The trade-off between lateral resolution and depth of focus (DOF) severely limits the capability of endoscopic optical coherence tomography (OCT) for high-resolution deep-tissue imaging. To address this issue, we developed a novel miniature all-fiber axicon OCT probe by inserting a segment of gradient-index (GRIN) fiber between a piece of single-mode fiber (SMF) and an axicon polished from a no-core fiber. The GRIN lens served as a beam expander extending the probe DOF by 5.2 times while maintaining a high lateral resolution of 2 μm. The DOF extension was experimentally verified by measuring the axial profile of the probe output beam and further by imaging multilayered polymer tapes and onion samples. The designed probe with a tight focus over a large DOF holds great potential in endoscopic OCT imaging of deep tissues at the cellular level.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Optical coherence tomography (OCT)  is a powerful biomedical imaging modality capable of non-invasively acquiring cross-sectional images of tissue microstructures in high spatial resolution. Due to the shallow penetration depth (< 3 mm) of light in tissue limited by optical scattering, it is difficult for OCT to directly image organs inside human . This limitation can be alleviated by introducing a flexible, compact imaging catheter or endoscopic probe into the organs. The endoscopic OCT systems have successfully in vivo visualized suspicious lesion tissues below the epithelial surfaces of internal organs such as gastrointestinal tracts  and pulmonary airways . Nevertheless, the axial (~10 μm) and lateral (~30 μm) resolutions of these endoscopic OCT systems are insufficient to resolve cellular and subcellular features of biological tissues [5–7]. The axial resolution can be improved by employing broadband supercontinuum light sources or pulsed lasers with low coherent lengths [8,9]. However, to improve the lateral resolution, using a conventional optical lens with a large numerical aperture would severely reduce the DOF that is inversely proportional to the square of the numerical aperture [10,11]. Various solutions such as using axicon lenses [12,13], phase spatial filters [14,15], and annular apodization [16,17] have been proposed to obtain high lateral resolution without significantly compromising the DOF.
Another attractive method to achieve high-resolution and large-DOF biomedical imaging was to use Bessel beam instead of Gaussian beam as the illumination light [18–20]. Bessel beam features a diffraction-free nature that allows it to maintain an invariable spot size as it propagates over a long distance . It can be readily generated by using an axicon, one of the most basic optical components with a cone shape [22,23]. Ding et al. have first used a free-space axicon to achieve both high lateral resolution and large DOF in the OCT . Unfortunately, this scheme cannot be employed in endoscopic OCT because the bulky axicon prevented compact integration of the whole sample arm optics into a small-diameter scanning probe. An alternative approach for Bessel beam generation is to utilize optical fibers, which are attractive for endoscopic applications considering their compactness, free of alignment and remote delivery capabilities. Using a fiber lens to focus light with an annular field  or chemical etching of the optical fiber tip into an axicon [19,20] were developed to generate Bessel beams. Although chemical-etching treatment is frequently used to modify the fiber structure, it required accurate control of the parameters including the concentration of etching solution, the ambient temperature, the etching time, and the fiber-core doping. Moreover, axicons produced by chemical etching provided the OCT probe a limited DOF improvement compared to a bulk axicon.
In this work, we report on the development of a miniature all-fiber axicon probe for forward-viewing OCT imaging with both enlarged DOF and high lateral resolution. The probe was fabricated by first splicing a segment of gradient-index (GRIN) fiber between a single-mode fiber (SMF) and a no-core fiber (NCF), and then mechanically polishing the NCF to produce an axicon at the tip of the fiber assembly. No sophisticated selective-chemical etching procedure was involved. The GRIN fiber with a 1/4 pitch length acted as a beam expander allowing a greatly increased DOF without compromise on the lateral resolution [19,20]. This improvement of DOF was experimentally verified by measuring the axial profile of the output beam from the probe and also by imaging a multilayered polymer tape and an onion sample. The superior performance of miniature all-fiber axicon probe is promising to improve OCT diagnostic results in clinical applications.
