Thickness Speed Progression Index: Machine Learning Approach for Keratoconus Detection

Shady T Awwad; Bassel Hammoud; Jad F Assaf; Lara Asroui; James Bradley Randleman; Cynthia J Roberts; Douglas D Koch; Jawad Kaisania; Carl-Joe Mehanna; Shadi Elbassuoni

doi:10.1016/j.ajo.2024.11.011

Thickness Speed Progression Index: Machine Learning Approach for Keratoconus Detection

Am J Ophthalmol. 2024 Nov 27:271:188-201. doi: 10.1016/j.ajo.2024.11.011. Online ahead of print.

Authors

Affiliations

¹ From the Department of Ophthalmology (S.T.A., B.H., L.A., and C.J.M.), American University of Beirut Medical Center, Beirut, Lebanon. Electronic address: [email protected].
² From the Department of Ophthalmology (S.T.A., B.H., L.A., and C.J.M.), American University of Beirut Medical Center, Beirut, Lebanon; Cole Eye Institute (B.H. and J.B.R.), Cleveland Clinic, Cleveland, Ohio, USA.
³ American University of Beirut (J.F.A.), Faculty of Medicine, Beirut, Lebanon.
⁴ From the Department of Ophthalmology (S.T.A., B.H., L.A., and C.J.M.), American University of Beirut Medical Center, Beirut, Lebanon.
⁵ Cole Eye Institute (B.H. and J.B.R.), Cleveland Clinic, Cleveland, Ohio, USA; Cleveland Clinic Lerner College of Medicine of Case Western Reserve University (J.B.R.), Cleveland, Ohio, USA.
⁶ Departments of Ophthalmology & Visual Sciences and Biomedical Engineering (C.J.R.), The Ohio State University, Columbus, Ohio, USA.
⁷ Cullen Eye Institute (D.D.K.), Baylor College of Medicine, Houston, Texas, USA.
⁸ Department of Computer Science (J.K. and S.E.), American University of Beirut, Beirut, Lebanon.

PMID: 39608740
DOI: 10.1016/j.ajo.2024.11.011

Abstract

Purpose: To develop and validate a pachymetry-based machine learning (ML) index for differentiating keratoconus, keratoconus suspect, and normal corneas.

Design: Development and validation of an ML diagnostic algorithm.

Methods: This retrospective study included 349 eyes of 349 patients with normal, frank keratoconus (KC), and KC suspect (KCS) corneas. KCS corneas included topographically/tomographically normal (TNF) and borderline fellow eyes (TBF) of patients with asymmetric KC. Six parameters were derived from the corneal thickness progression map on the Galilei Dual Scheimpflug-Placido system and fed into a machine-learning algorithm to create the Thickness Speed Progression Index. The model was trained with 5-fold cross-validation using a random search over 7 different ML algorithms, and the best model and hyperparameters were selected.

Results: A total of 133 normal eyes, 141 KC eyes, and 75 KCS eyes, subdivided into 34 TNF and 41 TBF eyes, were included. In experiment 1 (normal and KC), the best model (Random Forest) achieved an accuracy of 100% and area under the receiver operating characteristic (AUROC) of 1.00 for both normal and KC groups. In experiment 2 (normal, KCS, and KC), the model achieved an overall accuracy of 91%, and AUROC curves of 0.93, 0.83, and 0.99 in detecting normal, KCS, and KC corneas respectively. In experiment 3 (normal, TNF, TBF, and KC), the model achieved an accuracy of 87% with AUROC curves of 0.91, 0.60, 0.77, and 0.94 for normal, TNF, TBF, and KC corneas, respectively.

Conclusions: Using data solely based on pachymetry, ML algorithms such as the Thickness Speed Progression Index are able to discriminate normal corneas from KC and KCSs corneas with reasonable accuracy.