Open Access
Add to BibSonomyAbstract
We describe a new method to generate thin (thickness > 200 nm) and ultrathin (thickness < 200 nm) fluorescent layers to be used for microscope optical characterization. These layers are obtained by ultramicrotomy sectioning of fluorescent acrylic slides. This technique generates sub-resolution sheets with high fluorescence emission and uniform thickness, permitting to determine the z-response of different optical sectioning systems. Compared to the state of the art, the here proposed technique allows shorter and easier manufacturing procedure. Moreover, these fluorescent layers can be employed without protective coverslips, allowing the use of the Sectioned Imaging Property (SIP)-chart characterization method with coverslip-uncorrected objectives, water immersion objectives and micro-endoscopes.
© 2014 Optical Society of America
1. Introduction
Maintaining and optimizing optical microscopy set-ups requires powerful characterization methods which precisely quantify the microscope optical performances. Two fundamental parameters of any microscope are the spatial resolution and the dimension of the field of view (FOV). The first is the ability of the microscope to resolve the fine details of the observed sample, and it is quantified as the minimum distance necessary to distinguish two neighboring object points [1]; the latter is the extent of the sample area that can be visualized and its dimension is limited by the apertures along the optical path [2]. In general, the point spread function (PSF), which is defined as the irradiance distribution that results from a single point source in the object space, is used to determine the spatial resolution [1–4]. However, the PSF gives an estimate of the microscope resolution in a precise position within the FOV and it is known that resolution may vary across the FOV. In particular, in lateral parts of the FOV the microscope resolution may be reduced because of lower signal-to-noise ratio (SNR), or because of off-axis optical aberrations. These limitations are particularly important in optical devices with constrained lateral size (and thus FOV) as, for example, micro-endoscopes. For these latter devices the FOV is sometimes indicated [5] but a fast and quantitative method to evaluate how the resolution varies across the FOV is currently missing.
A useful figure of merit to describe the resolution of microscopes is the full width at half maximum (FWHM) of the image-plane intensity profiles of sub-resolved objects [3,4,6]. In the case of sectioning microscopy, a detailed characterization of the resolution along the axial direction over the field of view is necessary to fully describe the performances of the optical system. As described in Eqs. (1) for multi-photon microscopy, the resolution in the x-y plane of a scanning microscope is tightly related to the resolution along the z axis [7]. In these formulas, FWHMxy is the FWHM of lateral PSF, FWHMz is the FWHM of axial PSF function, n is the effective refractive index of the specimen, NA the numerical aperture of the objective and λ the light wavelength.
These equations are derived under the basic assumption that the illumination light at the objective input pupil has a distortion-free wavefront and its intensity is uniform in the lateral direction. It is important to underline, however, that these conditions are not always met; the intensity distribution of incident light may be non-uniform and may have a truncated Gaussian distribution, which significantly affects the performances of the optical system.
A powerful method to characterize the axial resolution and other image properties in sectioning microscopy is the use of Sectioned Imaging Properties (SIP) charts [8]. Utilizing a thin, sub-resolution, fluorescent layer, the SIP chart allows the simultaneous visualization of the z resolution and of the illumination intensity distribution for the optical system over the whole FOV. The use of a sub-resolution fluorescent layer to characterize the z-response of an optical system was proposed by Schrader et al. [9], but their samples exhibited limited uniformity. Zwier et al. [10] overcame this limitation by using polyvinylalcohol as a host material for fluorescein dyes. With this strategy, they reported uniform layers with thickness of 150-200 nm. These thickness values allow direct axial resolution measurements with error below 4-5%, only for systems with FWHMz larger than about 800 nm [8]. The samples developed by Zwier et al. [10] were used by Brakenhoff et al. [8] to propose the SIP chart as a standard method to characterize image properties in sectioning microscopy. A different technique for the preparation of ultrathin layers was later developed by Vicidomini et al. [11], following the layer-by-layer method presented in [12]. The Vicidomini et al. approach is particularly useful for monitoring the axial response of very high-resolution systems, like 4Pi microscopes. More recently, Model and Blank [13] demonstrated the use of a liquid concentrated solution of fluorescent dyes to create thin layers sealed in between two cover glasses.
