A fluorescence confocal microendoscope requires a high-performance miniature objective. We present a miniature objective comprising four glass lenses and one plastic aspheric lens. The 0.5 NA objective is achromatized in the wavelength range of 488–550 nm, has a field of view (FOV) of 360 μm, and an outer diameter of 2.6 mm. The assembled miniature objective can resolve features separated by as little as 0.78 μm. The imaging quality of the fluorescence confocal microendoscope with the miniature objective is similar to that of a commercial confocal microscope. It can resolve cellular structures such as crypt structures and epithelial cells.
© 2016 Optical Society of America
Cancer is a serious threat to human health. In 2012, there were 14 million new cases of cancer and 8.2 million deaths from cancer worldwide , and cancers of the digestive tract accounted for 30% of those deaths. Early diagnosis is effective in reducing cancer-related mortality and improving the 5-year survival rate. Currently, the gold standard process for diagnosing digestive tract cancers is an initial endoscopic examination, removal of tissue by biopsy, and a histopathological examination of the tissue slices.
Recently, some have used a fluorescence confocal microendoscope as a guide during the biopsy phase of conventional endoscopy because of its high-resolution real-time imaging capability [2,3]. This guidance can reduce the pain level for patients and make conventional biopsies more efficient. Fluorescence confocal microendoscopic imaging can detect regions of possible neoplasia, which then can be removed by biopsy forceps for histopathological examination. The high-resolution and noninvasive imaging capabilities of the fluorescence confocal microendoscope allow it to be used for an “optical biopsy” instead of a conventional biopsy for clinical diagnosis . The subcellular resolution, optical sectioning , and fluorescence real-time imaging of the fluorescence confocal microendoscope can provide imaging results similar to those of a histopathological examination [6,7].
The optical probe enters the human body via the biopsy channel of a conventional endoscope; therefore, it must fit into the biopsy channel. In addition, the optical probe plays a major role in the imaging capabilities of the fluorescence confocal microendoscope . The outer diameter of the optical probe must be <2.8 mm when used with gastrointestinal endoscopes from Olympus. The simplest optical probe for a high-resolution microendoscope (HRME) is the fiber bundle [9–11]. However, the resolution of the HRME (8 μm) is limited by the diameter of the individual fibers in the fiber bundle, making it difficult to resolve a single cell sufficiently. To improve the resolution, a miniature objective is added to the fiber bundle of the optical probe. For the fluorescence confocal microendoscope, the miniature objective should consist of a high numerical aperture (NA) lens system with minimal aberrations and a small outer diameter. Gradient index (GRIN) lenses are commonly used in microendoscopy because of their small outer diameter [12–14], but they are not suitable for fluorescence confocal microendoscopy because it is extremely difficult to correct chromatic aberration owing to how they form an image. However, GRIN lenses combined with a plano-convex lens can correct spherical aberration and improve the NA effectively in monochrome imaging [15,16].
A custom-designed complex lens system is a better choice for fluorescence confocal microendoscopy. However, because the miniature objective must be short and have a small outer diameter, its design is difficult . The goal is for the miniature achromatic objective to have imaging ability similar to that of commercial achromatic objectives, but have fewer lenses. Custom-designed miniature objectives composed of all glass lenses were first used in reflectance confocal endoscopy [18,19], followed by a miniature achromatic objective consisting of a combination of all glass lenses [20,21]. The miniature objective has an outer diameter of 3 mm and is 13 mm long. It comprises nine lenses and has a resolution of 1.8 μm in the wavelength range of 480–660 nm. Because it is difficult to fabricate a glass lens with an outer diameter smaller than 3 mm using the common grinding and polishing method, the flexibility in designing a miniature objective with custom-designed lenses is limited. A system with all glass lenses usually requires more lenses to correct aberration, thus complicating the assembly process.
