were two groups of vessels. As shown in Fig. 6(a), the morphology of vessels 1 and 2 are clearly displayed by PAI before ethanol injection, which would made it easy for a clinician to find target vessels or some key areas. These target vessels or regions are the core of the lesion and their blockages are the key to treatment. V1, V2, and V3 are three different vessels contained in vessels 1. Then an injection point in V1 was chosen. The PAI result after ethanol injection is shown in Fig. 6(b). Vessels 2 are still clear and the brightness of black tape used for reference is similar to that in Fig. 6(a). However, bright foci appear in the area of vessels 1 and the brightness of the area is heterogeneous, which is in accord with the results in Fig. 5. Therefore, the positions of thrombi in vivo could be recognized according to the positions of bright points in PAI. In Fig. 6(c), the average PA signal values of black reference tape, normal blood vessels 1 in Fig. 6(a) and thrombi in Fig. 6(b) were measured. The average PA signal value of thrombi was two times higher than that of normal vessels 1, which verifies the results in Fig. 5 and the ability of PAI to indicate positions of thrombi. In order to know more precisely the signal heterogeneity of vessels injected with ethanol, the average PA signal values of vessels 1 and their standard deviations were counted. The result is shown in Fig. 6(d). The standard deviations of V1 and V3 showed great differences between pre-injection and post-injection. The standard deviation ratio of V1 between pre-injection and post-injection was 3.53 and that of V3 was up to 5.56. Because V2 was far from the injection point, the standard deviations did not change. Therefore, the vessels that were injected with the agent could also be identified by the inhomogeneity of PA signal.
Fig. 6 In vivo photoacoustic imaging of superficial vessels in rabbit ear before ethanol injection (a) and after injection (b). (c) Average photoacoustic signal intensities of black tape, vessels 1 and thrombi in (a) and (b). (d) Average photoacoustic signal intensities of vessels 1 in (a) and (b). The error bars represent standard deviations of photoacoustic signal values.
3.4 In vivo imaging of deep vessels before and after ethanol injection in rabbit
In order to adapt to clinical demand, we also examined the feasibility of PAI guiding the sclerotherapy of deep vessels. We used a second New Zealand white rabbit (body weight: 3 kg). In order to simulate the invisibility of deep vessels, the PAI was performed on the reverse side of its ear. As shown in Figs. 7(a) and 7(b), a clear vascular network could not been seen on the reverse side of the ear. Meanwhile, to verify the morphology of blood vessels, the corresponding area on the obverse side is also shown. To assess the imaging ability at different depths, the focus was set on the skin surface first to image the reference black tape and then was deepened to get a clear view of the vascular network. As shown in Fig. 7, when the focus was set on the surface, the image of the black tape was clear. The PA intensity of the black tape is higher than that of vessels, and only the outline of large blood vessels can be seen. Then, when the focus was deepened (~1.5 mm below the surface), more vascular details are visible, providing a more accurate injection position. After injection, the vessel outlines became blurry and vessel 1 appeared to be punctured, where a thrombus was formed as expected. In Figs. 7(c) and 7(d), the average PA signal intensities of black tape, vessel 1 and thrombus were measured. The PA signals of vessel 1 and thrombus were greatly enhanced due to the deepening of the focus, which reduced the signal of the black tape. This demonstrates the importance of focal regulation for system imaging. All the results were in agreement with that of in vitro experiments, demonstrating the feasibility of PAI guiding sclerotherapy of deep vessels.
Fig. 7 In vivo photoacoustic imaging of deep vessels in rabbit ear before ethanol injection (a) and after injection (b). (c) Average photoacoustic signals of black tape and vessel 1 before and after focus deepening in (a). (d) Average photoacoustic signals of black tape and thrombus before and after focus deepening in (b).
