Open Access
Add to BibSonomy
Abstract
We demonstrate a miniature fiber-optic two two-photon endomicroscopy with microsphere-spliced double-cladding antiresonant fiber for resolution enhancement. An easy-to-operate process for fixing microsphere permanently in an antiresonant fiber core, by arc discharge, is proposed. The flexible fiber-optic probe is integrated with a parameter of 5.8 mm × 49.1 mm (outer diameter × rigid length); the field of view is 210 µm, the resolution is 1.3 µm, and the frame rate is 0.7 fps. The imaging ability is verified using ex-vivo mouse kidney, heart, stomach, tail tendon, and in-vivo brain neural imaging.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Optical microscopic imaging is a significant diagnostic method in the biomedical field, and optical endomicroscopic imaging is a notable approach of this type. The routine gold standard for disease diagnosis includes conventional video endoscopy/laparoscopy/gastroscopy with pathological biopsy [1,2]. During an operation, the attending doctor does not have real-time access to imaging results at subcellular resolution of the patients’ suspicious area. Therefore, an endoscopic imaging method is required that can image with subcellular resolution and offer guidance regarding the suspected diseased tissues for intraoperative decision-making.
Compared to endoscopic optical coherence tomography (OCT) and confocal endomicroscopy [3,4], nonlinear endomicroscopy has the advantages of inherent optical sectioning, deep penetration, and high resolution subcellular structural and functional imaging capabilities [5,6]. Nonlinear endomicroscopy techniques have been developed by many researchers in recent years [7–21]. According to whether the scanning / driving device is located in the probe, the endoscopic system can be divided into proximal scanning type [12,17] and distal scanning type [7–11,13–16,18–21]. Piezoelectric ceramic tube (PZT) scanner with double-cladding fiber is the main scheme among the distal scanning type [10,13–16,18–21], with a higher degree of common-path probe integration [22]. Double-cladding fiber realizes the dual role of femtosecond pulsed light transmission and fluorescence signal collection.
Compared to photonic crystal bandgap fibers, large mode field Kagomé fibers and antiresonant fibers possess unique characteristics of low dispersion, low loss, and wide bandwidth pulse transmission [23–25]. These fibers are attractive for nonlinear imaging applications, especially in coherent anti-stokes Raman scattering (CARS) imaging and multicolor imaging [26]. However, their output NA is low, which makes it difficult to achieve high imaging resolution for fiber-scanning nonlinear endoscope designs without using the fiber-tip engineering method or higher magnification objectives. The conventional method of generating a spherical/hemispherical lens by arc-discharging the fiber [27–29] can lead to local destruction of such photonic crystal fiber microstructures, and the method of attaching a lens at the fiber end face by using a polymer directly [30–32] is also difficult to achieve in hollow fibers, making the fiber-tip engineering scheme for increased output NA of photonic crystal fibers challenging.
From the current state of the art, two main fiber-tip engineering methods have been applied to increase the fiber output NA in nonlinear endomicroscopy: by the graded-index (GRIN) lens or ball lens. Liang et al. [33] proposed a cascaded numerical aperture (NA) strategy for double-cladding fiber. The silica rod and tiny GRIN lens composed a mode-field focuser that nearly tripled the fiber output NA from 0.12 to 0.35. Due to the enhancement of the fiber output NA, a small magnification objective can be used to ensure the imaging resolution of the system. Therefore, this strategy can ultimately enlarge the field of view (FOV) and imaging throughout (i.e., the total number of resolvable pixels per image). Lombardini et al. [34] proposed a microsphere-spliced hollow-core Kagomé-lattice double-clad fiber scheme for a multimodal flexible coherent Raman endoscope. The CO2 laser splicer forms a 30 µm microsphere fixed in the fiber core, and the microsphere acts as a ball lens to converge the 15 µm guided mode into a 1 µm spot. This fiber-tip engineering method improves the NA of the Kagomé-lattice fiber to 0.3. Kudlinski et al. [35] explored the method of microsphere attachment to a hollow-core antiresonant fiber by CO2 laser splicer. The test results of the fiber NA measured in the far-field showed that the fiber output NA is improved 8.7 times from 0.023 to 0.21. Compared to the GRIN lens method, the microsphere-spliced method demonstrates higher NA enhancement effect. Besides, this method has less impact on the fluorescence collection efficiency because the microsphere is spliced to the fiber core while the fiber cladding remains unchanged.
In this report, we promote a two-photon endomicroscopy with microsphere-spliced double-cladding antiresonant fiber for resolution enhancement. Firstly, in the theoretical part, we establish a microsphere model based on the ABCD matrix method. The effects of the microsphere lens radius and refractive index on the output beam waist radius, working distance and numerical aperture are analyzed. Secondly, in the experimental part, a fabrication process for fixing a microsphere in the core of an antiresonant fiber, based on arc discharge, is proposed. The results of near-field spot size and far-field NA tests are presented. Finally, we demonstrate the proposed two-photon endomicroscopy system ability to image ex-vivo mouse kidney, heart, stomach, tail tendon, and in-vivo brain neural imaging.