A drawback of all the samples manufactured with the previously described methods is that they cannot be used for the characterization of many common microscope objectives. Indeed, the coverslip, that is necessary to enclose the fluorescent dye, allows a proper characterization only for objectives corrected for the optical aberration introduced by this glass window. Water immersion objectives and micro-endoscopes [14], which are generally used for applications involving living cells and tissues, cannot be properly characterized with this method because of the presence of the coverslip. Removal of the cover glass in the samples described previously [10,11,13] would require developing a different chemistry for the fluorescent dye deposition, because the fluorescent dyes used in those studies are based on water soluble components.
We here report a new procedure to manufacture thin (> 200 nm) and ultrathin (< 200 nm) highly fluorescent layers by ultramicrotomy sectioning of thick fluorescent acrylic slides. This technique permits rapid manufacturing of samples with different thicknesses by using common equipment for electron microscopy sample preparation, with a procedure significantly easier than those reported previously. These films, made of acrylic plastic, are stable when immersed in water and we report effective characterization of imaging conditions in optical sectioning microscopy with water immersion objectives and micro-endoscopes using one- and two-photon excitation with this method. Moreover, the availability of fluorescent plastic with different excitation and emission spectra, permits the fabrication of layers with different spectral properties which is useful to visualize the chromatic properties of the optical system and to characterize different laser lines in confocal microscopy. The absence of cover glass in the methodology that we are proposing allows the use of these fluorescent slides also for the direct characterization of dry objectives without coverslip correction and with very short working distances.
In section 2 we give an overview of the method which has been adopted for characterization; section 3 is dedicated to the description of the fluorescent layer fabrication; in section 4, we report use of the proposed method with objectives of different types, including water immersion coverslip-uncorrected objectives and micro-endoscopes. Section 5 is dedicated to the discussion of the experimental results and to the conclusions.
2. Characterization method
SIP charts are a powerful method to characterize image properties in sectioning microscopy [8]. Following this method, a chart, collecting a large amount of information on the possible lateral-shift-variant behavior of axial resolution, is created by the analysis of the through-focus data set obtained during sectioning microscopy of a thin, sub-resolution, fluorescent layer. The chart contains the following characterizing parameters: ISum, Imax, Zmax, FWHMz, and Skew. Isum is the integrated intensity along the z-axis. In an ideal case, a uniform fluorescent layer gives a constant ISum distribution. However, in real cases, uneven illumination and detection generates non-uniform intensity of collected light. Imax is the projection onto one plane of the maximal intensity that is observed along the z-axis sectioning. In an optical system with small aberrations, the maximal intensity distribution is a good figure of merit of the optical performances over the FOV [1]. In a system with uniform optical performance, the Imax distribution equals the ISum distribution. It is important to remind that the Imax representation is not affected by possible inclination of the fluorescent layer with respect to the plane normal to the light propagation direction (x-y plane), because it collects the maximal values along the sectioning process. Zmax indicates the axial position where the maximal intensity is detected; so, it gives information on the inclination of the sample with respect to the x-y plane and on the field curvature distortion of the optics. The FWHMz distribution gives the full-width-half-maximum resolution along z-axis over the FOV; this map is obtained by plotting of the z-stack intensity distribution for each pixel in the FOV and subsequently measuring the FWHM. Finally, Skew describes the axial asymmetry of the z-response, and can reveal the presence of optical aberrations (e.g. spherical aberration). The axial intensity profiles of 5 different points on the FOV are also reported.
This chart can be generated from collected 3D image data sets using the freely available plug-in SIPchart.jar [15] in the Fiji (ImageJ) software. The plug-in resizes the original images into a 64 × 64 images. Since the performances of the optical system does not usually vary abruptly over the FOV, this binning procedure smoothes down point-like defects in the optics or in the fluorescent sample.
3. Materials and methods
In order to use the SIP-chart method for characterization of optical systems over their entire FOV, three properties for the ultrathin fluorescent samples are of fundamental importance: i) the sample thickness should be small with respect to the axial resolution of the microscope optics; ii) emissivity of the sample should be high enough to ensure good SNR; iii) the sample should be uniform, in term of thickness, fluorescent emission and bleaching properties.