4. Discussion
PAI may guide the percutaneous sclerotherapy of vascular malformations. Current sclerotherapy often results in complications such as skin ulcers, muscle atrophy, and nerve injury [41]. These complications may cause considerable pain and inconvenience to patients, especially for infants and young children. An accurate location and strict control of dosage are significant in reducing complications. Though US is a noninvasive method, its contrast and resolution are limited and it is not suitable for small and medium-size lesions. DSA and MRA usually need contrast agents, which undoubtedly aggravate the adverse reactions, and DSA or MRA cannot show the formation of thrombi, which is not beneficial in controlling the dosage of the sclerosing agent. PAI can show clear vascular morphology and the formation of thrombi without contrast agents and radiation, helping clinicians to localize the injection point and evaluate the therapeutic effects. The depths of lesions in clinical cases vary from mm to cm. The ranges of lesions are usually from a few cm to a dozen cm [42, 43]. For infants and young children, many lesions are in the early stage, and the depths and scopes are usually in mm and cm, respectively. PAI is especially suitable to help them receive timely treatment. The characteristics of PAI and three other methods are summarized in Table 1. Compared with the current guidance procedure, PAI-guided operational steps are similar. A key operational point is to ensure that the focus depths are the same before and after the injection. To better achieve this objective, auxiliary instruments should be considered to help physicians during surgery. Furthermore, PAI can be used multiple times with low cost and good operability, and is especially appropriate for cases requiring repeated treatments [43, 44]. Meanwhile, PAI may be combined with other endoscopic minimally invasive surgical instruments to guide the sclerotherapy of vascular malformations in internal organs such as bowel and liver. In clinic, sclerotherapy guided by optical endoscope system for intestinal vascular malformations has been applied. And PA endoscope system has also been developed to detect atherosclerosis and esophageal diseases in recent years [45, 46]. The size of transducer used in PA endoscope system has reached millimeter [46]. So a combination of PA endoscope and minimally invasive instruments is promising. Even though three capabilities of PAI to guide sclerotherapy including displaying vascular structure, reflecting the areas reached by sclerosing agent and showing the thrombi positions have been demonstrated, several problems should be solved before clinical application. First, the current scanning speed cannot meet the real-time requirements of clinical treatment. A fast scanning device and imaging algorithm are needed for real-time imaging. Linear or array transducer can be considered and special integrated acquisition circuit should be developed. In addition, imaging algorithm used here is 2D maximum amplitude projection, but a 3D PAI showing depth information will be better to clinical application. Also, this requires a faster imaging speed. Second, the imaging system now uses coupling fluid to couple PA signals from tissue into the transducer, requiring contact with the skin. Just as the ultrasound-guided process, this requires the needle to be inserted obliquely into the skin from the edge of lesion, reducing the convenience of guidance. Our group has developed non-contact PAI which has been applied to detect basal cell carcinoma and melanoma in mice with good imaging quality [47, 48]. Therefore, it can be an alternative to solve this problem in the future. Third, we focused on the treatment of superficial or middle-level vascular malformations that were not very deep, thus the wavelength of laser used here was 532 nm. For deeper vascular malformations, the near-infrared spectral range could also be used for this method based on the changes in the absorption spectrum of hemoglobin. Fourth, the dosage of sclerosing agents is not controlled precisely at present. In order to master the most suitable dosage, differences in PA images before and after sclerotherapy, dosage of sclerosing agents and treatment effect need to be counted and analyzed, which requires extensive research. Further study is necessary to take full advantage of PAI.
Table 1. Comparison of guiding methods for sclerotherapy of vascular malformations
5. Conclusions
In summary, we developed a new method based on PA technique for guiding percutaneous sclerotherapy of vascular malformations, achieving a treatment with higher accuracy and fewer side-effects. This method may be especially appropriate for small and medium-sized lesions and cases requiring repeated treatments. PAI may allow early and accurate treatment of vascular malformations, and potentially opens up a new field of application.
Funding
National Natural Science Foundation of China (11774101, 61331001, 61627827, 81630046); National High Technology Research and Development Program of China (2015AA020901); The Science and Technology Planning Project of Guangdong Province, China (2015B020233016, 2014B020215003, 2014A020215031); The Distinguished Young Teacher Project in Higher Education of Guangdong, China (YQ2015049); The Science and Technology Youth Talent for Special Program of Guangdong, China (2015TQ01X882).
Acknowledgments
The authors would like to thank Dr. Yujiao Shi, Dr. Fen Yang and Dr. Haigang Ma for their valuable advices to this paper. We thank Libby Cone, MD, MA, from Liwen Bianji, Edanz Group China (www.liwenbianji.cn/ac) for editing a draft and revision of this manuscript.