2. Method
2.1 Two-photon endomicroscopy with microsphere-spliced DC-ARF setup overview
Figure 1 (a) depicts the schematic setup of the two-photon endomicroscopy with a microsphere-spliced double-cladding antiresonant fiber. In short, femtosecond pulse generated by Ti:Sapphire femtosecond pulse laser (Chameleon Vision II, coherent) is coupled to the core of customed-designed double-cladding antiresonant fiber (DC-ARF) by a coupling lens (CL; 354850 B, Lightpath). The DC-ARF diameter of hollow core and silica outer cladding are 24 µm and 134 µm, respectively. The numerical aperture (NA) of the DC-ARF core and outer cladding are 0.034 and 0.5, respectively. More characteristics of DC-ARF can be found in our previous work [36]. A four-quadrant PZT (PT230.94, Physik Instruments) and DC-ARF are fixed by glue reversely [37], forming a PZT-driven fiber scanner. The dual-channel amplitude-modulated sine and cosine waves are amplified by the voltage amplifier (TD250, Piezodrive) and then act on the PZT-driven fiber scanner. The 30-mm DC-ARF cantilever realizes two-dimensional spiral scanning at the second-order resonance point of 700 Hz, resulting in frame rate of 0.7 fps with 512 scanning circles. The reverse fixing structure of PZT-driven fiber scanner minimizes the rigid length of the integrated two-photon endomicroscopy probe.
Fig. 1. (a) Schematic of the two-photon endomicroscopy system (ADC, analog-to digital module; DAC, digital-to-analog module; AMP, amplifier). The femtosecond pulse excitation light path is shown in red, and the fluorescence collection light path is shown in green. (b) SEM of the microsphere-spliced DC-ARF. (c) Photographs of integrated two-photon endomicroscopy probe.
The microsphere (9030 series, Thermofisher Scientific) with an average diameter of 32.5 µm is spliced in the DC-ARF core permanently by electrical discharge. The microsphere acts as a ball lens, and the $N{A_{fiber}}$ of the fiber core is increased approximately nine times to 0.3. The scanning electron microscope (SEM; GeminiSEM 300, Zeiss) image of the microsphere-spliced antiresonant fiber is shown in Fig. 1 (b). The six capillary structures in the inner cladding of antiresonant fiber support the microsphere. The structure and optical design of the miniature objective used in this two-photon endoscope have been described previously [38]. This finite-distance miniature objective theoretical $N{A_{image}}$ and $N{A_{object}}$ are 0.2 and 0.6, which corresponds to a 3× magnification. The two-photon fluorescence signal is reflected by dichroic mirror (DM; DMLP650R, Thorlabs), filtered by a bandpass filter (FF01-530/43-25, Semrock), and collected by a photomultiplier tube (PMT) detector (H10770PA-40-SEL, Hamamatsu). A photograph of an integrated fiber-optic two-photon endomicroscopy probe is depicted in Fig. 1 (c). The reverse-fixed PZT-driven scanner and miniature objective are assembled into the aluminum alloy probe, with an integrated parameter of 5.8 mm × 49.1 mm × 3.5 g (outer diameter × rigid length × weight).
2.2 ABCD matrix analysis of microsphere-spliced DC-ARF
As a typical finite-distance system, the imaging resolution of the two-photon endomicroscopy is determined by formula 1:
where $N{A_{object}}$ is the objective object NA, $N{A_{image}}$ is the objective image NA and M is the objective magnification. The mode field diameter of antiresonant fiber is large and the $N{A_{fiber}}$ is low, When the antiresonant fiber outputs directly, the spot does not fulfill the pupil of the miniature objective and the $N{A_{image}}$ is limited by $N{A_{fiber}}$.The FOV of the two-photon endomicroscopy is determined by formula 2:
where ${D_{scan}}$ is the scanning range of the PZT fiber scanner. As formulas 1 and 2 describe, the objective magnification M can both affect the full width at half-maximum (FWHM) of the point spread function (PSF) ${\delta _{PSF}}$ and the FOV. A higher objective magnification $M$ can induce the ${\delta _{PSF}}$ that achieves a higher resolution, but the FOV decreases at the same magnification.To enhance the lateral resolution while maintain the FOV, we have developed a fiber-tip engineering method for hollow-core DC-ARF. A microsphere is spliced permanently into the DC-ARF core and it acts as a ball lens to increase $N{A_{fiber}}$ by arc charge. In this case, the $N{A_{fiber}}$ is enhanced and the spot fulfills the pupil of the objective. The upper limit of the objective designed $N{A_{object}}$ determines the ultimate resolution. The geometric structure of the microsphere is depicted in Fig. 2. The ABCD transformation matrix can be used to describe Gaussian beam propagation and transformation conveniently. The following presents analyses of the output NA and waist parameters of the microsphere-spliced DC-ARF using the ABCD transformation matrix [39–41].
