Open Access
Add to BibSonomyAbstract
Three-dimensional confocal microscopy of a living in situ rabbit cornea, in a freshly excised eye, is demonstrated in two movie loops. A specimen chamber for the eye was designed and constructed to maintain the unstained, unfixed, in situ cornea in a viable physiological state during data acquisition. The 400 micron thick, transparent, cornea has been optically sectioned into 365 sections using a laser scanning confocal microscope. A high numerical aperture, water immersion microscope objective minimized the spherical aberrations which would occur with the use of an oil immersion objective. Depth dependent light attenuation due to absorption and scatter within the specimen was manually compensated at each sampled section. Isometric sampling resulted in near-cubic voxels which compensated for the reduced microscopic resolution in the z-axis as compared to x- and y- resolution.
©1998 Optical Society of America
1. Introduction
Three-dimensional reflected light confocal microscopy of a living unfixed, unstained 400 micron thick cornea is an example of the power of microscopic techniques in biomedicine. The three-dimensional visualization of thick, living tissues can be achieved with reflected light confocal microscopy [1–4]. A confocal microscope has enhanced resolution on the z-axis, as compared to a non-confocal light microscope, and this feature results in the ability to optically section thick tissues [1].
The resulting stack of optical sections can then be transformed in a computer into a three-dimensional volume which permits visualization of the stack from viewpoints that differ from the plane containing the individual optical sections [5–12]. These references explain and illustrate a variety of algorithms and three-dimensional visualization techniques for the viewing of the three-dimensional volume that is generated from the stack of optical sections.
The optical properties of the cornea are related to its unique structure [13–15]. The reflected light confocal images are understood in terms of the optical scattering and absorption of the cornea [13].
Optical sections from the full-thickness of the cornea were obtained with reflected light confocal microscopy [16]. Volume rendering computer techniques were used to convert the stack of optical sections from the living cornea into a three-dimensional visualization [17–23].
2. Materials and Methods
The maintenance of the freshly excised, rabbit cornea in a natural, viable, physiological state requires the use of a special chamber to hold the excised eye [24]. This specially designed chamber maintains the mechanical and physiological stability of the ex vivo eye during observation with confocal microscopy. Special care must be exercised during the excision of the eye to prevent mechanical stress to the cornea which could result in changes to its structure [24]. In cases when this special chamber was not used, and the correct excision technique was not employed, then the confocal images of the excised cornea show epithelial and stromal swelling and a time-dependent deterioration of the cornea. These changes are not observed when the correct procedures are followed [24].
The details of specimen preparation, the composition of the bicarbonate Ringer’s solution and a detailed description of confocal image acquisition, Kalman filtering, and the volume rendering techniques have been previously described [23]. Therefore, the technique is only briefly described here. The corneas were observed in situ with a BioRad© MRC-600 laser scanning confocal microscope. The microscope was operated in the back scattered light mode. The laser excitation was a 15 mW argon /krypton ion laser with excitation lines at 488 nm and 568 nm, and 647nm. The combined lines were used for data acquisition. The neutral density filter transmitted one percent of the laser output power. A series of 365 optical sections were acquired in a quarter screen format (384 × 256 pixels per image). Each image was a compilation of 4 separate scanned images averaged using the Kalman algorithm and required a total time of one second per image to acquire. A Nikon Optiphot microscope with an 10x ocular was coupled to the LSCM. The microscope objective was a Leitz 25X water immersion objective with an NA of 0.6. The objective had a free working distance of 1.7 mm. The long working distance of the microscope objective permits the observation of the full thickness of the rabbit cornea which is 400 microns. The fine focus of the LSCM was incriminated in 1.10 micron steps.
The light attenuation effects of observing a 400 micron thick transparent sample were corrected in the following manner. Within the stroma the light attenuation increased with sample depth from the anterior region of the stroma to the posterior regions. To compensate for the problem of light attenuation, the gain for each optical section was manually adjusted to maximize the contrast and the signal to noise ratio.
The data sets were reconstructed using VoxelView Software (Vital Images, Fairfield, IA). The VoxelView software uses a fast implementation of the volume rendering technique [6]. The movie loops contained in this paper were made with the same software. Motion and pseudo color are used to help with the three-dimensional visualization of the living cornea.
