Open Access
Add to BibSonomy
Abstract
In photoacoustic tomography (PAT), delivering high energy pulses through optical fiber is critical for achieving high quality imaging. A fiber coupling scheme with a beam homogenizer is demonstrated for coupling high energy pulses in a single multimode fiber. This scheme can benefit PAT applications that require miniaturized illumination or internal illumination with a small fiber. The beam homogenizer is achieved by using a cross cylindrical lens array, which provides a periodic spatial modulation on the phase of the input light. Thus the lens array acts as a phase grating which diffracts the beam into a 2D diffraction pattern. Both theoretical analysis and experiments demonstrate that the focused beam can be split into a 2D spot array that can reduce the peak power on the fiber tip surface and thus enhance the coupling performance. The theoretical analysis of the intensity distribution of the focused beam is carried out by Fourier optics. In experiments, coupled energy at 48 mJ/pulse and 60 mJ/pulse have been achieved and the corresponding coupling efficiency is 70% and 90% in a 1000-μm and a 1500-μm-core-diameter fiber, respectively. The high energy pulses delivered by the multimode fiber are further tested for PAT imaging in phantoms. PAT imaging of a printed dot array shows a large illumination area of 7 cm2 under 5 mm thick chicken breast tissue. In vivo imaging is also demonstrated on the human forearm. The large improvement in coupling energy can potentially benefit PAT with single fiber delivery to achieve large area imaging and deep penetration detection.
© 2017 Optical Society of America
1. Introduction
Photoacoustic tomography (PAT) is a photoacoustic (PA) imaging modality characterized by large area illumination and multi-directional detection [1,2]. Due to its relative deep penetration capability, PAT has been applied to study skin cancer [3], prostate cancer [4,5], breast cancer [6], abdominal organs [7], and brain activity [8]. PAT uses high energy laser pulses to excite PA signals. These pulses have been delivered traditionally in free space by prisms [3, 7]. However, to translate PAT into clinical applications, delivering the high-energy pulses through an optical fiber is necessary for operational convenience and laser safety. Several prototype PAT systems using optical fiber for light delivery have been developed for imaging cancers [9–13]. Manohar et al. developed a breast imaging system employing a 588-element, 1-MHz ultrasound array with a 1064 nm pulse laser for excitation [9–11]. To illuminate the entire breast, a 1-to-9 splitting fiber bundle was applied to deliver nearly 81 mJ of pulse energy to the breast surface. A total 30 cases of breast malignancies from 31 patients were visualized with high contrast, showing PAT as a potential alternative to X-ray mammography for breast cancer imaging. Horiguchi et al. developed a PAT system to monitor periprostatic tissue in prostate cancer patients undergoing radical prostatectomy [13]. Their light delivery was through two illumination apertures which were integrated with a 6.5 MHz transrectal ultrasound transducer, delivering 60 mJ per pulse. By applying a 756 nm Q-switched pulse laser for illumination, periprostatic microvessels were imaged and located, improving the identification of the neurovascular bundle. Seven patients were imaged in this study. Furthermore, the small fiber diameter enables PAT to perform internal illumination to image internal organs, which can overcome the problem of insufficient light penetration compared with external illumination [8, 13, 14]. For example, Lin et al. developed an illumination for oral cavity to provide sufficient light to image a mouse brain [8]. Using a multimode fiber with 2.8 mm core diameter, nearly 10 mJ energy at 780 nm was delivered, where the internal illumination enabled identification of internal brain structures.
A major challenge in delivering high-energy laser pulses is the fiber damage of the tip surface caused by the high peak power density when focusing and coupling light into the fiber [15]. Thus the output energy from optical fiber delivery has been limited, which restricts the illumination area and penetration of PAT. Several measurements at different wavelengths and pulse widths have reported the damage threshold for fused silica fibers [16–19]. Gallais et al. observed a damage threshold at a peak power density of 9.3 GW/cm2 for silica glass, obtained at 1064 nm wavelength and 7 ns pulse duration after 1000 continuous pulses [18]. Robinson and Ilev observed a damage threshold at a peak power density of 3.7 GW/cm2 in a 100-μm-core-diameter fiber and 3.9 GW/cm2 in a 200-μm-core-diameter fiber, where the corresponding pulse energy was 2.9 mJ and 9.7 mJ respectively [19]. However, in the same work, the damage threshold was only 0.86 GW/cm2, corresponding to 13 mJ in pulse energy, in a 700-μm-core-diameter fiber. Since there is not a definite value of damage threshold for different fibers, commercial fibers usually have their own practical and theoretical damage threshold as a guideline for usage. For example, Thorlabs Inc. specifies a practical damage threshold of 1 GW/cm2 and theoretical damage threshold of 5 GW/cm2.