The q parameter at the waist of the input Gaussian beam is defined as
where ${\omega _0}$ is the mode field radius of the DC-ARF, and $\lambda $ is the working wavelength. In this two-photon endomicroscopy, the mode field radius mode ${\omega _0}$ is 8.6 µm, and working wavelength is 920 nm.When the input Gaussian beam passes through the microsphere lens, it undergoes four transformations. The input Gaussian beam passes through the microsphere surface ${S_1}$, propagates in the uniform medium of the microsphere for distance L, passes through the microsphere surface ${S_2}$, and propagates in air medium for distance d.
The total transformation matrix T of the microsphere is obtained:
Parameter q at the waist of the output Gaussian beam is defined as ${q_c}$:
where ${\omega _c}$ is the waist radius of the Gaussian beam focused by the microsphere, and ${R_c}$ is the radius of curvature of the equal phase plane. Based on the ABCD transformation matrix of Gaussian-beam parameter q,The real and imaginary parts of the formula 6 and formula 7 are equal, and the curvature radius of the equal-phase surface on the focal plane is infinite. Thus, formula 8 and formula 9 can be obtained:
where $x = \frac{{\pi \omega _0^2}}{\lambda }$.By solving the formula 8, we can obtain the working distance $d$:
By analyzing formula 9, when the refractive index of the microsphere medium $n < 2$, the working distance $d > 0$, the output spot is outside the microsphere lens. Conversely, when the refractive index of the microbead medium $n > 2$, the output spot is inside the microsphere lens. This analysis also remains consistent with the analysis based on geometric modeling of the spherical lens [42].
The fiber output $N{A_{fiber}}$ is depicted by formula 11:
The simulation results of the numerical calculations are presented in Fig. 3. The diameter of the spherical lens ranges from 25 to 35 µm, and its refractive index ranges from 1.45 to 1.55. The simulation results indicate that the smaller the radius and higher the refractive index of the microsphere, the stronger the convergence effect on the spot, resulting in a larger $N{A_{fiber}}$. Section 2.2 describes the fabrication and evaluation of the micro-spliced DC-ARF. By utilizing the theoretical average microsphere diameter 32.5 µm and soda-lime glass refractive index 1.52 provided by the manufacturer, we can obtain that under the regulation of the microsphere, the theoretical waist radius ${\omega _c}$ at convergence is 0.80 µm, the working distance is 7.43 µm, and output $N{A_{fiber}}$ of the DC-ARF is 0.34.
Fig. 3. Simulation of the microsphere radius and medium refractive index for the DC-ARF output: (a) waist radius; (b) working distance and (c) numerical aperture. The dashed line in the figure is the equipotential line.
2.3 Fabrication and evaluation of microsphere-spliced DC-ARF
In this section, we describe the fabrication and evaluation of a hollow core DC-ARF with a microsphere-spliced attachment for numerical aperture enhancement. The microsphere (9030 series, Thermofisher Scientific) used in this experiment has a diameter of 32.5 µm ${\pm} $ 1.2 µm (theoretical average diameter and standard deviation, respectively). The microsphere material is uniform soda-lime glass with a refractive index of 1.52 at a wavelength of 589 nm. Figure 4 (a) depicts the SEM (GeminiSEM 300, Zeiss) image of the microsphere. Seven microspheres can be seen, and the diameter of the microsphere measured ranges from 27.67 µm to 36.31 µm.
Fig. 4. (a) SEM image of the microspheres. (b) Fabrication process of the DC-ARF with microsphere attachment.
As shown in Fig. 4 (b), the fabrication process of the DC-ARF with a spliced microsphere involves four main steps. In the first step, single-mode fiber (1060-XP, Nufern) with a cladding diameter of 125 µm is used to make the fiber taper. The taper with a conical-frustum tip facilitates the attraction of the microsphere. The taper is fabricated by the special mode of the fusion splicer (FSM 100P+, Fujikura). In the special mode, after the standard fusion of the left and right single-mode fibers, the left and right motors evacuate backward at a speed of 0.04 µm/ms, and the evacuation time is 1200 ms. Then, the fiber is cleaved from the middle and two conical-frustum-tip tapers are completed. In the second step, the fiber taper is in contact with the microsphere through a three-dimensional displacement table. The process is monitored in real time through a stereomicroscope (XTL-400, Guiguang). Because of the electrostatic force between the fiber taper and microsphere, the microsphere is attached to the conical-frustum tip of the taper. In the third step, the taper with microsphere attached by electrostatic force and the DC-ARF are placed on the left and right sides of the fusion splicer (FSM 100P, Fujikura). Through the positions of the taper and the DC-ARF can be adjusted by controlling the motors carefully, the microsphere is aligned with the core of the antiresonant fiber precisely. Then, the taper is withdrawn in the direction vertical to the placement of the fiber. We note that because the microsphere diameter is larger than the DC-ARF core, the microsphere can be attached to the DC-ARF and supported by the capillary structure around the DC-ARF core. In the final step, after fine adjustment of discharge power and discharge time, the microsphere is spliced permanently into the DC-ARF core. The fiber capillary structure does not collapse, and the microsphere melts slightly without obvious deformation.
