Official ESCRS | European Society of Cataract & Refractive Surgeons
Copenhagen 2016 Registration Programme Exhibitor Information Virtual Exhibition Satellite Meetings Glaucoma Day 2016 Hotel Star Alliance

10 - 14 Sept. 2016, Bella Center, Copenhagen, Denmark

This Meeting has been awarded 27 CME credits


escrs app advert yo advert

Corneal deformation parameters for diagnosing keratoconus

Search Title by author or title

Session Details

Session Title: Corneal Biomechanics

Session Date/Time: Tuesday 13/09/2016 | 16:30-18:00

Paper Time: 16:42

Venue: Auditorium C6

First Author: : B.Lopes BRAZIL

Co Author(s): :    I. Ramos   A. Luz   R. Vinciguerra   P. Vinciguerra   R. Ambrosio Jr.        

Abstract Details


To develop a support vector machine (SVM) model based on corneal deformation parameters (CDP) from the Corvis ST® (Oculus; Wetzlar, Germany) to distinguish normal and keratoconic corneas.


In a retrospective multicenter study, one eye randomly selected from 662 patients were enrolled from two clinics at different continents, being 482 normal eyes (255 clinic 1 and 227 from clinic 2) and 180 keratoconic cases (79 clinic 1 and 101 from clinic 2).


Patient selection was based on clinical criteria including Placido-disk based corneal topography and rotating Scheimpflug corneal tomography. All patients were analyzed with the Corvis ST (OCULUS Optikgeräte GmbH; Wetzlar, Germany) to extract CDP, intra ocular pressure and horizontal thickness data. In this study, only CPD parameters were considered to develop a support vector machine (SVM) model to distinguish both groups. ROC curves were used in a hold-out validation set to test the accuracy of the model.


The CDP-SVM model achieved an area under the curve (AUC) in the validation set of 0.970, with sensitivity of 93.3% and specificity of 90% for a cut-off value of 0.206. The CDP-SVM had a statistically significant higher AUC than Deformation Amplitude ratio at 2 mm, which had best AUC among individual CDP (AUC=0.933, sensitivity of 86.2% and specificity of 85.8% (De Long, p<0.0001).


This is possible to distinguish keratoconic from normal corneas exclusively based on corneal deformation response. The CDP-SVM model could be used in combination with geometric data (ie. Horizontal thickness profile and central corneal radius) to further improve accuracy for the diagnosis of ectatic corneal diseases and to characterize ectasia susceptibility.

Financial Disclosure:

... receives consulting fees, retainer, or contract payments from a company producing, developing or supplying the product or procedure presented

Back to previous