2. Principle, probe fabrication and performance test
For OCT imaging, the DOF and spot size of the illumination beam should be customized for specific applications. Theoretically, the DOF of the Bessel beam formed by an axicon lens is approximately equal to its maximum non-diffractive propagation distance, and can be expressed as :Fig. 1. ω0 is the beam waist illuminating on the axicon lens. The central spot radius RBB can be calculated from the first zero position of the zero-order Bessel function, and can be expressed as follows :
From Eqs. (1) and (2), it can be observed that a larger input beam waist ω0 corresponds to a longer depth of focus LDOF as well as an unaffected central spot radius RBB of the Bessel beam. Also, a small RBB requires a large refraction angle β, which reduces the depth of focus LDOF. Therefore, the parameters of the both the GRIN and axicon lens could be adjusted to optimize the depth of focus LDOF and the central spot radius RBB. As an example, by setting parameters α = 30°, ω0 = 39 μm, λ = 1.31 μm, and n = 1.4468 of the fabricated EDOF axicon probe into Eqs. (1) and (2), the depth of focus LDOF and the central spot radius RBB are theoretically estimated to be ~110 μm and ~1.8 μm, respectively. The LDOF is nearly 8.6 times larger than the 12.8 μm for a conventional SMF axicon probe with ω0 = 4.6 μm and the same central spot radius RBB.
2.2 Probe fabrication
Figure 1(a) shows the schematic of an EDOF axicon probe. The distal focusing optics consists of short sections of GRIN fiber (GI2016-F, YOFC, China) and NCF (FG125LA, Thorlabs, USA) directly spliced to the end of a piece of SMF (SMF-28, Corning, USA). The GRIN fiber and NCF segments have lengths of 371 μm (1/4 pitch length) and 50 μm, respectively. The NCF end was molded into an axicon lens with a conical angle of 120 degrees using a commercial fiber polisher (Ultrapol Fiber Lensing Machine, ULTRATEC Manufacturing Inc, USA). The monolithic all-fiber probe was encapsulated into a stainless hypodermic tube and fixed via a UV-curing optical adhesive (NOA81, Norland, USA). Figure 1(b) shows a typical microscope image of an EDOF axicon probe. The GRIN fiber of 1/4 pitch length served as a beam expander to increase the area of light filling into the axicon lens , and allowed the probe to obtain a larger DOF than conventional SMF axicon probes with a core diameter of 8.2 µm according to Eqs. (1) and (2) [19,20].
The detailed process of polishing the NCF tip to an axicon lens is similar to that for manufacturing near-field optical microscope probes . The pre-fabricated lensed fiber was fixed to a fiber chuck which was connected to an electrical motor via a belt, hence the lensed fiber could be rotated around its axis. The entire system was mounted over a rotating platform and the tilt angle with respect to the platform could be adjusted flexibly. During the polishing process, the lensed fiber was set slightly in contact with the uniformly spinning abrasive disk while it was slowly rotating along its axis, so that a symmetrical circular cone could be obtained. The distance between the fiber tip and the abrasive disk was precisely controlled by using an optical microscope. To obtain a fiber axicon with a very smooth surface in a short polishing time, the fiber tip has been firstly coarsely polished into a cone shape by using a grinding paper with a diamond grain size of 3 μm and then fine-grinded with a 1 μm grain size paper. The microscope images of the polished EDOF axicon probe and its endface are shown in Figs. 1(b) and 1(c), where a sharp and smooth cone at the probe tip are observed. For comparative study, a traditional SMF axicon probe without the GRIN lens was manufactured using the same technique, as shown in Fig. 1(d).