A near-infrared semiconductor laser with a center wavelength of 915 nm (LR-PSFJ-915/1–100 mW, Changchun Laser Technology) is selected as the light source for the $N{A_{fiber}}$ test after microsphere attachment. The 915 nm laser outputs by coupling into a single-mode fiber (PM-980, Nufern), and the PM-980 is spliced with the one-side DC-ARF. The far-field measurement results for the $1/{e^2}$ diameter width of the microsphere-spliced DC-ARF output spot is shown in Fig. 5(a), recorded by the beam quality analyzer (Photon NanoScan, Ophir Optronics Solutions). The far-field calculated NA results in the X and Y directions are 0.30 and 0.31, respectively. To test the near-field spot size of the fiber output transformed by the microsphere, we use a 20× objective (Plan N, Olympus) with a complimentary metal–oxide–semiconductor (CMOS) camera (Panda,4.2 M PCO). The objective is employed to magnify the near-field image and to collimate the spot to the CMOS camera. As shown in Fig. 5 (b), the spot size recorded by the CMOS camera is 25.8 µm, which corresponds to an FWHM of 1.24 µm for the PSF to the fiber output. Compared to the DC-ARF without microsphere attachment (NA = 0.034), the output NA is improved by approximately 9×. When the laser output passes through a 1.7-m-antiresonant fiber without microsphere attachment, the output power is 8.2 mW; when a microsphere is attached to the fiber end-face, it becomes 6.9 mW. Therefore, the power loss induced by the microsphere is approximately 15.8% at 915 nm. The fiber output is facing the center and close to the detection surface of the power detector.
Fig. 5. (a) Far-field microsphere-spliced DC-ARF NA measurement at 915 nm. The inset is the single-mode Gaussian beam spot measured at the output of the fiber. (b) Near-field spot recorded by the CMOS camera after the convergence of the microsphere. The scale bar is 100 µm.
3. Experimental results
Firstly, we evaluate the system's imaging key technical parameters such as FOV and lateral resolution. The resolution test samples are prepared as described in detail in [36]. The FOV was tested using 25-µm standard grid samples. As shown in Fig. 6 (a), the maximum scanning FOV of this nonlinear fiber-optic probe is approximately 210 µm. By linearly reducing the drive voltage, the scanning FOV is then reduced to 75 µm, which subsequent resolution tests are measured at this FOV.
Fig. 6. Two-photon imaging of 25 µm standard grid samples (each image shown is the average of 10 image acquisitions to improve the SBR).
Green-fluorescent polymer microspheres (G400 Fluoro-max, Thermo Scientific) with diameter of 380 nm were used to measure the lateral resolution. As depicted in Fig. 7 (a-b), the full width at half-maximum (FWHM) of the point spread function (PSF) for bead 1 and bead 2 were imaged. The FWHM for bead 1 and bead 2 is 1.32 µm and 1.29 µm, respectively, which is shown in Fig. 7 (c).
Fig. 7. (a-b) Two-photon imaging of 380 µm polymer microspheres for lateral resolution measurement (each image shown is the average of 10 image acquisitions to improve the SBR). (c) Normalized intensity profiles of bead 1 and bead 2.