Such high peak power occurs when focusing and coupling light into the fiber. A traditional coupling method uses a single lens to focus the collimated beam and couple the focused light into a multimode fiber with a large core diameter. However, for multimode fiber with relatively large core diameter (e.g. 1 mm), it is difficult to find a lens to generate a focal spot in the similar dimension. Therefore, the coupling of pulse laser usually places the fiber behind the focus to obtain an enlarged beam which can match with the core size of the fiber [20,21]. A fiber bundle can couple more energy than a single fiber due to its larger effective core diameter. Table 1 lists several systems that use a multimode fiber or a fiber bundle for pulse laser delivery in PAT. Variations in the coupled pulse energy depend on the different systems and applications. In a multimode fiber, the reported energy is typically limited to be less than 28 mJ. In a fiber bundle, energy up to 81 mJ has been reported [10, 21–23]. Although fiber bundle can couple higher power than multimode fiber, fiber bundle usually has a much larger size. Meanwhile, multimode fiber has relatively smaller size and it can benefit many applications that require miniaturized illumination or internal illumination. In PA system that uses a compact handheld probe to deliver light and collect signal [24], it is possible to reduce the probe dimension by using multimode fiber instead of fiber bundle. In applications which need internal illumination, such as transurethral illumination for prostate imaging [5] or PA endoscopy [7, 25], multimode fiber is necessary because small size is critical. However, the energy that can be coupled into a multimode fiber is still relative low. The ANSI laser safety standard has specified the maximum fluence rate for exposure on tissues like skin, e.g. 20 mJ/cm2 at 700 nm wavelength [26]. The signal to noise ratio in PAT is usually limited and higher fluence rate can generate higher PAT signal. Thus coupling high energy into multimode fiber is necessary for systems that require miniaturized or internal illumination in order to illuminate a larger area or achieve deeper penetration. Therefore, it is necessary to develop a new coupling scheme to improve the coupled pulse energy in multimode fiber without causing fiber damage.
Table 1. PAT systems with fiber delivery
Beam homogenization has been used in the area of laser shock process to avoid fiber damage when coupling high energy laser pulses [27,28]. With a beam homogenizer, Schmidt-Uhlig et al. reported delivering nearly 100 mJ/pulse energy in a 1500-μm-core-diameter fused silica fiber from a 5 ns Nd:YAG laser at 10 Hz repetition rate. Nevertheless, the authors did not provide a theoretical analysis on the beam homogenization, and this approach has not been applied to PAT imaging.
In this paper, the coupling scheme based on beam homogenization is applied to a PAT system with multimode fiber delivery. The goal is to improve upon 28 mJ/pulse in the multimode fiber, in order to develop the PAT system for imaging prostate through transurethral or transrectal illumination. A cross cylindrical lens array and a plano-convex spherical lens are used to homogenize and couple high-energy laser pulses into a multimode fiber. Compared with previous work [27,28], a detailed theoretical analysis is carried out here on the focusing effect of the lens array based on the theory of Fourier optics. Optical simulations and experiments are also carried out to study the effect of the lens array. The theoretical result, optical simulation, and experimental results match with each other and show that the laser energy is split into multiple focal spots on the fiber tip surface, lowering the risk of fiber damage. In the fiber coupling experiment with the lens array, a 1000-μm-core-diameter fiber can output 48 mJ/pulse energy and a 1500-μm fiber can output 60 mJ/pulse energy, which are much higher than other reported PAT systems with light delivery by a single fiber. The high-energy pulses delivered by the multimode fiber are further applied to PAT imaging on phantom and in vivo imaging. The improvement in fiber output energy enables PAT illumination and imaging of a larger tissue area.
2. Beam homogenization with a cross cylindrical lens array
2.1 Principle
Figure 1 shows the schematic of the beam homogenization and fiber coupling system consisting of a cross cylindrical lens array and a plano-convex lens. The cross cylindrical lens array contains two 1D arrays of cylindrical lenses, an x array and a y array, joined together back-to-back, resulting in a 2D lens array effectively. This beam homogenization and fiber coupling system can split the incident laser beam to generate a more distributed intensity profile on the focal plane, and thus reduce the peak energy density on the fiber tip.