2.3 Performance test
To evaluate the quality of the output beam from the fabricated probes, we measured the beam profile using a CCD beam profiler (WinCamD, DataRay Inc.). The beam emitted from the probe was expanded by a 25 × objective lens before received by the beam profiler [27,28]. An SLD light source (S5FC1018S, Thorlabs Inc.), with a bandwidth of 45 nm at the center wavelength of 1310 nm, optically fed the fabricated probes. The probe was placed on 5-axis stage to adjust perpendicularly to the CCD. The profiles of the output beam were captured at different locations where we moved the probe along the optical axis over the focal region of interest in 5 μm steps. Figure 2(a) shows the result of the beam profile at the focal plane of the EDOF axicon probe. A main lob locating at the beam center surrounded by a series of concentric rings is a typical pattern of a Bessel beam. Figures 2(b) and 2(c) plot the normalized intensity distributions at the x and y axis corresponding to Fig. 2(a), respectively. The measured beam diameters are 2 µm and 1.96 µm in the x-axis and y-axis respectively, which indicated a negligible astigmatism ratio of 1.02 (the ratio of the beam diameter of the x-axis to that of the y-axis) caused by the imperfect fabrication process . The energy of the sidelobes in the x-axis direction and y-axis direction is about 18.9% and 16.6% of the peak energy, respectively. These sidelobes were relatively weak and introduced no obvious artifacts into the image in our experiment .
Figure 3 shows the measured full-width at half-maximum (FWHM) beam diameter of the EDOF and SMF axicon probes at different depth. The focal plane was set as the zero-position along the depth direction. Here, the lateral resolution is defined as FWHM beam diameter of the focused spot and the DOF refers to a regime over which the FWHM beam diameter is smaller than twice its value at focus . Compared with the SMF axicon probe (the DOF is ~30 μm), the EDOF axicon probe provided a 5.2-fold better DOF of ~155 μm at the cost of a slightly reduced signal-to-noise ratio (SNR) which could be compensated by using a higher input power laser. The probe had a lateral resolution of ~2 μm, which was very close to the theoretically predicted value (~1.8 μm). This good agreement confirmed that the high quality of the axicon probes fabricated by the mechanical polishing process. We also simulated the evolution of the lateral resolution with the depth for a Gaussian beam, and a 11.3 times improvement of the DOF was achieved by using the EDOF axicon probe.
The collection efficiency in angular orientation was obtained by measuring the back-reflected signal power from a mirror mounted on a high-precision rotator (PR01, Thorlabs, USA). The angle between the optical axis of probe and the normal of the mirror was tuned by 2 degrees per step. A barely cleaved fiber as well as a SMF axicon probe were tested as references using the same experimental process. Figure 4 plots the normalized collection efficiency of a cleaved fiber, a SMF axicon probe, and an EDOF axicon probe. The collection efficiencies of both the SMF and EDOF axicon probes were less sensitive to the tilt angle than that of the simply cleaved fiber counterpart, due to the multi-focal characteristics of the axicon probes . The better collection efficiency of the SMF axicon probe than that of the EDOF axicon probe may be caused by the shorter distance between the center of the no-diffraction zone to the probe tip surface.
3. Experimental setup and results
3.1 OCT imaging system design
The experimental setup of the swept source OCT (SS-OCT) system is shown in Fig. 5. We employed a Fourier domain mode-locking (FDML)  swept laser (FDML-1310-4B-APC, Optores GmbH) with an A-line scanning rate of ~1.6 MHz as the OCT light source. The light source provided an output power up to 160 mW and a spectral scanning range of ~110 nm with a central wavelength of 1315 nm. In the system, output light signal from the swept source was divided by a 90:10 fiber coupler with 10% power routing to a reference arm while the remaining 90% output power was sent into a sample arm or a recalibration arm which could be selected by an optical switch. The OCT imaging signal is obtained via a Michelson interferometer comprising a reference arm and a sample arm. Another interferometer including a reference arm and a recalibration arm was used to resample the OCT imaging signal so that this signal was uniform in the k-space. The back-scattered sample signal and reference signal, after travelling through a 50:50 fiber coupler, were detected by a balanced photodetector (PDB480C-AC, Thorlabs Inc.) with a 1.6 GHz bandwidth and digitized by an oscilloscope (DPO70404B, Tektronix, Inc.) at 3.13 GS/s. The designed probe was attached to a precise linear translation stage (M-L01.8A1, Physik Instrumente Co., Ltd.) for transverse mechanical scanning. The stage operation and data acquisition/processing were performed by custom designed software program written in a LabVIEW program.
