Intraocular pressure estimation using proper orthogonal decomposition

Abstract

Glaucoma is the second leading cause of irreversible blindness. The primary indicator for glaucoma is an elevated intraocular pressure, which is estimated by means of contact or non-contact tonometry. However, these techniques do not accurately account for the cornea properties that deviate from the norm, thus leading to the inaccurate estimation of the intraocular pressure. This work builds on a previous study, in which a combination of an artificial neural network and a genetic algorithm was used to estimate the intraocular pressure and cornea properties. This paper proposes to use proper orthogonal decomposition to accurately estimate the intraocular pressure independent of the cornea properties. The results indicate that proper orthogonal decomposition is able to estimate the intraocular pressure, and that the cornea properties have a slight influence on the estimation. For thicker corneas, however, the intraocular pressure prediction is influenced. This study concluded that this deterministic technique avoids the ambiguity that could result from a method relying on a stochastic optimization routine.