Study and characterization of morphogeometric parameters to assist diagnosis of keratoconus

Francisco Cavas-Martínez,Jorge Alió,Dolores Parras,Francisco J F Cañavate,Laurent Bataille,Daniel G Fernández-Pacheco

doi:10.1186/s12938-018-0564-7

Abstract

BackgroundIn case of significant imperfections on the cornea, data acquisition is difficult and a significant level of missing data could require the interpolation of important areas of the cornea, resulting in a very ambiguous model. The development of methods to define in vivo customised geometric properties of the cornea based only on real raw data is extremely useful to diagnose and assess the progression of diseases directly related to the corneal architecture. The present work tries to improve the prognostic of corneal ectasia creating a 3D customised model of the cornea and analysing different geometric variables from this model to determine which variables or combination of them could be defined as an indicator of susceptibility to develop keratoconus.MethodsA corneal geometric reconstruction was performed using zonal functions and retrospective Scheimpflug tomography data from 187 eyes of 187 patients. Morphology of healthy and keratoconic corneas was characterized by means of geometric variables. The performance of these variables as predictors of a new geometric marker was assessed and their correlations were analysed.ResultsThe more representative variable to classify the corneal anomalies related to keratoconus was posterior apex deviation (area under receiver operating characteristic curve > 0.899; p < 0.0001). However, the strongest correlations in both healthy and pathological corneas were provided by the metrics directly related to the thickness, as deviations of the anterior/posterior minimum thickness points.ConclusionsThe presented morphogeometric approach based on the analysis and custom geometric modelling of the cornea demonstrates to be useful for the characterization and diagnosis of keratoconus disease, stating that geometrical deformation is an effective marker of the ectatic disease’s progression.

Highlights

In case of significant imperfections on the cornea, data acquisition is difficult and a significant level of missing data could require the interpolation of important areas of the cornea, resulting in a very ambiguous model
It is one of the leading indications for corneal transplantation. This pathology is characterised by a progressive corneal thinning and a structural weakening, resulting in corneal protrusion, irregular astigmatism and a gradual deterioration of the visual performance related to morphology changes in the corneal architecture
We found a satisfactory level of discriminative ability for anterior corneal surface area (AUROC, 0.853), posterior corneal surface area (AUROC, 0.813) and the anterior apex deviation (AUC, 0.742)

Summary

Introduction

In case of significant imperfections on the cornea, data acquisition is difficult and a significant level of missing data could require the interpolation of important areas of the cornea, resulting in a very ambiguous model. In clinical practice the ophthalmologists use corneal topographers based on the Scheimpflug technology [4] This diagnostic equipment gives a matrix of discrete points for the anterior and posterior surface of the cornea [5]. These raw data are not interpolated and are used to generate a geometric model with the modal methods called Zernike polynomials. These Zernike polynomials are defined for all discrete points of the anterior and posterior corneal surfaces for their reconstruction. In cases of pathological corneas, as for instance very aberrated corneas, it has been demonstrated that it is very difficult to define the Zernike polynomial order required to get the most relevant information about the corneal surfaces [5,6,7]

Methods

Results

Discussion

Conclusion