References and links
1. P. Redondo, “Vascular malformations (I). Concept, classification, pathogenesis and clinical features,” Actas Dermosifiliogr. 98(3), 141–158 (2007). [CrossRef] [PubMed]
2. J. B. Mulliken and J. Glowacki, “Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics,” Plast. Reconstr. Surg. 69(3), 412–422 (1982). [CrossRef] [PubMed]
3. X. Wang, X. Bai, K. Wu, and Q. Liu, “The evaluation of the management of haemangioma and vascular malformation in oral and maxillofacial region,” Int. J. Oral Maxillofac. Surg. 38(5), 529 (2009). [CrossRef]
4. S. Steinbach, A. J. Fasunla, C. M. E. Lahme, S. P. Schäfers, W. Hundt, P. Wolf, R. Mandic, J. A. Werner, and B. Eivazi, “Smell and taste in patients with vascular malformation of the extracranial head and neck region,” Am. J. Rhinol. Allergy 28(1), 45–51 (2014). [CrossRef] [PubMed]
5. I. T. Jackson, R. Carreño, Z. Potparic, and K. Hussain, “Hemangiomas, vascular malformations, and lymphovenous malformations: classification and methods of treatment,” Plast. Reconstr. Surg. 91(7), 1216–1230 (1993). [CrossRef] [PubMed]
6. L. Blum, S. Gallas, J. P. Cottier, C. B. Sonier Vinikoff, G. Lorette, and D. Herbreteau, “Percutaneous sclerotherapy for the treatment of soft-tissue venous malformations: a retrospective study of 68 patients,” J. Radiol. 85(2 Pt 1), 107–116 (2004). [CrossRef] [PubMed]
7. H. Furukawa, S. Sasaki, A. Oyama, E. Funayama, T. Hayashi, N. Saito, M. Nagao, W. Mol, and Y. Yamamoto, “Efficacy of percutaneous ethanol sclerotherapy for venous malformation in lower extremities: a retrospective review of 21 cases,” Eur. J. Plast. Surg. 36(2), 105–110 (2013). [CrossRef]
8. L. Su, X. Fan, L. Zheng, and J. Zheng, “Absolute ethanol sclerotherapy for venous malformations in the face and neck,” J. Oral Maxillofac. Surg. 68(7), 1622–1627 (2010). [CrossRef] [PubMed]
9. Y. H. Jeon, Y. S. Do, S. W. Shin, W. C. Liu, J. M. Cho, M. H. Lee, D. L. Kim, B. B. Lee, S. W. Choo, and I. W. Choo, “Ethanol embolization of arteriovenous malformations: results and complications of 33 cases,” Radiol. 235(2), 674–682 (2003).
10. J. Li, J. Chen, G. Zheng, G. Liao, Z. Fu, J. Li, T. Zhang, and Y. Su, “Digital subtraction angiography-guided percutaneous sclerotherapy of venous malformations with pingyangmycin and/or absolute ethanol in the maxillofacial region,” J. Oral Maxillofac. Surg. 68(9), 2258–2266 (2010). [CrossRef] [PubMed]
11. S. Blaise, M. Charavin-Cocuzza, H. Riom, M. Brix, C. Seinturier, J. M. Diamand, G. Gachet, and P. H. Carpentier, “Treatment of low-flow vascular malformations by ultrasound-guided sclerotherapy with polidocanol foam: 24 cases and literature review,” Eur. J. Vasc. Endovasc. Surg. 41(3), 412–417 (2011). [CrossRef] [PubMed]
12. D. T. Boll, E. M. Merkle, and J. S. Lewin, “Low-flow vascular malformations: MR-guided percutaneous sclerotherapy in qualitative and quantitative assessment of therapy and outcome,” Radiology 233(2), 376–384 (2004). [CrossRef] [PubMed]
13. G. Gianfranco, F. Eloisa, C. Vito, G. Raffaele, T. Gianluca, and R. Umberto, “Color-Doppler Ultrasound in the Diagnosis of Oral Vascular Anomalies,” N. Am. J. Med. Sci. 6(1), 1–5 (2014). [CrossRef] [PubMed]
14. L. Gold, L. N. Nazarian, A. S. Johar, and V. M. Rao, “Characterization of maxillofacial soft tissue vascular anomalies by ultrasound and color Doppler imaging: an adjuvant to computed tomography and magnetic resonance imaging,” J. Oral Maxillofac. Surg. 61(1), 19–31 (2003). [CrossRef] [PubMed]
15. D. M. O’Mara, P. A. Dicamillo, W. D. Gilson, D. A. Herzka, F. K. Wacker, J. S. Lewin, and C. R. Weiss, “MR-guided percutaneous sclerotherapy of low-flow vascular malformations: Clinical experience using a 1.5 tesla MR system,” J. Magn. Reson. Imaging 45(4), 1154–1162 (2017). [CrossRef] [PubMed]