The here presented fabrication process consist in using ultramicrotomy techniques to cut thin and ultrathin films (50-500 nm) from thick fluorescent plastic slides, which are widely used to check the uniformity of illumination of fluorescence imaging systems. The ultramicrotome is a common equipment for the preparation of ultrathin films in electron microscopy that allows cutting slices of the sample with precise control of their thicknesses between 20 nm and 2500 nm. The wide range of thicknesses that can be achieved with the ultramicrotome permits to tailor the sample according to the specific optics that needs to be characterized. This flexibility is important because, for example, while it is important to have ultrathin layers (50 nm) to allow the effective characterization of high NA optics, such ultrathin layers may be not optimal (or may have insufficient emissivity) when used with low NA objectives because of their lower fluorescence collection. In these latter cases, having the ability to generate thicker (but still sub-resolved) samples is crucial. With our method, the layer thickness can be easily tailored to be > 200 nm for low numerical aperture optics (NA < 0.5), and thinner (< 200 nm) for high NA objectives (NA ≥ 0.5).
Samples with thickness between 50 nm and 500 nm were prepared by cutting fluorescent plastic reference slides (Chroma Technology Corporation, Olching, Germany) with green excitation (FITC/GFP) and with red excitation (Rhodamine/Texas Red) spectra using an ultramicrotome EM UC6 (Leica Microsystem, Wetzlar, Germany) equipped with a Pelco SU diamond knife (cutting angle: 35°). Ultrathin sample area was approximately 550 x 550 μm2. The samples were first cut in de-ionized water. A few drops of water with the ultrathin samples floating on their surfaces were then placed on an anti-adherent surface on a Teflon foil (DuPont, Wilmington, DE, USA). The samples were left still until water completely evaporated to allow the deposition of the ultrathin layers onto the Teflon foil. The surface of the ultrathin samples on the opposite side of the Teflon foil was glued to a coverslip using UV adhesive NOA63 (Norland Products Inc., Cranbury, NJ, USA) with low fluorescent emission and refraction index of 1.56. The Teflon foil was finally removed, leaving one surface of the thin layers exposed to air. This configuration allows the use of the ultrathin layers for direct characterization of water immersion objectives, objectives uncorrected for aberration introduced by coverslips and micro-endoscopes. If required, ultrathin plastic films can also be immersed in oil and covered with specific coverslips.
We measured the thickness of some samples using the XP-2 profilometer (Ambios Technology Inc, Santa Cruz, CA) and analyzed its surface with the Atomic Force Microscope (AFM). Figure 1 shows the profilometer measurement of a 100 nm thick sample (sample dimension in the x-y plane: 550 x 550 μm2). Slice thickness was constant over the whole sample.
Fig. 1 Profilometer measurement of a yellow/green fluorescent layer which was cut at a nominal thickness of 100 nm.
We measured the uniformity of the fluorescence emitted by the ultrathin samples by slightly modifying the procedure indicated in [10]. We acquired two image stacks in two different portions of the ultrathin slice (each portion is shifted in the x-y plane with respect to the other one). We generated two images corresponding to the maximal intensity recorded in each stack at each focal point. One image was divided over the other and the histogram graph of the resulting ratio image was plotted (Fig. 2). Fiji software [16] was used for this calculation. This procedure corrects potential errors introduced by the ultrathin sample inclination during the x-y shift. Figure 2(c) shows the histogram graph for a water dipping Olympus LUMPLFLN 60XW objective (Olympus Corp. Tokyo, Japan), a ultrathin layer (thickness: 70 nm), and two-photon excitation (λ = 920 nm) provided by a Chameleon laser (Coherent Inc., Santa Clara, CA, USA). In this case, the imaged field of view was of 156 x 156 µm2. Under ideal conditions, this histogram graph should be formed by a narrow distribution with an average value of 1. The histogram graph displayed in Fig. 2 has a mean value of 0.999, a standard deviation of 0.02 and a FWHM of 0.04. We repeated this measurement in different regions of the ultrathin layer and always obtained histogram distributions with FWHM < 0.05.
Fig. 2 Uniformity test of layer’s fluorescence; a) and b) are maximal intensity distributions emitted over the FOV by one fluorescent layer (thickness: 70 nm), in two different regions; c) represents the histogram distribution obtained by dividing the intensity maps displayed in a) and b).