3. Results
The three-dimensional reconstruction is pseudocolored to enhance the perception of the internal structures. Red-orange codes for high light scatter; and light blue codes for lower levels of light scattering from the cornea. In the center of the rectangular volume is a highly scattering artifact from internal reflections within the confocal microscope-it is the red-orange cylinder traversing both volumes (a. and b.).
This movie loop begins with a build-up of the full thickness of the cornea. The build-up begins with the single layer of endothelial cells and adds the many layers of keratocyte cells (only their nuclei are shown) and finally adds the basal lamina layer and the corneal epithelium which is at the anterior (tear surface) of the cornea.
The curvature of the cornea is shown by observing the outer limiting surfaces in the reconstruction of the full-thickness of the cornea. The anterior surface consists of the 40 micron thick corneal epithelium. The highly reflecting basal lamina layer is the second highly reflecting layer located 40 microns posterior to the anterior surface of the cornea. The corneal endothelial layer is located on the posterior surface of the cornea. The two limiting cell layers, one each surface are 400 microns apart. Between the basal lamina layer and the posterior endothelial cell layer is a region called the corneal stroma. It contains keratocyte cells and large nerves. One large bifurcating nerve is shown within the middle of the full thickness of the cornea. Since the nerve is highly scattering it is pseudocolored red-orange. The cell bodies of the individual keratocytes are not observed within the full three-dimensional reconstruction. The highly scattering nuclei of the keratocytes are shown within the volume of the corneal stroma.
Fig. 2. Keratocyte nuclei in the posterior stroma (high resolution, partial thickness, 5 microns). This reconstructed volume of the stroma is near the posterior endothelial cell layer of the cornea. [Media 2]
The highly scattering, large (25 micron long), oval nuclei of the stromal keratocyte cells are shown in this movie loop. The oval nuclei of the stromal keratocytes are highly scattering and are pseudocolored red-orange. The cell bodies of the keratocytes are not observed due to their very weak reflectivity. A network of linear structures of one micron diameter are observed within the posterior stroma. It is presumed that these linear structures are within the cell bodies of the keratocytes which are not visible in this volume rendering.
4. Discussion
The three-dimensional volume visualization of the unstained, living rabbit cornea was obtained from a stack of optical sections. Reflected light confocal microscopy was used to acquire the stack of images. The use of a special chamber to maintain the physiological state of the in situ cornea and special nontraumatic surgical techniques to excise the eye are technical requirements to avoid artifacts. In addition, the use of high aperture water immersion microscope objectives, compensation for light attenuation with the 400 micron thick cornea by manual gain adjustments, and isotropic sampling are required. This paper demonstrates the use of reflected light confocal microscopy to obtain a stack of optical sections with submicron resolution (in the plane of the cornea) through a 400 micron thick, unstained, unfixed, living tissue. Computer graphics volume rendering techniques were employed to form a three-dimensional visualization of the living cornea. A current goal is to obtain three-dimensional reflected light confocal microscopy of the living human cornea in vivo [25–26].
Acknowledgments
This work was supported by a grant from NIH EY-06958. The author thanks M. A. Farmer, M. Forster, J. Czégé, and the Biomedical Instrumentation Center at USUHS.
References and links
1. B. R. Masters, Selected Papers on Confocal Microscopy, (SPIE, The International Society for Optical Engineering, Bellingham, WA, 1996).
2. B. R. Masters, “Confocal Microscopy of Ocular Tissue,” in Confocal Microscopy,T. Wilson, ed. (Academic Press, London, 1990), 305–324.
3. B. R. Masters and G. S. Kino, “Confocal Microscopy of the Eye,” in Noninvasive Diagnostic Techniques in Ophthalmology,B. R. Masters ed. (Springer-Verlag, New York, 1990), 152–171. [CrossRef]
4. B. R. Masters, G. Gonnord, and P. Corcuff, “Three-dimensional microscopic biopsy of in vivo human skin: a new technique based on a flexible confocal microscope,” J. Microsc. 185, 329–338 (1997). [CrossRef] [PubMed]
5. H. E. Keller, “Objective lenses for confocal microscopy,” in The Handbook of Biological Confocal Microscopy, 2 ed.,J. B. Pawley, ed. (Plenum Press, New York, 1995), 111–126.