16. D. T. Boll, E. M. Merkle, and J. S. Lewin, “MR-guided percutaneous sclerotherapy of low-flow vascular malformations in the head and neck,” Magn. Reson. Imaging Clin. N. Am. 13(3), 595–600 (2005). [CrossRef] [PubMed]
17. M. P. Hartung, T. M. Grist, and C. J. François, “Magnetic resonance angiography: current status and future directions,” J. Cardiovasc. Magn. Reson. 13(1), 19 (2011). [CrossRef] [PubMed]
18. W. Song, W. Zheng, R. Liu, R. Lin, H. Huang, X. Gong, S. Yang, R. Zhang, and L. Song, “Reflection-mode in vivo photoacoustic microscopy with subwavelength lateral resolution,” Biomed. Opt. Express 5(12), 4235–4241 (2014). [CrossRef] [PubMed]
19. L. Song, K. Maslov, R. Bitton, K. K. Shung, and L. V. Wang, “Fast 3-D dark-field reflection-mode photoacoustic microscopy in vivo with a 30-MHz ultrasound linear array,” J. Biomed. Opt. 13(5), 054028 (2008). [CrossRef] [PubMed]
20. G. He, D. Xu, H. Qin, S. Yang, and D. Xing, “In vivo cell characteristic extraction and identification by photoacoustic flow cytography,” Biomed. Opt. Express 6(10), 3748–3756 (2015). [CrossRef] [PubMed]
21. Y. Wang, D. Xu, S. Yang, and D. Xing, “Toward in vivo biopsy of melanoma based on photoacoustic and ultrasound dual imaging with an integrated detector,” Biomed. Opt. Express 7(2), 279–286 (2016). [CrossRef] [PubMed]
22. L. V. Wang and S. Hu, “Photoacoustic tomography: in vivo imaging from organelles to organs,” Science 335(6075), 1458–1462 (2012). [CrossRef] [PubMed]
23. Z. Chen, S. Yang, and D. Xing, “In vivo detection of hemoglobin oxygen saturation and carboxyhemoglobin saturation with multiwavelength photoacoustic microscopy,” Opt. Lett. 37(16), 3414–3416 (2012). [CrossRef] [PubMed]
24. X. Wang, X. Xie, G. Ku, L. V. Wang, and G. Stoica, “Noninvasive imaging of hemoglobin concentration and oxygenation in the rat brain using high-resolution photoacoustic tomography,” J. Biomed. Opt. 11(2), 024015 (2006). [CrossRef] [PubMed]
25. K. Peng, L. He, B. Wang, and J. Xiao, “Detection of cervical cancer based on photoacoustic imaging-the in-vitro results,” Biomed. Opt. Express 6(1), 135–143 (2014). [CrossRef] [PubMed]
26. L. Xi, S. R. Grobmyer, L. Wu, R. Chen, G. Zhou, L. G. Gutwein, J. Sun, W. Liao, Q. Zhou, H. Xie, and H. Jiang, “Evaluation of breast tumor margins in vivo with intraoperative photoacoustic imaging,” Opt. Express 20(8), 8726–8731 (2012). [CrossRef] [PubMed]
27. T. Wang, S. Nandy, H. S. Salehi, P. D. Kumavor, and Q. Zhu, “A low-cost photoacoustic microscopy system with a laser diode excitation,” Biomed. Opt. Express 5(9), 3053–3058 (2014). [CrossRef] [PubMed]
28. S. L. Chen, L. J. Guo, and X. Wang, “All-optical photoacoustic microscopy,” Photoacoustics 3(4), 143–150 (2015). [CrossRef]
29. Z. Xie, S. L. Chen, T. Ling, L. J. Guo, P. L. Carson, and X. Wang, “Pure optical photoacoustic microscopy,” Opt. Express 19(10), 9027–9034 (2011). [CrossRef] [PubMed]
30. W. Song, Q. Wei, T. Liu, D. Kuai, J. M. Burke, S. Jiao, and H. F. Zhang, “Integrating photoacoustic ophthalmoscopy with scanning laser ophthalmoscopy, optical coherence tomography, and fluorescein angiography for a multimodal retinal imaging platform,” J. Biomed. Opt. 17(6), 061206 (2012). [CrossRef] [PubMed]
31. X. Yang, B. Jiang, X. Song, J. Wei, and Q. Luo, “Fast axial-scanning photoacoustic microscopy using tunable acoustic gradient lens,” Opt. Express 25(7), 7349–7357 (2017). [CrossRef] [PubMed]
