Purchase this article with an account.
A. H. Kashani, G. Martin, M. Humayun; Spectral Determination of Retinal Oxygen Content in Dog Retina Using Computerized Tomographic Imaging Spectroscopy. Invest. Ophthalmol. Vis. Sci. 2010;51(13):5768.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To introduce a novel hyperspectral oximetry model for imaging the dog retina with computerized tomographic imaging spectroscopy.
We use a novel hyperspectral camera that attaches to a standard fundus camera and permits simultaneous recording of >50 wavelengths of spectral data from standard fundus images. Images were acquired from dilated eyes of two dogs. A custom-made unsupervised software algorithm was used to calculate retinal oxygen saturation in vessels and tissue. Oxygen saturation was determined based on the optical density for each pixel in vessels and tissue. A least squares method is applied to calculate oxygen saturation from 28 wavelengths of spectral data. Results are displayed as pseudocolor map of relative oxygen saturation in the retinal vessels and tissue.
Our oximetry maps of the dog retina are able to resolve tissue and vessel hemoglobin oxygen saturation. Figure 1 below shows one example of a normal fundus photograph (left panel) and the corresponding pseudocolored tissue oximetry map (right panel). Red represents the highest relative hemoglobin oxygen saturation (within arteries). Blue shades represent the lowest hemoglobin oxygen saturation.
We present an unsupervised oximetry algorithm for analysis of the oxygen context of the dog retina. This algorithm successfully maps relative hemoglobin oxygen saturation in retinal vessels and surrounding tissue. Future experiments will be aimed at quantitative calibration of oximetry values from retinal vessels and tissue.
This PDF is available to Subscribers Only