To test the bleaching property of the ultrathin samples, we focused a pulsed laser beam (repetition rate, 80 MHz; nominal pulse duration, 140 fs; wavelength, 920 nm) for a dwell time of 60 s and measured the emitted fluorescence with a photomultiplier tube (PMT).
We repeated this measurement on several points across the fluorescent layer (layer dimension: 550 × 550 µm2) keeping the laser intensity constant. As shown in Fig. 3, the temporal profile of the light intensity collected by the PMT has the same decay for all the randomly tested points.
Fig. 3 Fluorescence intensity decay due to photo-bleaching in a yellow-green layer (thickness: 70 nm). Bleaching is induced by two-photon excitation (λ = 920 nm) on 6 different points which were randomly located on the thin layer.
4. Characterization of coverslip-uncorrected objectives, water immersion objectives and micro-endoscopes
The ultrathin fluorescent layers were used to test different objectives, an Olympus MPFLN 100X (NA 0.9, WD 1 mm, dry objective, uncorrected for coverslip), an Olympus LUMPLFLN 60XW (NA 0.9, WD 2 mm, water dipping objective), and two micro-endoscopes: GT-MO-080-018-810 (NA 0.8, WD 0.2 mm, 1.4 mm diameter, water immersion) and a NEM-050-25-10-860-S (NA 0.5, WD 0.2 mm, 0.5 mm diameter, water immersion), both by Grintech (Grintech Gmbh, Jena, Germany). All the objectives and micro-endoscopes were tested with two-photon excitation provided by a mode-locked Ti:Sapphire laser source (Ultra II Chameleon, Coherent Inc., Santa Clara, CA, USA), a commercial Prairie Ultima IV scanhead (Prairie Inc., Madison, WI) and an upright epi-fluorescence microscope (BX61 Olympus Corp. Tokyo, Japan). In addition, the MPFLN 100X was also tested with an inverted confocal microscope Nikon Confocal A1 (Nikon Corp., Tokyo, Japan) and single-photon excitation.
We employed ultrathin layers of thickness 100 nm and 200 nm obtained by cutting a plastic slide with excitation/emission spectra similar to those of FITC. The thickness used for these samples was below the theoretical FWHMz resolution of the abovementioned objectives at the employed wavelengths. This measurement thus provides a good estimate of the axial resolution of the objective [6,17]. We also report the z-response of the Olympus MPFLN 100X objective with a 100 nm ultrathin layer with TRITC emission/excitation.
4.1 Two-photon measurements
Figure 4 shows the SIP chart obtained for the Olympus MPLFLN 100X (NA 0.9, WD 1 mm, dry, uncorrected objective) mounted onto the two-photon microscope, by using a green-yellow fluorescent layer (thickness: 100 nm). The z-step for the through-focus scanning data set was 0.25 µm, producing data oversampling for the 3D reconstruction.
Fig. 4 SIP chart for the Olympus MPLFLN 100X objective over a field of view of 95 × 95µm2, obtained by using a 100 nm-thick fluorescent layer and two-photon excitation (λ = 920nm).
A x-z image of the collected fluorescence intensity is reported in Fig. 5 (top). In the central region of the FOV, the z-response FWHM is obtained by fitting the axial intensity profile with a Gaussian curve, as represented in Fig. 5 (bottom). The elaboration displayed in Fig. 5 allows the simultaneous visualization of the axial resolution along a linear portion of the FOV. This analysis was obtained using Fiji software [16], cropping a thin portion of the field of view, re-slicing it to get its x-z projection and averaging the intensity on the selected field. The fit with a Gaussian curve can be immediately performed using the freely available ImageJ Plug-in MetroloJ [18]. The objective properties appear to be quite constant over the whole field of view of the system (95 x 95 µm2). The minor unevenness along x direction of the intensity distributions (Imax and ISum) was most likely due to a slightly asymmetric incident beam. This was verified by direct observation of the beam wavefront with a Shack-Hartmann wavefront sensor (model WFS150-5C, Thorlabs, Newton, NJ). The FWHM resolution had a slight drift along the y-direction, consistently with the minor shift of the maximal intensity in the Imax map. The specimen was also slightly inclined (slope ~1/100) along the x-axis, as it results visible from the Zmax graph.