Fig. 1 Schematic of the coupling optics. The incident beam from the pulse laser is split into beamlets by a cross cylindrical lens array, and the beamlets are focused by a plano-convex lens at its focal plane. To couple the light into a fiber, the fiber input end is placed at the focal plane of the plano-convex lens.
The analysis of the intensity distribution on the focal plane is carried out based on the theory of Fourier optics. The optical model is shown in Fig. 2. In Fig. 2, defines the scalar electric field right after the lens array and presents the scalar electric field on the focal plane of the coupling lens. The distance between the lens array and the coupling lens is . The detection plane is the back focal plane of the coupling lens and is the focal length of the coupling lens. A plane wave with constant amplitude is assumed as the input in front of the lens array and can be expressed as,
where is the amplitude of the incident plane wave and denotes the transmittance function of the lens array. The transmittance function for a single cylindrical lens focusing in the x direction can be expressed as [29],where is the wave number of the incident plane wave and is the focal length of the cylindrical lens. By neglecting the thickness of the cylindrical lens array, its transmittance function can be expressed as,where is the number of elements of the lens array in both x and y, and denote the central position of the x and y cylindrical lens array, respectively, and is the pitch size of each cylindrical lens, which equals to the width of each lens element. The rectangular function is defined as,Fig. 2 (a) Optical model of the coupling scheme with the lens array. The lens array transforms the incident plane wave into multiple spherical wavelets. Constructive interference happens where the propagation angle satisfies , which forms the mth diffraction order on the focal plane. (b) 1D plot of the periodic phase modulation function introduced by the lens array. P is the pitch of the lens array.
The rectangular function in Eq. (4) defines the aperture boundary for each cylindrical lens. The first summation term in Eq. (3) represents one side of the lens array, which only focuses light in the x direction. The exponential term indicates the phase change caused by each cylindrical lens. The second summation term presents the other side of the lens array, which only focuses light in the y direction. Because the focusing effect along the two directions is independent, the two summation terms can be multiplied directly to obtain the transmittance function of the lens array in 2D.
The electric field at the focal plane can be obtained from the electric field using Fourier optics [30],
where is the Fourier transform operator. The field at the focal plane is the Fourier transform of the field , multiplied with a phase term related to the distances d and . The detailed derivation of Eq. (5) can be found in [30]. The intensity distribution can be obtained as the modulus squared of the field amplitude,As shown in Fig. 2, the lens array acts similarly as a phase grating [30,31]. Each element of the lens array introduces a phase delay according to its transmittance function,
The transmittance function can be viewed as a quadratic approximation to a spherical wave [30]. For an input of plane wave, it is transformed into an array of secondary spherical wavelets after the lens array. The lens array introduces a periodic spatial phase modulation that can be written as,where the comb function is a sequence of delta impulses uniformly distributed in space and defined as [30],A modulation depth is defined as below to describe the strength of the phase modulation,When the input light is coherent, the multiple spherical wavelets lead to multi-beam interference. At certain propagation angles , constructive interference happens, where the path length difference between two adjacent beams is an integer number of wavelength. The constructive interference condition is similar as the diffraction orders of phase grating. The coupling lens focuses the beams at the same angle to one spot on its focal plane, forming the mth diffraction order. The position of the mth order diffraction spot can be written as:For the effective 2D lens array, the light intensity distribution on the focal plane of the coupling lens forms a 2D diffraction pattern. The period of the diffraction pattern isA phase grating can be much more efficient in diffracting light than an amplitude grating. The diffraction efficiency for each order depends on the modulation depth and the shape of the periodic phase modulation function. For small modulation depth, the zero-order beam contains most of the power. As the modulation depth increases, more and more power can be diffracted into the higher order beams.2.2 Theoretical and experimental validation of the focal intensity distribution
Both theoretical analysis and experimental measurement have been carried out to study the intensity distribution on the focal plane. The cross cylindrical lens array is obtained from SUSS MicroOptics (Nr.18-00142, Hauterive, Switzerland). Its specifications are listed in Table 2. The focal length of the cylindrical lens array is calculated to be 133 mm. The coupling lens is a plano-convex spherical lens with 75 mm focal length (LA1608, Thorlabs, Newton, United States). The wavelength is set to be 700 nm.