On the other hand, the single-point diamond-turning (SPDT) technique can be used to fabricate plastic lenses with high surface accuracy, thus allowing greater flexibility in designing the miniature objective. Miniature objectives comprising all plastic lenses have been used in both reflectance and fluorescence confocal endoscopy [22,23]. The selection of plastic materials available for lens-making is limited compared with glass materials, which makes it difficult to design a miniature achromatic objective with plastic lenses. A miniature achromatic objective, which is composed of 6 lenses with four different plastic materials, was achieved with an outer diameter of 2.1 mm [6,23]. The miniature objective provides a resolution of 4.4 μm integrating with the high-resolution microendoscope (HRME) system. The miniature objective was very complex and required superior technology for its fabrication and assembly. The plastic lenses were aligned using V-shaped grooves and sub-millimeter precision balls instead of spacer rings. The high-performance miniature objective was assembled using self-aligning holders fabricated using LIGA (German acronym for lithographie, galvanoformung, abformung, or lithography, electroplating, and molding) technology . Alternatively, aspheric plastic lenses have been combined with glass spherical lenses to reduce the number of optical elements, making it easy to implement the miniature objective, but it designed for only one wavelength [24,25].
So far, the design of an easy-to-fabricate high-NA miniature objective is still challenging. In this paper, we present a five-lens miniature objective that can be easily implemented using common optical fabrication and assembly technology. The miniature achromatic objective comprises four glass lenses and one plastic aspheric lens. The NA of the object space is 0.5 with a working wavelength range of 488–550 nm. With the miniature objective combined with a fluorescence confocal microendoscope, we obtain images of fresh colon tissue similar to those from a commercial confocal microscope. In addition, the fluorescence confocal microendoscope with the miniature objective has the potential to be used for optical biopsy or as a guide for conventional biopsy for clinical diagnosis in the future.
2. Optical design
2.1 Fluorescence confocal microendoscope system overview
Figure 1 is a schematic diagram of the fluorescence confocal microendoscope. Excitation light generated by a 488-nm laser (CYAN 488-50 CDRH, Spectra-Physics, Santa Clara, CA, USA) is expanded by a telescoped beam expander and then passing a dichroic mirror before entering the X-Y scanner. Next, the 4f system relays the excitation light to the back focal plane of the coupling objective, where it is focused on one fiber of the fiber bundle (IGN 08/30, Sumitomo, Tokyo, Japan). The fiber bundle and the miniature objective focus the excitation light onto the tissue space, collect the emitted light, and return it to the coupling objective. Thus, the miniature objective is an important component of the fluorescence confocal microendoscope. This paper focuses on the design of a high-performance miniature objective;a detailed description of the fluorescence confocal microendoscope was provided in .
2.2 Design of the miniature objective
The specifications of the miniature objective with respect to the requirements for the fluorescence confocal microendoscope are listed in Table 1. The miniature objective is designed to operate in the wavelength range of 488–550 nm to accommodate the fluorescent dyes acridine yellow and sodium fluorescein, which are commonly used for fluorescence confocal imaging of gastrointestinal tissue.
The NA of 0.5 for the object space provides a diffraction-limited resolution of <1.0 μm, allowing visualization of one cell. We assumed that the index of refraction for tissue is similar to that of water, so the object space of the miniature objective is designed for immersion in water. To reduce the likelihood of crosstalk between individual fibers of the fiber bundle, the NA of the image space is set to 0.25 to match the 0.3 NA of the fiber bundle; this improves the efficiency of light coupling and collection. A 360-μm field of view (FOV) is magnified 2 × to match the 720-μm active diameter of the fiber bundle. Considering the penetration depth of confocal microscopic imaging in gastrointestinal tissue, the working distance is set at 150 μm . Telecentricity in fiber space ensures uniform efficiency of fiber coupling across the entire active diameter of the fiber bundle. To ensure that the packaged diameter of the miniature objective is 2.6 mm, the outer diameters of all lenses are 2.2 mm, taking assembly processing into account. To achieve diffraction-limited performance, the root-mean-square (RMS) spot size in fiber bundle space should be less than the diffraction-limited Airy disk spot size that is calculated by the formula 2.44λF , where λ is the primary wavelength and F is the f-number. According to Nyquist’s sampling theorem, the modulation transfer function (MTF) value in the image space should be greater than 50% when the spatial frequency of the fiber space is 167 lp/mm. In this case, the lateral resolution of fluorescence confocal microendoscopy is limited by the fiber bundle, not the miniature objective.
After choosing the appropriate structure for the miniature objective, with the right types of lenses of different glass materials for aberration correction, the design process is complete. Our final design is a combination of only four glass lenses and one plastic lens. The lens parameters are listed in Table 2 and the layout of the lens system is displayed in Fig. 2. The optical design software used in this paper was ZEMAX.
