Corneal topography reinterpretation through separate analysis of the projected rings

Abstract

We present a new algorithm to process captured images of reflected Placido rings. Up to our knowledge, conventional topographers transform from Cartesian to polar coordinates and vice-versa, thus extrapolating corneal data and introducing noise and image artefacts. Moreover, captured data are processed by the device according to proprietary algorithms and offering a final map of corneal curvature. Corneal topography images consists of concentric rings of approximately elliptical shape. Our proposal consists of considering the information that provides each separate ring. A snake-annealing-like method permits identifying the ring even with discontinuities due to eye-lashes and reflections. By analysing the geometrical parameters of rings (centre, semi-axis and orientation), one can obtain information about small morphological micro-fluctuations and local astigmatisms. These parameters can be obtained with sub-pixel accuracy so the method results of high precision. The method can be easily adapted to work on any topographer, so that it can provide additional information about the cornea at no additional cost.