Table 2. Specifications of the cross cylindrical lens array
The theoretical analysis of the intensity distribution on the focal plane can be calculated based on Eqs. (5) and (6). The transmittance function of the cross cylindrical lens array can be obtained based on the parameters in Table 2, where , and . The calculated intensity distribution map on the focal plane is shown in Fig. 3(a) and a corresponding intensity line profile from the center position is shown in Fig. 3(b). A periodic spot pattern can be clearly seen, which covers an area larger than 0.5 mm × 0.5 mm. The pattern contains more than a nine by nine spot array, caused by interference. The calculated period based on Eq. (12) is 52.5 µm which matches with the period in Figs. 3(a) and 3(b). The uneven light intensity distribution among the spots is a property of the phase grating, which depends on the modulation depth and shape of the phase grating [30, 31].
Fig. 3 The light intensity distribution map (left column) and a corresponding intensity line profile (right column) at the focal plane. (a) and (b) are obtained by theoretical analysis; (c) and (d) are obtained by Zemax simulation; (e) and (f) are obtained by experiment. The location for the line profile is marked by an arrow.
Zemax simulation can also be carried out to simulate the optical system. The Zemax file of the cross cylindrical lens array is provided by SUSS MicroOptics. The light distribution on the focal plane can be simulated by the physical optics analysis module. The 2D intensity distribution map is shown in Fig. 3(c) and a corresponding line profile is shown in Fig. 3(d). The Zemax simulation result matches very well with the above analytical result. The number of spots and the spacing between spots are all matched. A small difference is that the spot size obtained by Zemax simulation is larger than the ideal focal spot depicted in Fig. 3(a). This may have been caused by the finite value of the thickness of the cross cylindrical lens array used in the Zemax simulation, where the lens thickness is neglected in the theoretical calculation in Fig. 3(a).
In the experiment, a Fabry-Perot laser operated at 660 nm wavelength (S1FC660, Thorlabs, Newton, United States) is used instead of a nanosecond pulse laser only for the characterization of the beam profile, concerning the potential damage. The collimated beam from the fiber coupled laser is expanded to nearly 9 mm in diameter to match with the output beam size of the nanosecond pulse laser. A CCD camera (MU300, AmScope, Irvine, United States) monitors the light intensity at the focal plane. The 2D intensity distribution map is shown in Fig. 3(e) and a line profile is shown in Fig. 3(f). As we can see, the experimental results match reasonably well with the analytical and Zemax simulation results. With the cross cylindrical lens array, the light distribution on the focal plane is split into a spot array. Compared with the single focal spot generated by a single lens without the lens array, such a spot array distribution can lower the peak power density on the fiber tip and thus enable coupling higher energy into the fiber without damage.
For some high energy pulsed lasers used in PAT, “hot spots” exist in the beam profile due to beam inhomogeneity, where the power density is higher than the surrounding area. The lens array based beam homogenizer can also reduce the effect of “hot spot”. If a “hot spot” exists in the input wave, it will be transformed to spherical wavelet after the lens array. The spherical wavelet contains different propagation angles, which will be focused to different locations on the focal plane of the coupling lens. Therefore, the energy from the “hot spot” will be redistributed to a relatively large area on the focal plane, reducing the potential of thermal damage.
To check whether the beam distribution pattern at the input of the fiber affects the fiber output, the beam profile at the output of the fiber is measured experimentally. The diverged beam output from a 1000-μm-core-diameter fiber (FT1000UMT, Thorlabs, Newton, United States) is imaged by the CCD camera and shown in Fig. 4. The 2D intensity distribution map is shown in Fig. 4(a) and an intensity line profile is shown in Fig. 4(b). A quasi-Gaussian beam is obtained at the fiber output. This shows that the spot pattern at the input end of the fiber can be eliminated after propagating inside the multimode fiber. The elimination of the spot pattern is likely caused by the effect of a large number of spatial modes propagating in the multimode fiber. The speckles observed in the beam profile are also caused by the nature of multimode fiber [32]. The quasi-Gaussian beam profile at the output of the multimode fiber shows that using the cross cylindrical lens array for light coupling does not affect the uniformity of illumination on sample for PA imaging. In fiber bundle, the beam profile is usually discretized because the bundle contains many fibers. Therefore, a multimode fiber can provide a more well-defined beam profile than fiber bundle.