Secondly, we performed experiments on the ex-vivo and in-vivo mice tissues to verify the two-photon endomicroscopy capability. 920 nm is the effective wavelength for excitation of biosensors such as gcamp6 and GFP. Figure 8 (a) depicts the renal tubules in ex-vivo eGFP (enhanced green fluorescent protein) mouse kidney. Figure 8 (b) depicts the muscle tissue in ex-vivo eGFP mouse heart tissue. Figure 8 (c) depicts gastric pit in ex-vivo eGFP stomach tissue inner wall. Mouse tendon tissue contains a large amount of highly-ordered collagen fibers, which are the main source of second harmonic generation. Therefore, we selected ex-vivo mouse tail tendon tissue to verify the imaging ability for the second-harmonic generation signal. Forty-week-old male Thy1 mice weighing 30 g were used. Tail amputation was performed after anesthesia. After the end-face of the tail was crosscut, it was fixed with 4% low gelling temperature agarose (a9414-25 g, Sigma) for two-photon imaging. Figure 8 (d) depicts the collagen in crosscut tail tendon tissue. Finally, we performed in-vivo static brain neural imaging on the eGFP-labeled mouse. The mouse is fixed to a body fixator under a microscope (SZN, Soptop), and the handheld cranial drill (RWD Life Science) is activated to polish the skull until the brain tissue and blood vessels are exposed. Then, the bone tissue is peeled off with tissue forceps and protected with 0.9% sodium chloride injection. The mouse is fixed to a body fixator and placed under the two-photon endomicroscopy probe for two-photon neural imaging. As shown in the Fig. 9 (e), the structure of microvessels and neural cell bodies can be seen. In the imaging experiments above, the power under the probe is less than 70 mW.
Fig. 8. Two-photon imaging results of (a) ex-vivo eGFP mouse kidney tissue. (b) ex-vivo eGFP mouse heart tissue. (c) ex-vivo eGFP inner wall of stomach tissue. (d) ex-vivo second-harmonic generation imaging of Thy1 mice tail tendon tissue. (e) In-vivo eGFP static brain neural imaging (Each image shown is the average of 10 image acquisitions to improve the SBR).
Fig. 9. (a) The photograph of the miniature objective. (b) The schematic diagram of the miniature objective resolution test system. (c) Resolution test image of the USAF target on the CMOS camera. (d) Normalized intensity measurement at 360 lp/mm and 400 lp/mm.
4. Discussion
In this section, we discuss the current state of the two-photon endomicroscopy resolution with microsphere-spliced double-cladding antiresonant fiber. When the double-cladding antiresonant fiber is directly applied to the two-photon endomicroscopy system with no fiber engineering method, the lateral resolution is 3.1 µm. The system resolution is limited by the large mode field diameter and low $N{A_{fiber}}$[36]. By attaching the microsphere to the antiresonant fiber core, the $N{A_{fiber}}$ is enhanced, and the resolution of the two-photon endomicroscopy is improved to 1.3 µm. The microsphere spliced in the antiresonant fiber core acts as a ball lens to expand $N{A_{fiber}}$ so that it fills the entrance pupil of the micro-objective. In this case, we note that the resolution is limited by the spatial resolution of the miniature objective.
The photographs of the miniature objective used in the two-photon endomicroscopy probe are shown in Fig. 9 (a), and the structural design and simulation results of the modulation transfer function (MTF) has been reported in our previous report [38]. The design results showed that MTF at 370 lp / mm is > 0.5, which can meet the design requirements. We conducted supplementary tests on the resolution measurement of the micro-objective. The schematic diagram of the miniature objective resolution test system is depicted in Fig. 9 (b). The test wavelength of 920 nm is transmitted through the high-resolution microscopy USAF target (Edmund Optics). The objective (Plan N, Olympus) and the focusing lens with focal length =35 mm (LA1027-A, Thorlabs) form a relay system to converge the light beam to the CMOS camera (Panda,4.2 M PCO). The resolution test image of the USAF target is shown in Fig. 9 (c), and the normalized intensity measurement at spatial frequency 360 lp/mm and 400 lp/mm are described in Fig. 9 (d). The experimental test results show that the MTF of the miniature objective is > 0.5 at 360 lp / mm and < 0.5 at 400 lp/mm. The MTF is > 0.5 at spatial frequency 360 lp / mm, which indicates that the miniature objective can distinguish objects separated by 1.38 µm with a high contrast. Therefore, when MTF = 0.5, the corresponding spatial frequency should be between 360–400 lp/mm.
Following the analysis in Refs. [33,43], we use the figure of merit (FOM) to evaluate the impact of this microsphere attachment method on the imaging throughout of the two-photon endomicroscopy. The FOM is used to quantify the number of points that can be resolved in the horizontal direction of the FOV, which can be seen in formula 12:
The imaging throughput (the total number of points that can be resolved in a spiral scanning pattern) can be estimated by applying formula 13:
In our former work [36], the FOV of the fiber-optic scanning two-photon endomicroscopy was 200 µm, and the lateral resolution was 3.1 µm. The FOM and imaging throughout of the system was calculated to be 64.5 and 4160, respectively. In this work, the FOV is 210 µm, and the lateral resolution is enhanced to 1.3 µm. The FOM and imaging throughout of the system are calculated to be 151.5 and 26094, respectively. By applying the microsphere-spliced method for double-cladding antiresonant fiber, the two-photon imaging throughout is increased by 6.2 times.
