Purchase this article with an account.
HA Elzembely, J Reynaud, RW Beuerman; Confocal Stereoscopic Imaging of the Rabbit Cornea . Invest. Ophthalmol. Vis. Sci. 2002;43(13):1715.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To determine the feasibility of using a conventional white-light confocal microscope to produce stereoscopic images of the rabbit cornea. Method: Initial imaging trials tested stationary objects such as cotton and paper fibers, grapes, and tomatoes to eliminate specimen movement as a variable. A computer-controlled stepper motor provided precise angular positioning of the microscope with accuracy of 0.05 degrees. After developing the stereo imaging procedures, all corneal layers of anesthetized New Zealand white rabbits were imaged. To obtain stereo pairs, the focal axis of rotation was located by tracking a landmark while rotating the microscope 6 degrees from center in 1-degree increments and precisely adjusting the focal plane. The axis of rotation was set when the landmark remained stationary at all angles. A composite of the left and right image (492 x 369µm or 640 x 480 pixels) was then produced by placing the images 2.5 inches apart using Adobe Photoshop. Stereo pairs were then viewed using stereoscopic glasses. Results: Initial stereo imaging procedures were developed and 6 degrees proved optimal for stereopsis. Within the cells of the grapes and tomatoes, positioning of organelles within the cells could be observed. In the rabbit cornea, stereo pairs of the epithelium and stroma were readily obtained. The depth relationship between stromal nerves and processes of adjacent keratocytes was visualized. Images of the endothelium did not reveal much additional depth information, most likely due to the 12µm depth of field and the single cell layer. Conclusion: Early results show that the confocal microscope can be used to obtain stereoscopic images of living tissue. This may be particularly useful in spatial localization of corneal foreign bodies and structural abnormalities of corneal dystrophies. The process requires further refinement and efficient imaging algorithms must be developed to reduce errors induced by slight tissue movements during imaging.
This PDF is available to Subscribers Only