Fig. 4 Measured light intensity distribution map (a) and line profile; (b) at the output of the multimode fiber. The location for the line profile is marked by an arrow.
3. Coupling high energy pulses into a multimode fiber for PAT
3.1 Experimental setup
The cylindrical lens array is used to couple high-energy pulses into a multimode fiber. A PAT system using the multimode fiber to deliver light is developed. A schematic of the PAT system is shown in Fig. 5. A Q-switched Nd:YAG laser (Surelite OPO Plus SLIII-10, Continuum, San Jose, United States) generates 3-5 ns laser pulses at 10 Hz repetition rate. The laser output is frequency doubled to 532 nm, which then pumps an optical parametric oscillator (OPO). The OPO output wavelength is tunable from 675 nm to 2500 nm and the maximum output energy is 120 mJ. After the OPO, the light is coupled into a multimode fiber (FT1000UMT/FT1500UMT, Thorlabs, Newton, United States) by the cylindrical lens array and the coupling lens. The multimode fiber delivers laser pulses to the sample. The beam diameter after OPO is 9.5 mm. Hence, the lens array can be fully covered. The PA signal is detected by a wide-bandwidth linear transducer array (L14-5/38, Analogic, Richmond, Canada) with 128 channels, a central frequency of 7.2 MHz, and a 70% fractional bandwidth at −6 dB. The received data is then transferred to a data acquisition system (DAQ) and is sent to the ultrasound machine (Ultrasonix MDP, Analogic, Richmond, Canada). Synchronization between the laser firing and data acquisition is achieved by using a signal from the laser to trigger the DAQ. The reconstruction method uses the basic back-projection algorithm [33].
Fig. 5 Experimental setup of the photoacoustic tomography with fiber delivering high energy pulses. The laser beam from the optical parametric oscillator (OPO) is homogenized and coupled into a multimode fiber to deliver light to phantom. DAQ: data acquisition system.
3.2 Performance of fiber coupling
To evaluate the coupling performance with the lens array, the pulse energy at the fiber output is measured by an energy sensor (J-50MB-YAG, Coherent, Palo Alto, United States). The output pulse energy is averaged over one hundred consecutive pulses. The laser wavelength is tuned over the spectrum range from 675 nm to 900 nm. Light coupling into multimode fibers with 1000 µm and 1500 µm core diameters are investigated respectively. The fiber length is about 1 to 2 m. The results of the output pulse energy at different wavelengths are shown in Figs. 6(a) and 6(b) for the 1000 µm and 1500 µm fiber, respectively. The maximum output for the 1000-μm-core-diameter fiber is about 48 mJ/pulse at 750 nm wavelength. By comparison, the maximum output for the 1500-μm-core-diameter fiber is about 64 mJ/pulse at 700 nm wavelength. In Figs. 6(a) and 6(b), the fluctuation of the output energy of the fiber is mainly affected by the energy variation of the laser output of the pulse laser. Figure 6(c) shows the coupling efficiency of the two types of fiber. The coupling efficiency is calculated as the ratio of the energy at the output end of the fiber to the energy at the input end of the fiber. The coupling efficiency reaches 70% and 90% for the 1000-μm and 1500-μm-core-diameter fibers, respectively.
Fig. 6 Output power characterization at different wavelengths and fiber core sizes. (a) Averaged pulse energy coupled into a 1000-µm-core-diameter fiber (blue curve). The error bar indicates the energy fluctuation measured over one hundred consecutive pulses. The laser energy measured at the input of the fiber is shown in red curve. (b) Similar power characterization as in (a) but with the light coupled into a 1500-µm-core-diameter fiber. (c) Comparison of the coupling efficiency of the two types of fiber.
The lens array and the coupling lens introduce some additional insertion loss, which is measured to be ~10%. The laser output from the OPO is slightly divergent, which affects the intensity distribution at the focal plane and makes the dimension of the focusing pattern slightly larger than the theoretical size estimated for a collimated beam. As a result, the 1500-μm-core-diameter fiber under the same optical setup obtains higher coupling efficiency than the 1000-μm-core-diameter fiber. However, 1500-μm-core-diameter is more rigid than the 1000-μm-core-diameter fiber and fiber with smaller core diameter is more preferable for transurethral illumination in prostate imaging. Hence, the 1000-μm-core-diameter fiber is selected and applied in the following PAT imaging.