3.2 System performance
To characterize the probe performance with the SS-OCT system, the probe was fixed on a three-dimensional linear translation stage, and a metallic mirror was employed as the sample which was placed in front of the imaging probe. Then, we adjusted the spatial orientation of the probe and aligned it to maximize the recoupling power into the probe. In order to prevent detector saturation, an additional loss of 47.4 dB was introduced in the sample arm by loosening the FC/APC connector. Due to the short working distance of the designed probe, it was not feasible here to insert a neutral density filter into the sample arm as widely used . The axial resolution of the system was acquired from the FWHM of the linear point spread function (PSF). The result in Fig. 6(a) manifests that the linear PSF at an imaging depth of 0.4 mm has a 11.7 μm FWHM in air, corresponding to axial resolution of 8.5 µm in tissue (assuming the refractive index of the tissue n = 1.37). The measured axial resolution was lower than the theoretical value, which was about 7 µm for a light source with the sweep bandwidth of 110 nm. The discrepancy may be caused by the dispersion mismatch between two arms of interferometer and the non-Gaussian spectrum shape of the light source .
The relationship between the sensitivity of the probe-based system and the imaging depth was studied by measuring the PSFs of a mirror located at different axial positions in the reference arm as shown in Fig. 6(b). The sensitivity reduced by 6 dB at a depth of 2 mm, caused by the limited coherence length of the laser source . The sensitivity was calculated by adding an additional loss of 47.4 dB to the measured SNR. The SNR can be expressed as SSNR = 10log10 (Max I 2 /σ2), where I is the linear magnitude of the OCT image and σ2 is the variance of I in a background noise region . The measured SNR at the 0.4 mm imaging depth was 45 dB, corresponding to the system sensitivity of 92.4 dB. The lateral resolution of the systems was ~2 μm, determined by the spot size focused onto the sample.
3.3 OCT imaging test
The EDOF axicon probe with an extended DOF was then used for OCT imaging of a phantom made by stacking multi-layer translucent tape on a microscope slide. The thickness of the translucent tape was ~50 μm. We carefully adjusted the axial distance between the probe tips and the sample surface to eliminate effects of the depth-dependent sensitivity of the SS-OCT system. The same dynamic range was used for log compression of the images in Fig. 7. The phantom was imaged by scanning the fiber probe laterally in a 2 μm step using a precise linear translation stage. Images of the multi-layer translucent tape obtained via the EDOF axicon probe and a conventional SMF axicon probe are shown in Figs. 7(a) and 7(b), respectively. We can clearly observe 10 layers translucent tape in Fig. 7(a), while only 6 layers in Fig. 7(b). Consequently, the EDOF probe have proven a significantly increased DOF. We also carried out a cross sectional OCT imaging of an onion tissue using SS-OCT system with the EDOF axicon probe. As the EDOF axicon had a short working distance, the onion sample was pressed to flatten its surface before imaging. The image is displayed in logarithmic gray scale as shown in Fig. 8, and the image size is 140 × 500 pixels corresponding to an area of 1.2 × 1 mm. Tiny structure of onion cell can be identified up to 500 μm deep. The slight blurring of the onion structure may be caused by the damage of the internal structure of the onion during the pressing process. To verify the high later resolution of our OCT imaging system, instead of onion with large cell size, samples such as microsphere/agar mixture will be used as image sample in the future. It also needs to point out that the short working distance of the probe might limit its capability for endoscope applications. Considering that the currently used fibers have diameters as small as 125 μm, employment of fibers with larger diameters can effectively increase the probe working distance without severe compromise of the probe compactness.
To conclude, a miniature all-fiber axicon probe with an extended Bessel focus of depth for forward-viewing OCT imaging has been developed in this work. We first theoretically proved that the DOF of the Bessel beam can be effectively expanded by increasing the diameter of incident light field on the axicon lens without sacrificing the spot size. And then, we experimentally introduced a 1/4 pitch GRIN fiber into the designed fiber probe to generate a Bessel beam with a 155 μm DOF and a 2 μm FWHM beam diameter. This design effectively alleviated the mutual restriction between DOF and lateral resolution of traditional optical lenses. Finally, we imaged a multi-layer translucent tape and an onion sample to verify the improved DOF. The designed probe described here, is promising to improve the capability of OCT for high-resolution deep-tissue imaging in clinical applications.