Fig. 5 Collected fluorescence intensity z-profile and fitting parameters for the Olympus MPLFLN 100X objective. Fitted FWHMz = 1.79 μm. Theoretical FWHM z resolution was 1.12 µm.
Using the same optical system and within the same measurement session, we recorded a 3D image stack of sub-resolved fluorescent beads (diameter, 200 nm polystyrene, yellow dyed spherical beads, Polybeads by Polyscience Inc., Eppelheim, Germany). We measured FWHMz = 1.55 ± 0.1 µm (mean ± standard deviation; n = 8 beads), following the procedure reported in [3]. As reported in [6,17,19], it is important to note that, when using a sub-resolution fluorescent layer, the z-response FWHM results approximately 9% larger than the axial PSF FWHM obtained by using a fluorescent sub-resolution bead (which is described by Eqs. (1)). This additional measurement confirms the effectiveness of the method that we are proposing to characterize axial resolution.
In Fig. 6, we show the SIP chart for the water dipping objective Olympus LUMPLFLN 60XW over a 200 × 200 µm2 FOV. The z-step was 0.25 µm. The measured axial FWHM resolution was 2.8 µm. The discrepancy between the measured axial FWHM resolution and the theoretical limit (1.8 µm) is most likely to be ascribed to laser beam apodization at the entrance pupil of the objective and to slight wavefront deformations due to local distortions along the optical path of the laser beam.
Fig. 6 SIP chart and z profile of Olympus LUMPLFLN 60XW over a field of view of 200 × 200 µm2, obtained using an ultrathin layer (thickness:100 nm) and two-photon excitation by pulsed laser at 920 nm. FWHMz = 2.8 μm (theoretical FWHMz resolution was 1.8 µm).
We then used the ultrathin layers to characterize two commercially available micro-endoscopes, with different optical and mechanical properties. Both micro-endoscopes were optically coupled with an Olympus LMPLFLN-BD 50X objective (NA 0.5, WD 10.6 mm, dry objective, uncorrected for coverslip) at their image side. Figure 7 shows the SIP chart and x-z intensity map of the GT-MO-080-018-810 for a FOV of 38 × 38 µm2. The measured FHWM was of about 3.1 µm, consistently with what reported in literature [14,20] for micro-endoscopes of similar optical characteristics. The FWHM increases rapidly when moving from the optical axis toward the outer part of the endoscope FOV, accordingly with the characteristics delivered by the producer. The fluorescent layer used for this characterization was 100 nm-thick.
Fig. 7 SIP chart and z profile of Grintech micro-endoscope GT-MO-080-018-810 over a 38 × 38 µm2 FOV, obtained by using a 100 nm-thick layer and two-photon excitation by a pulsed laser at 920 nm.
Figure 8 shows the SIP chart of the micro-endoscope NEM-050-06-00-520-S over a FOV of 175 × 175 µm2. This endoscope consists of a singlet (0.94 pitch) of gradient refractive index (GRIN) rod lens (diameter, 0.5 mm; length, 1.86 mm). The best obtained z-resolution was 11 µm. In this case, we used a thicker fluorescent layer (thickness: 200 nm), in order to enhance the fluorescent response.
Fig. 8 SIP chart and z profile of Grintech micro-endoscope NEM-050-06-00-520-S over a 175 × 175 µm2 FOV, obtained by using a 200 nm-thick layer and two-photon excitation (λ = 920 nm).
4.2 Single-photon measurements
We also tested the effectiveness of our method by using single-photon excitation microscopy, employing two ultrathin (thickness: 100 nm) fluorescent layers made of different materials, with spectra similar to those of the commonly used fluorophores FITC/GFP (green) and Rhodamine /Texas Red (red) excitation, respectively. We set a pinhole aperture of 1 Airy unit in the confocal microscope. With the adopted configuration, the theoretical axial FWHM for a sub-resolution fluorescent layer of negligible thickness can be obtained using Eq. (2) [6,17].