In comparison, the traditional coupling scheme of using a single lens (without the beam homogenizing lens array) is also tested. The coupling energy and efficiency are found to be much lower than the situation with the lens array. The observed coupled energy without the lens array is typically less than 20 mJ/pulse and the coupling efficiency is typically less than 40%.
In Section 2, we have shown that the beam homogenization system with the cross cylindrical lens array generates a 2D array of beam spots on the focal plane of the coupling lens. The distributed intensity on the focal plane reduces the peak energy density on the fiber tip. Thus higher energy pulses can be coupled into the multimode fiber without damage. The 2D spot array on the focal plane is a result of the coherence and interference property of the spherical wavelets after the lens array. To further improve the coupled energy, the beam can be homogenized into a more uniform distribution by reducing the coherence of the light from the pulsed laser. For incoherent light, the superposition of the multiple spherical wavelets becomes simple intensity summation, which will form a relatively uniform intensity distribution on the focal plane.
3.3 Preliminary PAT Imaging
The high-energy pulses delivered by the multimode fiber are tested for PAT imaging. The 1000-μm-core-diameter fiber is used and the maximum output is about 48 mJ/pulse. A phantom containing a piece of paper with printed dots array placed in gelatin is imaged. The diameter of the dots is 0.1 mm and the spacing between two adjacent dots is 1 mm. The sample is illuminated by the fiber placed 4 cm away, which forms a circular beam around 10 mm in diameter. The PAT image of the dots array is shown in Fig. 7(a). The excited area is determined by the beam size. Next, a 5 mm thick chicken breast tissue is added on top of the phantom to increase the optical depth and simulate optical diffusion in biological tissue. Figure 7(b) shows the PAT image of the dots array with the 5 mm thick chicken breast tissue on top. As the light is diffused inside the chicken breast tissue, the beam size is increased significantly and PA signal from a larger area is excited. The reconstructed image in Fig. 7(b) shows at least 7 cm2 excitation area. Good signal level is still obtained over this large excitation area, as the signal-to-noise ratio drops from 16.1 dB to 10.5 dB due to the decrease of fluence.
Fig. 7 PAT image of printed dots array. (a) Imaging result without the cover of chicken breast and the effective imaged area in a 10-mm-diameter circle. (b) Imaging result with the cover of 5 mm thick chicken breast and the increased effective imaged area in a 30-mm-diameter circle.
In vivo PAT imaging is also performed on the human forearm of a volunteer. The excitation wavelength is set to be 750 nm and the illumination area on the skin is a circle of ~3 cm diameter. The energy density at the skin surface is controlled to be under 10 mJ/cm2, which is lower than the maximum permissible exposure (26 mJ/cm2 at 750 nm) by ANSI standard [26]. A coupling pad is placed between the forearm and the linear transducer array for acoustic coupling. The illumination fiber is placed on the same side with the transducer, forming a diverged circular pattern on the region of interest. Figure 8(a) shows the reconstructed PAT image (colored) on top of the ultrasound image (gray scale) and its SNR for PA signal is 15.8 dB. PAT signal is obtained from the skin surface and blood vessels. Three veins can be identified from the PAT image. One shallow capillary with lower contrast appears between the second and the third vein, which is also labeled in Fig. 8(a). Figure 8(b) shows the photography of the imaging location in the forearm and the corresponding blood vessels. These experimental results demonstrate the large-area imaging capability of the PAT system with the high energy pulses delivered by the multimode fiber.
Fig. 8 In vivo PAT image of forearm. (a) Overlay of PAT image (colored) on top of ultrasound image (grey scale). The PAT image highlights the skin surface and locations of several blood vessels labeled with numbers. (b) Photograph of the forearm where the blood vessels imaged by PAT are also marked. The dash line shows the place of the cross-sectional image.
4. Conclusion
To increase the pulse energy that can be coupled into a multimode fiber and reduce the fiber damage, a fiber coupling scheme with beam homogenization is developed and demonstrated for PAT imaging. The beam homogenization is achieved by using a cross cylindrical lens array, where the focused pulse energy is distributed over a 2D array of beam spots. Our results demonstrate that by adding the beam homogenization, the pulse energy delivered in a single fiber can be significantly improved. We have obtained the delivery of 48 mJ/pulse by a 1000-μm-core-diameter fiber and 64 mJ/pulse by a 1500-μm-core-diameter fiber. To the best of our knowledge, this is the highest output energy reported to date using a single fiber delivery for PAT imaging. With the increased output energy, the PAT system can illuminate a larger area at the same fluence or provide deeper penetration.