32. L. D. Liao, M. L. Li, H. Y. Lai, Y. Y. Shih, Y. C. Lo, S. Tsang, P. C. P. Chao, C. T. Lin, F. S. Jaw, and Y. Y. Chen, “Imaging brain hemodynamic changes during rat forepaw electrical stimulation using functional photoacoustic microscopy,” Neuroimage 52(2), 562–570 (2010). [CrossRef] [PubMed]
33. P. C. Li, C. R. C. Wang, D. B. Shieh, C. W. Wei, C. K. Liao, C. Poe, S. Jhan, A. A. Ding, and Y. N. Wu, “In vivo photoacoustic molecular imaging with simultaneous multiple selective targeting using antibody-conjugated gold nanorods,” Opt. Express 16(23), 18605–18615 (2008). [CrossRef] [PubMed]
34. A. Taruttis and V. Ntziachristos, “Advances in real-time multispectral optoacoustic imaging and its applications,” Nat. Photonics 9(4), 219–227 (2015). [CrossRef]
35. M. Jeon, W. Song, E. Huynh, J. Kim, J. Kim, B. L. Helfield, B. Y. C. Leung, D. E. Goertz, G. Zheng, J. Oh, J. F. Lovell, and C. Kim, “Methylene blue microbubbles as a model dual-modality contrast agent for ultrasound and activatable photoacoustic imaging,” J. Biomed. Opt. 19(1), 16005 (2014). [CrossRef] [PubMed]
36. X. Feng, F. Gao, and Y. Zheng, “Thermally modulated photoacoustic imaging with super-paramagnetic iron oxide nanoparticles,” Opt. Lett. 39(12), 3414–3417 (2014). [CrossRef] [PubMed]
37. S. Iskander-Rizk, P. Kruizinga, G. Springeling, H. J. Vos, A. F. van der Steen, and G. van Soest, “Photoacoustic imaging of sub-diffraction objects with spectral contrast,” Opt. Lett. 42(2), 191–194 (2017). [CrossRef] [PubMed]
38. Z. Deng, W. Li, and C. Li, “Slip-ring-based multi-transducer photoacoustic tomography system,” Opt. Lett. 41(12), 2859–2862 (2016). [CrossRef] [PubMed]
39. “American national standard for the safe use of lasers,” ANSI Standard Z136.1 (2007).
40. “Denaturation (biochemistry),” https://en.wikipedia.org/wiki/Denaturation_(biochemistry).
41. S. O. Odeyinde, L. Kangesu, and M. Badran, “Sclerotherapy for vascular malformations: complications and a review of techniques to avoid them,” J. Plast. Reconstr. Aesthet. Surg. 66(2), 215–223 (2013). [CrossRef] [PubMed]
42. P. Redondo, “Classification of vascular anomalies (tumours and malformations). Clinical characteristics and natural history,” An. Sist. Sanit. Navar. 27(1Suppl 1), 9–25 (2004). [PubMed]
43. J. R. Costa, M. A. Torriani, E. S. Hosni, O. P. D’Avila, and P. J. de Figueiredo, “Sclerotherapy for vascular malformations in the oral and maxillofacial region: treatment and follow-up of 66 lesions,” J. Oral Maxillofac. Surg. 69(6), e88–e92 (2011). [CrossRef] [PubMed]
44. M. Mishra, G. Singh, A. Gaur, S. Tandon, and A. Singh, “Role of sclerotherapy in management of vascular malformation in the maxillofacial region: Our experience,” Natl. J. Maxillofac. Surg. 8(1), 64–69 (2017). [CrossRef] [PubMed]
45. J. Zhang, S. Yang, X. Ji, Q. Zhou, and D. Xing, “Characterization of lipid-rich aortic plaques by intravascular photoacoustic tomography: ex vivo and in vivo validation in a rabbit atherosclerosis model with histologic correlation,” J. Am. Coll. Cardiol. 64(4), 385–390 (2014). [CrossRef] [PubMed]
46. D. Jin, F. Yang, Z. Chen, S. Yang, and D. Xing, “Biomechanical and morphological multi-parameter photoacoustic endoscope for identification of early esophageal disease,” Appl. Phys. Lett. 111(10), 103703 (2017). [CrossRef]
47. Z. Chen, S. Yang, and D. Xing, “Optically integrated trimodality imaging system: combined all-optical photoacoustic microscopy, optical coherence tomography, and fluorescence imaging,” Opt. Lett. 41(7), 1636–1639 (2016). [CrossRef] [PubMed]
48. W. Zhou, Z. Chen, S. Yang, and D. Xing, “Optical biopsy approach to basal cell carcinoma and melanoma based on all-optically integrated photoacoustic and optical coherence tomography,” Opt. Lett. 42(11), 2145–2148 (2017). [CrossRef] [PubMed]
