Prediction Of Keratoconus Progression Using High Resolution Oct Measurements And Bayes Statistics

Published 2023 - 41st Congress of the ESCRS

Reference: PP11.06 | Type: Free paper | DOI: 10.82333/08z9-fp22

Authors: Andreas Honeder* ¹ , Leon Pomberger ¹ , Klemens Waser ¹ , Haidar Khalil ¹ , Matthias Bolz ¹ , Peter Laubichler ¹ , Nino Hirnschall ¹

¹Ophthalmology and Optometry,Johannes Kepler University, Kepler University Clinic ,Linz,Austria

Purpose

Aim of this study was to predict keratoconus progression using high resolution OCT measurments and Bayes statistics.

Setting

This Study is conducted retrospectively at the Kepler University Clinic Linz, MCIII, Department of Ophthalmology and Optometry in its entirety.

Methods

Patient data was collected from our Keratoconus outpatient clinic in a retrospective fashion. In all cases a Scheimpflug measurement (Pentacam HR, Oculus, Germany) and a spectral domain OCT (MS-39, CSO, Italy) were available. Bayes statistics were used for the prediction of Keratoconus progression.

Results

In total 345 eyes of 173 patients were included in this study. Mean age was 33.13 years (SD 13.56; median: 30; range 8 to 74). Kmax (D) Scheimpflug front was (Median: 50.23, Mean: 52.37, SD: 7.89). Kmax (D) OCT front was (Median: 50.54, Mean: 52.47, SD: 7.85). Astigmatism (D) Scheimpflug was (Median: 2.60, Mean: 3.04, SD: 2.10). Cyl (D) OCT was (Median: 2.72, Mean: 3.08, SD: 2.12). The thinnest pachy (µm) of the Scheimpflug device was (Median: 480.00, Mean: 475.57, SD: 65.12) The thinnest pachy (µm) of OCT was (Median: 475.61, Mean: 470.64, SD: 65.10) A novel Bayes based statistical approach for Keratoconus progression prediction will be presented at the meeting.

Conclusions

Prediction of keratoconus progression with artificial intelligence is possible, but more difficult in later stages of keratoconus