One limitation of this work is the mismatch between the fiber output $N{A_{fiber}}$ (0.3) and the miniature objective $N{A_{image}}$ (0.2), which leads to energy loss. Two solutions can be considered: reducing the $N{A_{fiber}}$, such as using an antiresonant fiber with a larger mode field and a larger microsphere, or improving the miniature objective design $N{A_{image}}$ at the fiber end. The $N{A_{fiber}}$ should be consistent with the design $N{A_{image}}$ of the miniature objective. At this time, the system resolution and energy utilization are optimal.
It is worth noting that compared with the former method of fixing the microsphere by CO2 laser splicer [34,35], the method of fixing the microsphere by arc discharge is easier to implement at lower equipment cost. Furthermore, this microsphere-spliced NA enhancement scheme can also be applied to other hollow core fiber structures, such as conjoined-tube negative-curvature fiber [44], which can expand the application of hollow core fiber in nonlinear imaging.
5. Conclusion
In summary, a two-photon endomicroscopy prototype with microsphere-spliced double-cladding antiresonant fiber has been demonstrated with an integrated outer diameter of 5.8 mm, a rigid length of 49.1 mm, and weight of 3.5 g. In the theoretical part, a microsphere model based on the ABCD transformation matrix method is established; The effects of the microsphere radius and refractive index on the output beam waist, working distance, and numerical aperture are analyzed. In the experimental part, a method of fixing microspheres on the core of an antiresonant fiber, based on arc discharge, is proposed and verified experimentally. The results of near-field spot size and far-field NA tests are presented. The imaging FOV of this flexible probe is 210 µm, the resolution is 1.3 µm, and the frame rate is 0.7 fps. The ex-vivo mouse kidney, heart, stomach, tail tendon tissue, and in-vivo brain neuronal imaging results verify the capability of this two-photon endomicroscopy probe. This optical imaging instrument shows promising potential for in-vivo real-time internal organ disease diagnosis.
Funding
National Key Research and Development Program of China (2020YFB1312802); National Natural Science Foundation of China (31830036, 61973019, 61975002).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. W. Liang, G. Hall, B. Messerschmidt, M. J. Li, and X. Li, “Nonlinear optical endomicroscopy for label-free functional histology in vivo,” Light: Sci. Appl. 6(11), e17082 (2017). [CrossRef]
2. M. J. Gora, M. J. Suter, G. J. Tearney, and X. Li, “Endoscopic optical coherence tomography: technologies and clinical applications [Invited],” Biomed. Opt. Express 8(5), 2405–2444 (2017). [CrossRef]
3. M. Minsky, “Microscopy apparatus,” US Patent 3013467, December 19, 1961.
4. J. B. Pawley, Handbook of Biological Confocal Microscopy, 3rd ed. (Springer, 2006).
5. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning fluorescence microscopy,” Science 248(4951), 73–76 (1990). [CrossRef]
6. W. R. Zipfel, R. M. Williams, and W. W. Webb, “Nonlinear magic: multiphoton microscopy in the biosciences,” Nat. Biotechnol. 21(11), 1369–1377 (2003). [CrossRef]
7. M. T. Myaing, D. J. Macdonald, and X. Li, “Fiber-optic scanning two-photon fluorescence endoscope,” Opt. Lett. 31(8), 1076–1078 (2006). [CrossRef]
8. L. Fu, A. Jain, H. Xie, C. Cranfield, and M. Gu, “Nonlinear optical endoscopy based on a double-clad photonic crystal fiber and a MEMS mirror,” Opt. Express 14(3), 1027–1032 (2006). [CrossRef]
9. S. Tang, W. Jung, D. T. McCormick, T. Xie, J. Su, Y. C. Ahn, B. J. Tromberg, and Z. Chen, “Design and implementation of fiber-based multiphoton endoscopy with microelectromechanical systems scanning,” J. Biomed. Opt. 14(3), 034005 (2009). [CrossRef]