6. M. Levoy, “Display of Surfaces from Volume Data,” PhD dissertation, (University of North Carolina, Chapel Hill, N.C.1989)
7. N. S. White, “Visualization Systems for Multidimensional CLSM Images,” in Handbook of Biological Confocal Microscopy,J. B. Pawley, ed. (Plenum Press, New York1990), 211–254.
8. H. T. M. Van der Voort, G. J. Brankenhoff, and M. W. Baarlslag, “Three-dimensiona visualization methods for confocal microscopy,” J. Microsc. 153, 123–132 (1989). [CrossRef] [PubMed]
9. H. T. M. van der Voort and A. W. M. Smeudlers, “Acquisition, Processing, and Visualization of Three-Dimensional Confocal Images,” in Electronic Light Microscopy,D. Shotton, ed. (Wiley-Liss, New York, 1992), 247–271.
10. A. Kriete, Visualization in Biomedical Microscopies, (VCH, New York, 1992).
11. M. J. Cookson, R. A. Reynolds, and D-C. Abrams, “Three-dimensional visualization,” in Image analysis in histology, (R. Wootton, D. R. Springall, and J. M. Polak, eds. (Cambridge University Press, Cambridge, UK, 1995), 313–335.
12. O. Minet, G. Müller, and J. Beuthan, Selected Papers on Optical Tomography, Fundamentals and Applications in Medicine, (SPIE, The International Society for Optical Engineering, Bellingham, WA, 1998).
13. R. L. McCally and R. A. Farrell, “Light scattering from cornea and corneal transparency,” in Noninvasive Diagnostic Techniques in Ophthalmology, B. R. Masters, ed. (Springer Verlag, New York, 1990), 189–210. [CrossRef]
14. V. Tuchin, Selected Papers on Tissue Optics, Applications in Medical Diagnostics and Therapy, (SPIE, The International Society for Optical Engineering, Bellingham, WA, 1994).
15. M. A. Lemp, P. N. Dilly, and A. Boyde, “Tandem-scanning (confocal) microscopy of the full-thickness cornea,” Cornea 4, 205–209 (1986).
16. B. R. Masters and S. W. Paddock, “In vitro confocal imaging of the rabbit cornea,” J. Microsc. 158, 267–274 (1990). [CrossRef] [PubMed]
17. A. Kriete and B. R. Masters, “Three-dimensional visualization of the living cornea,” MICRO 90, 1, (Royal Microscopical Society, Oxford, 1990), 401–404.
18. B. R. Masters and S. W. Paddock, “Three-dimensional reconstruction of the rabbit cornea by confocal scanning optical microscopy and volume rendering,” Appl. Opt. 29, 3816–3822 (1990). [CrossRef] [PubMed]
19. B. R. Masters, “Two and three-dimensional visualization of the living cornea and ocular lens,” Machine Vision and Applications 4, 227–232 (1991). [CrossRef]
20. G. Q. Xiao, G. S. Kino, and B. R. Masters, “Observation of the rabbit cornea and lens with a new real-time confocal scanning optical microscope,” Scanning, 12,161–166 (1990). [CrossRef]
21. B. R. Masters and G. S. Kino, “Charge-coupled devices for quantitative Nipkow disk real-time scanning confocal microscopy,” in Electronic Light Microscopy: The principles and practice of intensified fluorescence, video-enhanced contrast and confocal scanning optical microscopy,D. Shotton, ed. (Wiley/Liss, Inc., New York, 1993), 315–327.
22. B. R. Masters, “Two- and Three-Dimensional Confocal Microscopy of the Eye,” in Multidimensional Microscopy,P. C. Cheng, T. H. Lin, W. L. Wu, and J. L. Wu, eds. (Springer-Verlag, New York, 1993), 321–338.
23. B. R. Masters and M. A. Farmer, “Three-dimensional confocal microscopy and visualization of the in situ cornea,” Comput. Med. Imag. Graph. 17, 211–219 (1993). [CrossRef]
24. B. R. Masters, “Specimen preparation and chamber for confocal microscopy of the eye,” Scanning Microscopy 7, 645–651 (1993). [PubMed]
25. B. R. Masters and A. A. Thaer, “Real-time scanning slit confocal microscopy of the in vivo human cornea,” Appl. Opt. 33, 695–701 (1994). [CrossRef] [PubMed]
26. B.R. Masters, “Scanning Slit Confocal Microscopy of the in vivo Cornea,” Opt. Eng. 34, 684–692 (1995). [CrossRef]