To correct the field curvature, the object plane has an aspheric surface, which is acceptable for confocal imaging of thick biological tissue. During the design process, the sag of the object surface (the distance between the vertex and the projection of the edge point of the object surface to the optical axis ) should be <5 μm, which is less than the usual axial resolution of the confocal microendoscope. Element 1, a plano-convex lens to facilitate direct contact with the tissue sample, is made of high-refractive-index glass material that strongly refracts the light beam. To correct chromatic aberration, Element 2 is a doublet lens, i.e., a combination of two lenses with nearly identical refractive indices but significantly different dispersions. Element 3 is a plastic aspheric lens that corrects spherical aberration effectively and Element 4 is a biconvex lens that corrects residual aberration.
Performance analysis results for the miniature objective are shown in Figs. 3, 4, and 5. Figure 3 presents the spot diagrams with diffraction-limited Airy disks for four radial image positions: on-axis, 0.5 field, 0.707 field, and full field. The RMS spot sizes of the four different positions are <2.47 μm, the diffraction-limited Airy disk spot size, which indicates diffraction-limited performance. Figure 4(a) presents the MTF curves of the tangential and sagittal planes of four radial image positions. The MTF contrast at 167 lp/mm is >0.5 for the eight MTF curves. Figure 4(b) shows the chromatic focal shift in the image space; the maximum focal shift is 12.8 μm, which is a slightly greater than the diffraction-limited range of 7.7 μm. The maximum focal shift in tissue space is calculated using the following equation :Eq. (1), ztissue = 4.3 μm, which is less than the usual axial resolution of the fluorescence confocal microendoscope ; therefore, the chromatic focal shift of the miniature objective would not affect the imaging capability of the fluorescence confocal microendoscope.
Figure 5 presents field curvature and distortion plots of fiber space. Converting fiber space to tissue space using Eq. (1), the maximum field curvature and astigmatic split at the edge of the field in Fig. 5(a) are 2.2 μm and 5 μm, respectively. The maximum distortion in Fig. 5(b) is 0.3% at the edge of field, which is acceptable to the human eye.
These performance analysis results show that the miniature objective adequately corrects the three aberrations, and that its performance meets the design requirements.
We performed tolerance analysis to confirm whether the fabrication was successful; the results are given in Table 3. The inverse limit mode was used and the criterion was that the RMS spot size should be smaller than the diffraction-limited Airy disk spot size, 2.47 μm. The tolerance analysis results confirmed the ability to fabricate the miniature objective to specifications. All tolerances were within the usual capabilities of fabrication and assembly. The results of the Monte Carlo analysis, which estimated the RMS spot size of the miniature objective after fabrication and assembly, are given in Table 4. The results show that 50% of the Monte Carlo simulations resulted in an RMS spot size ≤2.43 μm, which is smaller than the diffraction-limited Airy disk spot. This meets the acceptance criterion that 50% of the Monte Carlo simulations must have an RMS spot size less than or equal to that required by the design. The analysis results affirm that the miniature objective could be fabricated and assembled successfully.
3. Fabrication and assembly processes
The fabrication and assembly processes were undertaken by two companies. The Hong Kong Productivity Council fabricated the plastic lenses using SPDT technology and the Shangrao Qineng Optical Instrument Factory fabricated and assembled the glass lenses and lens barrel.
The miniature objective assembly, a cross-sectional view of which is shown in Fig. 6, has only four units besides the lenses: two barrels and two spacers, which can be easily fabricated and assembled with conventional technology. The barrels are made from surgical stainless steel for clinical applications. The hardness of surgical stainless steel makes barrel fabrication difficult, so a barrel should be kept to less than 6 mm in length to ensure the accuracy of the machining. Thus, barrel 1 is divided into two parts that have same outer diameter but different inner diameters to accommodate the different diameters of the lenses. Element 1 is assembled last and fixed in place with epoxy resin glue. Spacers 1 and 2 are oxidized black to reduce reflected light. All glass lenses have an antireflective coating to improve the light transmission. The opening at the end of Barrel 2 is for insertion of the fiber bundle. The photographs in Fig. 7 show a fully assembled miniature objective and the optical probe comprising a miniature objective and fiber bundle. The outer diameter and length of the assembled miniature objective are 2.6 mm and 10.3 mm, respectively, which meet the operational requirements of the commercial gastrointestinal endoscope. The fiber bundle was moved to the back opening of the miniature objective when the optical probe was fixed with epoxy resin glue behind the fluorescence confocal microendoscope when it was used to obtain images of fresh colon tissue in mice.