National Natural Science Foundation of China (U1701268, 61705082).
1. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef] [PubMed]
2. D. C. Adams, Y. Wang, L. P. Hariri, and M. J. Suter, “Advances in endoscopic optical coherence tomography catheter designs,” IEEE J. Sel. Top. Quantum Electron. 22(3), 1–12 (2016). [CrossRef]
3. A. M. Rollins, R. Ung-Arunyawee, A. Chak, R. C. Wong, K. Kobayashi, M. V. Sivak Jr., and J. A. Izatt, “Real-time in vivo imaging of human gastrointestinal ultrastructure by use of endoscopic optical coherence tomography with a novel efficient interferometer design,” Opt. Lett. 24(19), 1358–1360 (1999). [CrossRef] [PubMed]
4. S. Lam, B. Standish, C. Baldwin, A. McWilliams, J. leRiche, A. Gazdar, A. I. Vitkin, V. Yang, N. Ikeda, and C. MacAulay, “In vivo optical coherence tomography imaging of preinvasive bronchial lesions,” Clin. Cancer Res. 14(7), 2006–2011 (2008). [CrossRef] [PubMed]
5. 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] [PubMed]
6. L. Liu, J. A. Gardecki, S. K. Nadkarni, J. D. Toussaint, Y. Yagi, B. E. Bouma, and G. J. Tearney, “Imaging the subcellular structure of human coronary atherosclerosis using micro-optical coherence tomography,” Nat. Med. 17(8), 1010–1014 (2011). [CrossRef] [PubMed]
8. B. Bouma, G. J. Tearney, S. A. Boppart, M. R. Hee, M. E. Brezinski, and J. G. Fujimoto, “High-resolution optical coherence tomographic imaging using a mode-locked Ti:Al(2)O(3) laser source,” Opt. Lett. 20(13), 1486–1488 (1995). [CrossRef] [PubMed]
9. W. Yuan, J. Mavadia-Shukla, J. Xi, W. Liang, X. Yu, S. Yu, and X. Li, “Optimal operational conditions for supercontinuum-based ultrahigh-resolution endoscopic OCT imaging,” Opt. Lett. 41(2), 250–253 (2016). [CrossRef] [PubMed]
10. W. Drexler, M. Liu, A. Kumar, T. Kamali, A. Unterhuber, and R. A. Leitgeb, “Optical coherence tomography today: speed, contrast, and multimodality,” J. Biomed. Opt. 19(7), 071412 (2014). [CrossRef] [PubMed]
12. Z. Ding, H. Ren, Y. Zhao, J. S. Nelson, and Z. Chen, “High-resolution optical coherence tomography over a large depth range with an axicon lens,” Opt. Lett. 27(4), 243–245 (2002). [CrossRef] [PubMed]
13. K. S. Lee and J. P. Rolland, “Bessel beam spectral-domain high-resolution optical coherence tomography with micro-optic axicon providing extended focusing range,” Opt. Lett. 33(15), 1696–1698 (2008). [CrossRef] [PubMed]
14. L. Liu, C. Liu, W. C. Howe, C. J. R. Sheppard, and N. Chen, “Binary-phase spatial filter for real-time swept-source optical coherence microscopy,” Opt. Lett. 32(16), 2375–2377 (2007). [CrossRef] [PubMed]
15. J. Kim, J. Xing, H. S. Nam, J. W. Song, J. W. Kim, and H. Yoo, “Endoscopic micro-optical coherence tomography with extended depth of focus using a binary phase spatial filter,” Opt. Lett. 42(3), 379–382 (2017). [CrossRef] [PubMed]
16. K. K. Chu, C. Unglert, T. N. Ford, D. Cui, R. W. Carruth, K. Singh, L. Liu, S. E. Birket, G. M. Solomon, S. M. Rowe, and G. J. Tearney, “In vivo imaging of airway cilia and mucus clearance with micro-optical coherence tomography,” Biomed. Opt. Express 7(7), 2494–2505 (2016). [CrossRef] [PubMed]