10. Y. Zhao, H. Nakamura, and R. J. Gordon, “Development of a versatile two-photon endoscope for biological imaging,” Biomed. Opt. Express 1(4), 1159–1172 (2010). [CrossRef]
11. D. R. Rivera, C. M. Brown, D. G. Ouzounov, I. Pavlova, D. Kobat, W. W. Webb, and C. Xu, “Compact and flexible raster scanning multiphoton endoscope capable of imaging unstained tissue,” Proc. Natl. Acad. Sci. U. S. A. 108(43), 17598–17603 (2011). [CrossRef]
12. D. M. Huland, C. M. Brown, S. S. Howard, D. G. Ouzounov, I. Pavlova, K. Wang, D. R. Rivera, W. W. Webb, and C. Xu, “In vivo imaging of unstained tissues using long gradient index lens multiphoton endoscopic systems,” Biomed. Opt. Express 3(5), 1077–1085 (2012). [CrossRef]
13. Y. Zhang, M. L. Akins, K. Murari, J. Xi, M. Li, K. L. Phelps, M. Mahendroo, and X. Li, “A compact fiber-optic SHG scanning endomicroscope and its application to visualize cervical remodeling during pregnancy,” Proc. Natl. Acad. Sci. U. S. A. 109(32), 12878–12883 (2012). [CrossRef]
14. D. Do, H. Yoo, and D. G. Gweon, “Fiber-optic raster scanning two-photon endomicroscope using a tubular piezoelectric actuator,” J. Biomed. Opt. 19(6), 066010 (2014). [CrossRef]
15. G. Ducourthial, P. Leclerc, T. Mansuryan, M. Fabert, J. Brevier, R. Habert, F. Braud, R. Batrin, C. Vever-Bizet, G. Bourg-Heckly, L. Thiberville, A. Druilhe, A. Kudlinski, and F. Louradour, “Development of a real-time flexible multiphoton microendoscope for label-free imaging in a live animal,” Sci. Rep. 5(1), 18303 (2015). [CrossRef]
16. Y. Wang, Z. Li, X. Liang, and L. Fu, “Four-plate piezoelectric actuator driving a large-diameter special optical fiber for nonlinear optical microendoscopy,” Opt. Express 24(17), 19949–19960 (2016). [CrossRef]
17. A. Lukic, S. Dochow, H. Bae, G. Matz, I. Latka, B. Messerschmidt, M. Schmitt, and J. Popp, “Endoscopic fiber probe for nonlinear spectroscopic imaging,” Optica 4(5), 496–501 (2017). [CrossRef]
18. F. Akhoundi, Y. Qin, N. Peyghambarian, J. K. Barton, and K. Kieu, “Compact fiber-based multi-photon endoscope working at 1700nm,” Biomed. Opt. Express 9(5), 2326–2335 (2018). [CrossRef]
19. D. Y. Kim, K. Hwang, J. Ahn, Y. H. Seo, J. B. Kim, S. Lee, J. H. Yoon, E. Kong, Y. Jeong, S. Jon, P. Kim, and K. H. Jeong, “Lissajous scanning two-photon endomicroscope for in vivo tissue imaging,” Sci. Rep. 9(1), 3560 (2019). [CrossRef]
20. A. Li, H. Guan, H. C. Park, Y. Yue, D. Chen, W. Liang, M. Li, H. Lu, and X. Li, “Twist-free ultralight two-photon fiberscope enabling neuroimaging on freely rotating/walking mice,” Optica 8(6), 870–879 (2021). [CrossRef]
21. E. Pshenay-Severin, H. Bae, K. Reichwald, G. Matz, J. Bierlich, J. Kobelke, A. Lorenz, A. Schwuchow, T. Meyer-Zedler, M. Schmitt, B. Messerschmidt, and J. Popp, “Multimodal nonlinear endomicroscopic imaging probe using a double-core double-clad fiber and focus-combining micro-optical concept,” Light: Sci. Appl. 10(1), 207 (2021). [CrossRef]
22. V. Kučikas, M. P. Werner, T. Schmitz-Rode, F. Louradour, and V. M. A. M. J. Zandvoort, “Two-Photon Endoscopy: State of the Art and Perspectives,” Mol. Imaging Biol 1, 1–15 (2021). [CrossRef]
23. C. Martin, B. Debord, F. Gérôme, F. Benabid, and A. Ben-Yakar, “Multiphoton imaging using a high transmission broadband hypocycloid-shape Kagome HC-PCF,” in Conference on Lasers and Electro-Optics, OSA Technical Digest, 2016, JW2A.6.
24. H. Cui, C. Wang, J. Ma, H. Liu, D. Wu, A. Wang, and L. Feng, “Forward-detection fiber scanning endomicroscopy for two-photon fluorescence,” OSA Biophotonics Congress: Optics in the Life Sciences, 2021, JTu4a.21.