4. Characterization and testing results
Before integration with the fluorescence confocal microendoscope, we measured the lateral resolution, MTF, and chromatic focal shift of the miniature objective using the custom testing system shown in Fig. 8. The theoretical value of the lateral resolution of a miniature objective, with an NA of 0.5 and λ between 488 and 550 nm, ranges from 0.58 to 0.63 μm. Hence, to test the lateral resolution of our miniature objective, we used the standard 1951 USAF (United States Air Force) hi-resolution target (Edmund Optics, Barrington, NJ, USA) for the object target and a white-light source with frosted glass for uniform illumination. A 0.25 NA commercial objective (Olympus America, Center Valley, NJ, USA) was chosen to match the image NA of the miniature objective. The amplification system, which consisted of the commercial objective and a tube lens (f = 100 mm), relayed the image of the resolution target to the CMOS camera (UI-1495LE, IDS Imaging Development Systems GmbH, Obersulm, Germany). The result of the lateral resolution test is shown in Fig. 9(a). The enlarged figure shows that the miniature objective resolved the smallest elements on the resolution target, i.e., the group 9, element3 bars that have a spatial frequency of 645 lp/mm. Thus, we determined that the miniature objective can resolve features separated by 0.78 μm, which is close to the theoretical lateral resolution.
The imaging capability of the miniature objective was further analyzed by calculating the MTF curve using the slanted-edge method and the MATLAB program . The MATLAB program extracted eight MTF curves from the vertical and horizontal edges of four different positions of the resolution target image shown in Fig. 9(a), while referring to the ISO 12233 standard . The vertical and horizontal MTF curves were averaged and compared with the diffraction limit MTF curve, as seen in Fig. 9(b), which shows that the MTF at 167 lp/mm is >0.5 and meets the design requirements.
The chromatic focal shift of the miniature objective was tested by placing a series of 10-nm-wide band-pass filters (Thorlabs, Newton, NJ, USA) between the white-light source and the resolution target and measuring the focal shift in object space by changing the filters and moving the resolution target. When both the object space and the image space of the miniature objective are in air, the focal shift in the image space is calculated using the following equation:Fig. 10. The measured results match the predicted chromatic focal shift shown in Fig. 3(b), illustrating that the miniature objective is achromatic in the wavelength range of 488–550 nm.
5. Imaging results of the fluorescence confocal microendoscope
5.1 Lateral resolution test
The lateral resolution of the fluorescence confocal microendoscope was tested before and after integration with the miniature objective. The test sample was the 1951 USAF hi-resolution target with one drop of acridine yellow solution. The results of the test are shown in Fig. 11. Without the miniature objective, the smallest resolvable features are group 7, element 3 bars, which correspond to a spatial frequency of 161.3 lp/mm, approximately the fiber space frequency. With the miniature objective, the smallest resolvable features are group 8, element 3 bars, which correspond to a spatial frequency of 322.5 lp/mm, twice the fiber space frequency, indicating that the fluorescence confocal microendoscope can resolve features separated by 1.55 μm.
5.2 Tissue imaging
Fresh colon tissue was used to test the imaging capability of the fluorescence confocal microendoscope with the integrated miniature objective. C57BL/6 mice were sacrificed and dissected to obtain fresh colon tissue, which then was stained with the acridine yellow solution (0.1%) 1 min before washing with phosphate-buffered saline (PBS) solution. The tissue was imaged with and without the miniature objective by the fluorescence confocal microendoscope; and for comparison purposes, the fresh colon tissue was imaged by a commercial confocal microscope (FV1000, Olympus, 10X / NA 0.4 objective). The imaging results are shown in Fig. 12.
We showed that the lateral resolution of the fluorescence confocal microendoscope improved after integrating the miniature objective. The crypt structures, indicated by the red arrows in Fig. 12, can be clearly seen without [Fig. 12(a)] and with [Fig. 12(b)] the miniature objective, but epithelial cells, indicated by the white arrows, are seen only when the miniature objective is in place [Fig. 12(b)]. The image acquired by the fluorescence confocal microendoscope with the integrated miniature objective was similar to that acquired by the commercial confocal microscope [Fig. 12(c)]. One limitation to these comparisons is that the fresh colon tissue samples used for Figs. 12(a)–12(c) were from three different mice because colon tissue stays fresh for only 10 min.