17. Y. Luo, D. Cui, X. Yu, E. Bo, X. Wang, N. Wang, C. S. Braganza, S. Chen, X. Liu, Q. Xiong, S. Chen, S. Chen, and L. Liu, “Endomicroscopic optical coherence tomography for cellular resolution imaging of gastrointestinal tracts,” J. Biophotonics 11(4), e201700141 (2018). [CrossRef] [PubMed]
18. P. Dufour, M. Piché, Y. De Koninck, and N. McCarthy, “Two-photon excitation fluorescence microscopy with a high depth of field using an axicon,” Appl. Opt. 45(36), 9246–9252 (2006). [PubMed]
19. K. M. Tan, M. Mazilu, T. H. Chow, W. M. Lee, K. Taguichi, B. K. Ng, W. Sibbett, C. S. Herrington, C. T. A. Brown, and K. Dholakia, “In-fiber common-path optical coherence tomography using a conical-tip fiber,” Opt. Express 17(4), 2375–2384 (2009). [CrossRef] [PubMed]
20. J. Kim, J. Han, and J. Jeong, “Common-Path Optical Coherence Tomography Using a Conical-Frustum-Tip Fiber Probe,” IEEE J. Sel. Top. Quantum Electron. 20, 6800407 (2014).
22. G. Scott and N. McArdle, “Efficient generation of nearly diffraction-free beams using an axicon,” Opt. Eng. 31(12), 2640–2643 (1992). [CrossRef]
23. J. H. McLeod, “The axicon: a new type of optical element,” J. Opt. Soc. Am. 44(8), 592–597 (1954). [CrossRef]
24. J. Kim, Y. Jeong, S. Lee, W. Ha, J. S. Shin, and K. Oh, “Fourier optics along a hybrid optical fiber for Bessel-like beam generation and its applications in multiple-particle trapping,” Opt. Lett. 37(4), 623–625 (2012). [CrossRef] [PubMed]
26. T. Held, S. Emonin, O. Marti, and O. Hollricher, “Method to produce high-resolution scanning near-field optical microscope probes by beveling optical fibers,” Rev. Sci. Instrum. 71(8), 3118–3122 (2000). [CrossRef]
28. K. M. Tan, M. Shishkov, A. Chee, M. B. Applegate, B. E. Bouma, and M. J. Suter, “Flexible transbronchial optical frequency domain imaging smart needle for biopsy guidance,” Biomed. Opt. Express 3(8), 1947–1954 (2012). [CrossRef] [PubMed]
29. L. Scolaro, D. Lorenser, R. A. McLaughlin, B. C. Quirk, R. W. Kirk, and D. D. Sampson, “High-sensitivity anastigmatic imaging needle for optical coherence tomography,” Opt. Lett. 37(24), 5247–5249 (2012). [CrossRef] [PubMed]
30. Y. Ryu, Y. Shin, D. Lee, J. Y. Altarejos, E. Chung, and H.-S. Kwon, “Lensed fiber-optic probe design for efficient photon collection in scattering media,” Biomed. Opt. Express 6(1), 191–210 (2014). [CrossRef] [PubMed]
31. R. Huber, M. Wojtkowski, and J. G. Fujimoto, “Fourier Domain Mode Locking (FDML): A new laser operating regime and applications for optical coherence tomography,” Opt. Express 14(8), 3225–3237 (2006). [CrossRef] [PubMed]
34. T.-H. Tsai, H.-C. Lee, O. O. Ahsen, K. Liang, M. G. Giacomelli, B. M. Potsaid, Y. K. Tao, V. Jayaraman, M. Figueiredo, Q. Huang, A. E. Cable, J. Fujimoto, and H. Mashimo, “Ultrahigh speed endoscopic optical coherence tomography for gastroenterology,” Biomed. Opt. Express 5(12), 4387–4404 (2014). [CrossRef] [PubMed]
35. D. C. Adler, T. H. Ko, and J. G. Fujimoto, “Speckle reduction in optical coherence tomography images by use of a spatially adaptive wavelet filter,” Opt. Lett. 29(24), 2878–2880 (2004). [CrossRef] [PubMed]