Year: 2026
Type: preprint
Abstract: Keratoconus(KC) is a common degenerative condition of the cornea delineated by corneal steeping caused by corneal protrusion and thinning, impacting the quality of vision or simply affecting the vision negatively. In the early stages, non-surgical treatment options, such as spectacles and soft contact lenses can address vision problems. As the disease progresses, irregular astigmatism may require different treatment options. Screening progress of the condition takes place usually at vision-centres every 6 months. It can be done by using optical coherence tomography. Indeed, topographic profiles can be utilized to track the advancement of keratoconus. In this study, machine learning~(ML) algorithms were employed to detect the severity of keratoconus in a step to monitor the disease. Two long-short Term Memory~(LSTM) based models were applied on a dataset which contains the parameter values of patients' eyes obtained from OCT images with date information. One of the models works on one eye, and the second model considers the effect of both eyes on the progression of the disease. The reported test results are promising, demonstrating the effectiveness and applicability of the developed approach.
Source: Research Square
Author: Sibel Tariyan Ozyer
Institution: Ankara Medipol Üniversitesi
Language: English
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7204155481