Dual wavelength Fundus Oximetry (DWFO) is a useful tool for tissue oximetry and monitoring hyperoxia and hypoxia. The accuracy of retinal oximetry is affected by multiple factors. A recent numerical simulation study showed that the absolute measured SO2 error could be up to 20% when the vessel size is as large as 160µm or the melanin concentration in the retinal pigment epithelium is as high as 8micromoles/liter. We tested the hypothesis of oxygen exchange with retinal tissue using the Wang’s model to improve the accuracy of oxygen saturation calculations in retinal oximetry.

The accuracy of So2 measurement was validated in a bio-mimicking phantom, which was 3D, printed based on a schematic eye model. The phantom consisted of transparent capillaries of 160 μm inner diameters with the ability to include varied So2 levels. The fundus images were obtained by a recently developed DWFO. Dual wavelength fundus oximetry, a simple approach to oximetry, uses the approximately linear relationship between SO2 and the ratio of measured optical density at an isobestic waveband to the optical-density at an SO2-sensitive waveband. To the best of our knowledge this is the first study that has used the Wang’s model for improving the accuracy of oxygen saturation calculations in retinal oximetry.

The average standard deviation for repeated measurements of Optical Density Ratio (ODR) for individual capillaries sections was 2.27%. In each case, standard deviations were calculated for two images of each subject, and then were averaged across all subjects. ODR of arterial SO2 in capillaries was 0.90 ± 0.08 (mean ± standard deviation). The difference with the control group value was statistically significant (p<0.05).

We demonstrate that the repeatability of ODR measurements was found to be 0.5% (standard deviation of 10 repeated consecutive measurements in a single sample of 100% SO2 blood). This is much less than the change in ODR seen due to deoxygenation (~ 25%). There was however, some variability in ODR measurements compared to blood gas analyser measurements which may be due to inhomogeneities in the optical properties of the blood sample.