SyntEyes: A Higher-Order Statistical Eye Model for Healthy Eyes

PURPOSE. Stochastic eye models are a method to generate random biometry data with the variability found in the general population for use in optical calculations. This work improves the accuracy of a previous model by including the higher-order shape parameters of the cornea. METHODS. The right eye biometry of 312 subjects (40.8 +/- 11.0 years of age) were measured with an autorefractometer, a Scheimpflug camera, an optical biometer, and a ray tracing aberrometer. The corneal shape parameters, exported as Zernike coefficients, were converted to eigenvectors for dimensional reduction. The remaining 18 parameters were modeled as a sum of two multivariate Gaussians, from which an unlimited number of synthetic data sets (SyntEyes) were generated. After conversion back to Zernike coefficients, the data were introduced into ray tracing software. RESULTS. The mean values of nearly all SyntEyes parameters were statistically equal to those of the original data (two one-sided t-test, P > 0.05/109, Bonferroni correction). The variability of the SyntEyes parameters was similar to the original data for most important shape parameters and intraocular distances (F-test, P < 0.05/109), but significantly lower for the higher-order shape parameters (F-test, P > 0.05/109). The same was seen for the correlations between higher-order shape parameters. After applying simulated cataract or refractive surgery to the SyntEyes model, a very close resemblance to previously published clinical outcome data was seen. CONCLUSIONS. The SyntEyes model produces synthetic biometry that closely resembles clinically measured data, including the normal biological variations in the general population.