In this article, we presented our five-lens miniature objective, designed for use with a fluorescence confocal microendoscope. The miniature objective is achromatic in the 488–550-nm wavelength range across the whole FOV of 360 μm and has an NA of 0.5. It resolved the smallest element in the 1951 USAF hi-resolution target. The images obtained by the fluorescence confocal microendoscope integrated with the miniature objective show that it can be used for optical biopsy or as a guide in conventional biopsy for cellular visualization. The miniature objective improves the resolution of the fluorescence confocal microendoscope effectively, enabling recognition of the morphology and size of cells, which is important for early cancer diagnosis [34–36].
The simple structure of the miniature objective can facilitate the development of fluorescence confocal microendoscopy. With improved optical material and fabrication and assembly processes [37,38], the NA of the miniature objective can be increased to improve resolution, and chromatic aberration can be easily corrected for more dyes. There is the potential that a smaller and more easily implemented miniature objective with greater imaging capability can be realized in the future.
This work was funded by the National Natural Science Foundation of China (Grant Nos. 61178077 and 61205197) and the Science Fund for Creative Research Group of China (Grant No. 61421064).
References and links
1. B. W. Stewart and C. P. Wild, World Cancer Report 2014, Lyon, France: International Agency for Research on Cancer (World Health Organization, 2014).
3. P. L. Hsiung, J. Hardy, S. Friedland, R. Soetikno, C. B. Du, A. P. Wu, P. Sahbaie, J. M. Crawford, A. W. Lowe, C. H. Contag, and T. D. Wang, “Detection of colonic dysplasia in vivo using a targeted heptapeptide and confocal microendoscopy,” Nat. Med. 14(4), 454–458 (2008). [CrossRef] [PubMed]
5. N. Chen, S. Rehman, and C. J. R. Sheppard, “Advanced optical microscopy methods for in vivo imaging of sub-cellular structures in thick biological tissues,” J. Innov. Opt. Health Sci. 07(05), 1440001 (2014). [CrossRef]
7. S. M. Landau, C. Liang, R. T. Kester, T. S. Tkaczyk, and M. R. Descour, “Design and evaluation of an ultra-slim objective for in-vivo deep optical biopsy,” Opt. Express 18(5), 4758–4775 (2010). [CrossRef] [PubMed]
10. T. J. Muldoon, M. C. Pierce, D. L. Nida, M. D. Williams, A. Gillenwater, and R. Richards-Kortum, “Subcellular-resolution molecular imaging within living tissue by fiber microendoscopy,” Opt. Express 15(25), 16413–16423 (2007). [CrossRef] [PubMed]
11. M. Pierce, D. Yu, and R. Richards-Kortum, “High-resolution fiber-optic microendoscopy for in situ cellular imaging,” J. Vis. Exp. 47(47), 2306 (2011). [PubMed]
12. J. Knittel, L. Schnieder, G. Buess, B. Messerschmidt, and T. Possner, “Endoscope-compatible confocal microscope using a gradient index-lens system,” Opt. Commun. 188(5–6), 267–273 (2001). [CrossRef]
14. 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] [PubMed]
15. B. Messerschmidt, A. Kraeplin, S. Schenkl, I. Riemann, M. Stark, A. Ehlers, A. Tchernook, R. Le Harzic, and K. Koenig, “Novel concept of GRIN optical systems for high resolution microendoscopy. Part 1: Physical aspects,” Proc. SPIE 6432, 643202 (2007). [CrossRef]