Funding
Natural Sciences and Engineering Research Council of Canada (NSERC); Collaborative Health Research Project (CHRP 446576-13); Canadian Institutes of Health Research (CIHR), Collaborative Health Research Project (CPG-127772).
Acknowledgments
Thanks SUSS MicroOptics for sharing the Zemax file of the lens array.
References and links
1. L. V. Wang and S. Hu, “Photoacoustic tomography: in vivo imaging from organelles to organs,” Science 335(6075), 1458–1462 (2012). [CrossRef] [PubMed]
2. L. V. Wang, “Multiscale photoacoustic microscopy and computed tomography,” Nat. Photonics 3(9), 503–509 (2009). [CrossRef] [PubMed]
3. J. J. Niederhauser, M. Jaeger, R. Lemor, P. Weber, and M. Frenz, “Combined ultrasound and optoacoustic system for real-time high-contrast vascular imaging in vivo,” IEEE Trans. Med. Imaging 24(4), 436–440 (2005). [CrossRef] [PubMed]
4. N. Kuo, H. J. Kang, D. Y. Song, J. U. Kang, and E. M. Boctor, “Real-time photoacoustic imaging of prostate brachytherapy seeds using a clinical ultrasound system,” J. Biomed. Opt. 17(6), 066005 (2012). [CrossRef] [PubMed]
5. M. A. Lediju Bell, X. Guo, D. Y. Song, and E. M. Boctor, “Transurethral light delivery for prostate photoacoustic imaging,” J. Biomed. Opt. 20(3), 036002 (2015). [CrossRef] [PubMed]
6. R. A. Kruger, C. M. Kuzmiak, R. B. Lam, D. R. Reinecke, S. P. Del Rio, and D. Steed, “Dedicated 3D photoacoustic breast imaging,” Med. Phys. 40(11), 113301 (2013). [CrossRef] [PubMed]
7. J. M. Yang, K. Maslov, H. C. Yang, Q. Zhou, K. K. Shung, and L. V. Wang, “Photoacoustic endoscopy,” Opt. Lett. 34(10), 1591–1593 (2009). [CrossRef] [PubMed]
8. L. Lin, J. Xia, T. T. Wong, L. Li, and L. V. Wang, “In vivo deep brain imaging of rats using oral-cavity illuminated photoacoustic computed tomography,” J. Biomed. Opt. 20(1), 016019 (2015). [CrossRef] [PubMed]
9. S. Manohar, A. Kharine, J. C. G. Hespen, W. Steenbergen, and T. G. Leeuwen, “The Twente Photoacoustic Mammoscope: system overview and performance,” Phys. Med. Biol. 50(11), 2543–2557 (2005). [CrossRef] [PubMed]
10. W. Xia, D. Piras, M. K. A. Singh, J. C. van Hespen, T. G. van Leeuwen, W. Steenbergen, and S. Manohar, “Design and evaluation of a laboratory prototype system for 3D photoacoustic full breast tomography,” Biomed. Opt. Express 4(11), 2555–2569 (2013). [CrossRef] [PubMed]
11. M. Heijblom, D. Piras, F. M. van den Engh, M. van der Schaaf, J. M. Klaase, W. Steenbergen, and S. Manohar, “The state of the art in breast imaging using the Twente Photoacoustic Mammoscope: results from 31 measurements on malignancies,” Eur. Radiol. 26(11), 3874–3887 (2016). [CrossRef] [PubMed]
12. T. Kitai, M. Torii, T. Sugie, S. Kanao, Y. Mikami, T. Shiina, and M. Toi, “Photoacoustic mammography: initial clinical results,” Breast Cancer 21(2), 146–153 (2014). [CrossRef] [PubMed]
13. A. Horiguchi, K. Tsujita, K. Irisawa, T. Kasamatsu, K. Hirota, M. Kawaguchi, M. Shinchi, K. Ito, T. Asano, H. Shinmoto, H. Tsuda, and M. Ishihara, “A pilot study of photoacoustic imaging system for improved real-time visualization of neurovascular bundle during radical prostatectomy,” Prostate 76(3), 307–315 (2016). [CrossRef] [PubMed]
14. H. S. Salehi, T. Wang, P. D. Kumavor, H. Li, and Q. Zhu, “Design of miniaturized illumination for transvaginal co-registered photoacoustic and ultrasound imaging,” Biomed. Opt. Express 5(9), 3074–3079 (2014). [CrossRef] [PubMed]
15. S. Seidel and G. Phillipps, “Pulse lengthening by intracavity stimulated Brillouin scattering in a Q-switched, phase-conjugated Nd:YAG laser oscillator,” Appl. Opt. 32(36), 7408–7417 (1993). [CrossRef] [PubMed]
16. B. C. Stuart, M. D. Perry, R. D. Boyd, J. A. Britten, B. W. Shore, M. D. Feit, and A. M. Rubenchik, “Development of high damage threshold optics for petawatt-class short-pulse lasers,” Proc. SPIE 2377, 247–359 (1995). [CrossRef]