In Fig. 9 we show the SIP chart over a FOV of 125 × 125 µm2 and the fit of the intensity along the z-axis for the green fluorescent layer using the Olympus MPFLN 100X objective. The measured axial FWHM for this layer was of about 0.86 µm, in good agreement with the theoretical value of 0.82 µm calculated using Eq. (2).
Fig. 9 SIP chart and z profile of Olympus MPFLN 100X over a 125 × 125 µm2 FOV, obtained by using a 100 nm-thick yellow/green layer excited in an inverted Nikon Confocal A1 microscope. Excitation beam wavelength: 488 nm.
In Fig. 10, we show the results obtained with the same sectioning optical system for the measurement of a red, ultrathin (thickness:100 nm) layer. Compared to the previous measurement the FWHM is larger (0.97 µm), accordingly with the shift of excitation/emission towards longer wavelengths, which leads to a theoretical FWHMz of about 0.93 µm.
Fig. 10 SIP chart and z profile of Olympus MPFLN 100X over a 125 × 125 µm2 FOV, obtained using a 100 nm-thick red layer excited an inverted Nikon Confocal A1 microscope. Excitation beam wavelength: 550 nm.
5. Conclusions
In this study we present a novel method to produce thin and ultrathin fluorescent layers that can be used to characterize the properties of sectioning optical microscopes. The main advantages of this methodology are: i) layers are coverslip-free and can thus be used to characterize virtually any microscope objective, including cover glass-uncorrected objectives, water immersion objectives and micro-endoscopes; ii) samples can be produced with various thicknesses (range 20 – 2500 nm). This allows to customize the sample thickness according to the NA aperture of the optics that needs to be characterized. For example, thicker layers that emit stronger fluorescence signals can be used with low NA optics; iii) samples are cheap, easy to produce and stable over time; iv) layers have good uniformity and suffer from limited photobleaching; v) layer with different excitation/emission spectra can be easily produced with the same method; vi) samples for the characterization of coverslip-corrected objectives can be easily fabricated with an appropriate glass coverslip on top.
The use of thin fluorescent layers is an easy and effective way to obtain sub-resolution samples for the characterization and calibration of sectioning microscopy systems and optical components through the use of SIP charts [8]. Moreover, these samples can be efficiently used for image calibration in fluorescence microscopy [10]. This graphical representation describes in one shot some fundamental characteristics of the analyzed optics over the whole field of view. In spite of the high effectiveness of this approach, the restricted availability of thin uniform and fluorescent layers limited its diffusion. Moreover, previously published methods for fabrication of reference layers confined the proper use of the SIP chart analysis to system equipped with coverslip-corrected objectives [9–11,13]. The novel manufacturing method that we describe in this study allows extending the SIP chart approach to objectives that are not corrected for coverslip aberrations, to water immersion objectives and to micro-endoscopic probes. Micro-endoscopes, in particular, have usually short working distances and thus a precise and fast characterization of the optical performances of these probes is, most of the time, difficult. The methodology described in this manuscript is an efficient solution to these limitations. Moreover, the wide range of thicknesses allowed by ultramicrotome cutting procedure together with the high and uniform fluorescent emission of the produced samples permits to manufacture ultrathin layers (down to 20 nm) for characterization of high resolution optics as well as thicker layers (200 nm −1000 nm) for the analysis of optical systems/components with low numerical apertures. Finally, the possibility to use fluorescent plastic slides with different excitation/emission spectra allows to estimate chromatic effects of the optics under evaluation. Based on the aforementioned characteristics, we propose this methodology as a powerful and flexible method to probe the optical properties of optical microscopes with virtually any objective type.
Acknowledgments
We thank Roberta Ruffilli for her contribution to the preparation of the fluorescent layers, Mattia Pesce for his assistance in confocal measurements, Claudio Moretti for his help with two-photon measurements and Giuseppe Vicidomini for useful discussion. This work was supported by the Italian Institute of Technology “Interdisciplinary/Interdepartmental project” to TF and CL and grants from San Paolo “Programma in Neuroscienze”, MIUR FIRB (RBAP11X42L), Telethon-Italy (GGP10138) and a FP7 DESIRE project to TF.