25. M. Andreana, T. Le, W. Drexler, and A. Unterhuber, “Ultrashort pulse Kagome hollow-core photonic crystal fiber delivery for nonlinear optical imaging,” Opt. Lett. 44(7), 1588–1591 (2019). [CrossRef]
26. H. Guan, W. Liang, A. Li, Y. A. Gau, D. Chen, M. J. Li, D. E. Bergles, and X. Li, “Multicolor fiber-optic two-photon endomicroscopy for brain imaging,” Opt. Lett. 46(5), 1093–1096 (2021). [CrossRef]
27. H. Y. Choi, S. Y. Ryu, J. Na, B. H. Lee, I. B. Sohn, Y. C. Noh, and J. Lee, “Single-body lensed photonic crystal fibers as side-viewing probes for optical imaging systems,” Opt. Lett. 33(1), 34–36 (2008). [CrossRef]
28. H. Yoo, J. W. Kim, M. Shishkov, E. Namati, T. Morse, R. Shubochkin, J. R. McCarthy, V. Ntziachristos, B. E. Bouma, F. A. Jaffer, and G. J. Tearney, “Intra-arterial catheter for simultaneous microstructural and molecular imaging in vivo,” Nat. Med. 17(12), 1680–1684 (2011). [CrossRef]
29. W. Yuan, R. Brown, W. Mitzner, L. Yarmus, and X. Li, “Super-achromatic monolithic microprobe for ultrahigh-resolution endoscopic optical coherence tomography at 800 nm,” Nat. Commun. 8(1), 1531 (2017). [CrossRef]
30. M. Zhao, Y. Huang, and J. U. Kang, “Sapphire ball lens-based fiber probe for common-path optical coherence tomography and its applications in corneal and retinal imaging,” Opt. Lett. 37(23), 4835–4837 (2012). [CrossRef]
31. S. B. Wen, V. M. Sundaram, D. McBride, and Y. Yang, “Low-cost, high-precision micro-lensed optical fiber providing deep-micrometer to deep-nanometer-level light focusing,” Opt. Lett. 41(8), 1793–1796 (2016). [CrossRef]
32. M. Marrese, H. Offerhaus, E. Paardekam, and D. Iannuzzi, “70 µm diameter optical probe for common-path optical coherence tomography in air and liquids,” Opt. Lett. 43(24), 5929–5932 (2018). [CrossRef]
33. W. Liang, H. C. Park, K. Li, A. Li, D. Chen, H. Guan, Y. Yue, Y. A. Gau, D. E. Bergles, M. J. Li, H. Lu, and X. Li, “Throughput-speed product augmentation for scanning fiber-optic two-photon endomicroscopy,” IEEE Trans. Med. Imaging 39(12), 3779–3787 (2020). [CrossRef]
34. A. Lombardini, V. Mytskaniuk, S. Sivankutty, E. R. Andresen, X. Chen, J. Wenger, M. Fabert, N. Joly, F. Louradour, A. Kudlinski, and H. Rigneault, “High-resolution multimodal flexible coherent Raman endoscope,” Light: Sci. Appl. 7(1), 10 (2018). [CrossRef]
35. A. Kudlinski, A. Cassez, O. Vanvincq, D. Septier, A. Pastre, R. Habert, K. Baudelle, M. Douay, V. Mytskaniuk, V. Tsvirkun, H. Rigneault, and G. Bouwmans, “Double clad tubular anti-resonant hollow core fiber for nonlinear microendoscopy,” Opt. Express 28(10), 15062–15070 (2020). [CrossRef]
36. C. Wang, H. Liu, J. Ma, H. Cui, Y. Li, D. Wu, Y. Hu, D. Wu, Q. Fu, L. Liang, F. Yu, R. Wu, A. Wang, and L. Feng, “Spiral scanning fiber-optic two-photon endomicroscopy with a double-cladding antiresonant fiber,” Opt. Express 29(26), 43124–43135 (2021). [CrossRef]
37. N. Zhang, T. H. Tsai, O. O. Ahsen, K. Liang, H. C. Lee, P. Xue, X. Li, and J. G. Fujimoto, “Compact piezoelectric transducer fiber scanning probe for optical coherence tomography,” Opt. Lett. 39(2), 186–188 (2014). [CrossRef]
38. D. Wu, L. Feng, and A. Wang, “Optical design of two-photon endoscopy objective with high collection efficiency,” Laser Optoelectron. Prog. 55(07), 400–405 (2018).
39. W. Jung, W. Benalcazar, A. Ahmad, U. Sharma, H. Tu, and S. A. Boppart, “Numerical analysis of gradient index lens–based optical coherence tomography imaging probes,” J. Biomed. Opt. 15(6), 066027 (2010). [CrossRef]
40. B. Zhou, Y. Gao, T. Chen, and J. Chen, Laser Principle, 6th ed. (National Defense Industry Press, 2009).
41. Y. Lu and B. Lyu, Matrix Optics (Dalian University of Technology Press, 1989), Chap. 2.
42. F. Träger, Handbook of Lasers and Optics (Springer, 2007), Chap. 3.
43. D. R. Rivera, C. M. Brown, D. G. Ouzounov, W. W. Webb, and C. Xu, “Use of a lensed fiber for a large-field-of-view, high-resolution, fiber-scanning microendoscope,” Opt. Lett. 37(5), 881–883 (2012). [CrossRef]
44. S. Gao, Y. Wang, W. Ding, D. Jiang, S. Gu, X. Zhang, and P. Wang, “Hollow-core conjoined-tube negative-curvature fibre with ultralow loss,” Nat. Commun. 9(1), 2828 (2018). [CrossRef]