17. K. Walter, Solid-state Laser Engineering (Springer, 2013).
18. L. Gallais, J. Natoli, and C. Amra, “Statistical study of single and multiple pulse laser-induced damage in glasses,” Opt. Express 10(25), 1465–1474 (2002). [CrossRef] [PubMed]
19. R. A. Robinson and I. K. Ilev, “Design and optimization of a flexible high-peak-power laser-to-fiber coupled illumination system used in digital particle image velocimetry,” Rev. Sci. Instrum. 75(11), 4856–4862 (2004). [CrossRef]
20. J. Laufer, C. Elwell, D. Delpy, and P. Beard, “Spatially resolved blood oxygenation measurements using time-resolved photoacoustic spectroscopy,” in Oxygen Transport to Tissue XXVII, Vol. 578 of Advances in Experimental Medicine and Biology, G. Cicco, D. F. Bruley, M. Ferrari, and D. K. Harrison, eds. (Springer, 2006).
21. L. G. Montilla, R. Olafsson, D. R. Bauer, and R. S. Witte, “Real-time photoacoustic and ultrasound imaging: a simple solution for clinical ultrasound systems with linear arrays,” Phys. Med. Biol. 58(1), N1–N12 (2013). [CrossRef] [PubMed]
22. D. Razansky, A. Buehler, and V. Ntziachristos, “Volumetric real-time multispectral optoacoustic tomography of biomarkers,” Nat. Protoc. 6(8), 1121–1129 (2011). [CrossRef] [PubMed]
23. A. Buehler, E. Herzog, D. Razansky, and V. Ntziachristos, “Video rate optoacoustic tomography of mouse kidney perfusion,” Opt. Lett. 35(14), 2475–2477 (2010). [CrossRef] [PubMed]
24. X. L. Deán-Ben, T. F. Fehm, M. Gostic, and D. Razansky, “Volumetric hand-held optoacoustic angiography as a tool for real-time screening of dense breast,” J. Biophotonics 9(3), 253–259 (2016). [CrossRef] [PubMed]
25. H. He, A. Buehler, and V. Ntziachristos, “Optoacoustic endoscopy with curved scanning,” Opt. Lett. 40(20), 4667–4670 (2015). [CrossRef] [PubMed]
26. American National Standards Institute, Safe Use of Lasers (Laser Institute of America, 2000).
27. T. Schmidt-Uhlig, P. Karlitschek, M. Yoda, Y. Sano, and G. Marowsky, “Laser shock processing with 20 MW laser pulses delivered by optical fibers,” Eur. Phys. J. Appl. Phys. 9(3), 235–238 (2000). [CrossRef]
28. T. Schmidt-Uhlig, P. Karlitschek, G. Marowsky, and Y. Sano, “New simplified coupling scheme for the delivery of 20 MW Nd:YAG laser pulses by large core optical fibers,” Appl. Phys. B 72(2), 183–186 (2001). [CrossRef]
29. M. C. Teich and B. E. A. Saleh, Fundamentals of Photonics (Wiley Interscience, 2007).
30. J. W. Goodman, Introduction to Fourier Optics, 3rd ed. (Roberts and Company, 2005).
31. R. Magnusson and T. K. Gaylord, “Diffraction efficiencies of thin phase gratings with arbitrary grating space,” J. Opt. Soc. Am. 68(6), 806–808 (1978). [CrossRef]
32. J. W. Goodman, Speckle Phenomena in Optics: Theory and Applications (Roberts and Company Publishers, 2007).
33. M. Xu and L. V. Wang, “Universal back-projection algorithm for photoacoustic computed tomography,” Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 71(1), 016706 (2005). [CrossRef] [PubMed]