References and links
1. M. Born and E. Wolf, Principles of Optics, 7th ed. (Cambridge University Press, 1999).
2. M. Abramowitz, Microscope Basics and Beyond (Olympus America Inc., 2003), Vol.1.
3. R. W. Cole, T. Jinadasa, and C. M. Brown, “Measuring and interpreting point spread functions to determine confocal microscope resolution and ensure quality control,” Nat. Protoc. 12, 1929–1941 (2011).
4. A. R. Hibbs, G. MacDonald, and K. Garsha, “Practical confocal microscopy,” in Handbook of Biological Confocal Microscopy, 3rd ed., J.B. Pawley, ed. (Springer, 2006), Chap. 36.
5. R. P. Barretto and M. J. Schnitzer, “In vivo optical microendoscopy for imaging cells lying deep within live tissue,” Cold Spring Harb Protoc 2012(10), 1029–1034 (2012). [CrossRef] [PubMed]
6. S. Wilhelm, B. Gröbler, M. Gluch, and H. Heinz, “Confocal Laser Scanning Microscopy” (Zeiss Jena Inc.). http://zeiss-campus.magnet.fsu.edu/referencelibrary/laserconfocal.html.
7. 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] [PubMed]
8. G. J. Brakenhoff, G. W. H. Wurpel, K. Jalink, L. Oomen, L. Brocks, and J. M. Zwier, “Characterization of sectioning fluorescence microscopy with thin uniform fluorescent layers: Sectioned Imaging Property or SIPcharts,” J. Microsc. 219(3), 122–132 (2005). [CrossRef] [PubMed]
9. M. Schrader, U. G. Hofmann, and S. W. Hell, “Ultrathin fluorescent layers for monitoring the axial resolution in confocal and two-photon fluorescence microscopy,” J. Microsc. 191(2), 135–140 (1998b). [CrossRef] [PubMed]
10. J. M. Zwier, G. J. Van Rooij, J. W. Hofstraat, and G. J. Brakenhoff, “Image calibration in fluorescence microscopy,” J. Microsc. 216(1), 15–24 (2004). [CrossRef] [PubMed]
11. G. Vicidomini, M. Schneider, P. Bianchini, S. Krol, T. Szellas, and A. Diaspro, “Characterization of uniform ultrathin layer for z-response measurements in three-dimensional section fluorescence microscopy,” J. Microsc. 225(1), 88–95 (2007). [CrossRef] [PubMed]
12. G. Decher, “Fuzzy nanoassemblies: Toward layered polymeric multicomposites,” Science 277(5330), 1232–1237 (1997). [CrossRef]
13. M. A. Model and J. L. Blank, “Concentrated dyes as a source of two-dimensional fluorescent field for characterization of a confocal microscope,” J. Microsc. 229(1), 12–16 (2008). [CrossRef] [PubMed]
14. R. P. J. Barretto, B. Messerschmidt, and M. J. Schnitzer, “In vivo fluorescence imaging with high-resolution microlenses,” Nat. Methods 6(7), 511–512 (2009). [CrossRef] [PubMed]
15. N. Vischer, M. Savenije, and G. J. Brakenhoff, “SIPchart plugin for ImageJ,” University of Amsterdam (2014) http://simon.bio.uva.nl/sipcharts/SIPchart.html.
16. C. A. Schneider, W. S. Rasband, and K. W. Eliceiri, “NIH Image to ImageJ: 25 years of image analysis,” Nat. Methods 9(7), 671–675 (2012). [CrossRef] [PubMed]
17. B. Amos, G. McConnell, and T. Wilson, “Confocal microscopy,” in Handbook of Comprehensive Biophysics (Elsevier, 2012).
18. C. Matthews and F. P. Cordelieres, “œMetroloJ: an ImageJ plugin to help monitor microscopes' health,” in Proceedings of ImageJ User & Developer Conference (2010).€
19. S. Hell, G. Reiner, C. Cremer, and E. H. K. Stelzer, “Aberrations in confocal fluorescence microscopy induced by mismatches in refractive index,” J. Microsc. 169(3), 391–405 (1993). [CrossRef]
20. C. Wang and N. Ji, “Characterization and improvement of three-dimensional imaging performance of GRIN-lens-based two-photon fluorescence endomicroscopes with adaptive optics,” Opt. Express 21(22), 27142–27154 (2013). [CrossRef] [PubMed]