17. C. Liang, Design of Miniature Microscope Objective Optics for Biomedical Imaging (PhD thesis, 2002).
18. C. Liang, K.-B. Sung, R. R. Richards-Kortum, and M. R. Descour, “Design of a high-numerical-aperture miniature microscope objective for an endoscopic fiber confocal reflectance microscope,” Appl. Opt. 41(22), 4603–4610 (2002). [CrossRef] [PubMed]
19. A. Osdoit, M. Genet, A. Perchant, S. Loiseau, B. Abrat, and F. Lacombe, “In vivo fibered confocal reflectance imaging: totally non-invasive morphological cellular imaging brought to the endoscopist,” Proc. SPIE 6082, 608208 (2006). [CrossRef]
20. A. R. Rouse, A. Kano, J. A. Udovich, S. M. Kroto, and A. F. Gmitro, “Design and demonstration of a miniature catheter for a confocal microendoscope,” Appl. Opt. 43(31), 5763–5771 (2004). [CrossRef] [PubMed]
21. T. Y. Wu, A. F. Gmitro, and A. R. Rouse, “An achromatized miniature objective for microendoscopy,” in International Optical Design Conference, 2014 OSA Technical Digest Series (Optical Society of America, 2014), paper IW4A.2. [CrossRef]
22. M. D. Chidley, K. D. Carlson, R. R. Richards-Kortum, and M. R. Descour, “Design, assembly, and optical bench testing of a high-numerical-aperture miniature injection-molded objective for fiber-optic confocal reflectance microscopy,” Appl. Opt. 45(11), 2545–2554 (2006). [CrossRef] [PubMed]
23. M. Kyrish, J. Dobbs, S. Jain, X. Wang, D. Yu, R. Richards-Kortum, and T. S. Tkaczyk, “Needle-based fluorescence endomicroscopy via structured illumination with a plastic, achromatic objective,” J. Biomed. Opt. 18(9), 096003 (2013). [CrossRef] [PubMed]
25. R. T. Kester, T. S. Tkaczyk, M. R. Descour, T. Christenson, and R. Richards-Kortum, “High numerical aperture microendoscope objective for a fiber confocal reflectance microscope,” Opt. Express 15(5), 2409–2420 (2007). [CrossRef] [PubMed]
26. J. Wang, M. Yang, L. Yang, Y. Zhang, J. Yuan, Q. Liu, X. H. Hou, and L. Fu, “A confocal endoscope for cellular imaging,” Engineering 1(3), 351–360 (2015). [CrossRef]
27. A. A. Tanbakuchi, A Surgical Confocal Microlaparoscope for Real-Time Optical Biopsies (PhD thesis, 2009).
28. W. J. Smith, Modern Optical Engineering, 4th ed. (McGraw-Hill, 2008)
29. J. M. Geary, Introduction to Lens Design: with Practical ZEMAX (Willmann-Bell, 2002).
30. A. R. Rouse, Multi-Spectral Confocal Microendoscope for In-Vivo Imaging (PhD thesis, 2004).
31. Y. S. Sabharwal, A. R. Rouse, L. Donaldson, M. F. Hopkins, and A. F. Gmitro, “Slit-scanning confocal microendoscope for high-resolution in vivo imaging,” Appl. Opt. 38(34), 7133–7144 (1999). [CrossRef] [PubMed]
32. P. D. Burns, “Slanted-edge MTF for digital camera and scanner analysis,” in Is and Ts Pics Conference (Society for Imaging Science & Technology, 2000), pp. 135–138.
33. ISO, “Photography—electronic still-picture cameras—resolution measurements,” ISO 12233:2000, http://www.iso.org/iso/home/store/catalogue_tc/catalogue_detail.htm?csnumber=33715.
34. R. Cicchi and F. S. Pavone, “Multimodal nonlinear microscopy: A powerful label-free method for supporting standard diagnostics on biological tissues,” J. Innov. Opt. Health Sci. 07(05), 1330008 (2014). [CrossRef]
35. J. N. Zhang, Y. Q. Li, Y. A. Zhao, T. Yu, J. P. Zhang, Y. T. Guo, and H. Liu, “Classification of gastric pit patterns by confocal endomicroscopy,” Gastrointest. Endosc. 67(6), 843–853 (2008). [CrossRef] [PubMed]
36. Y. Kakeji, S. Yamaguchi, D. Yoshida, K. Tanoue, M. Ueda, A. Masunari, T. Utsunomiya, M. Imamura, H. Honda, Y. Maehara, and M. Hashizume, “Development and assessment of morphologic criteria for diagnosing gastric cancer using confocal endomicroscopy: an ex vivo and in vivo study,” Endoscopy 38(9), 886–890 (2006). [CrossRef] [PubMed]
37. G. Lan, T. F. Mauger, and G. Li, “Design of high-performance adaptive objective lens with large optical depth scanning range for ultrabroad near infrared microscopic imaging,” Biomed. Opt. Express 6(9), 3362–3377 (2015). [CrossRef] [PubMed]
38. L. Guoqiang, “Adaptive lens,” Prog. Opt. 55, 199–283 (2010). [CrossRef]