Detecting Keratoconus of Cornea Using Real Clinical Dataset

By Tingting Han, Raid Rafi Omar Al-Nima, Mustafa Ali Malla

The cornea is essential for vision, providing most of the eye's refractive power, so even minor structural changes can significantly affect sight. Keratoconus is a progressive condition that thins the cornea, leading to visual distortion and making early diagnosis challenging.This work introduces a clinical dataset of corneal maps covering curvature, thickness, and elevation from different layers of the cornea. The images are processed through a specialized pipeline to enhance their quality before analysis. An artificial intelligence approach is applied using adapted deep learning models, where each type of corneal map is analyzed separately to extract meaningful features and predict the presence of keratoconus.The outputs from these individual models are then combined using fusion techniques to improve overall diagnostic performance. Comparative evaluation of different fusion strategies identifies the most effective approach, resulting in highly accurate detection of keratoconus and demonstrating the reliability of the proposed system for clinical use.

• Format: Paperback
• Language: English
• ISBN: 6209824536
• ISBN13: 9786209824531
• Release Date: April 2026
• Publisher: Scholars' Press
• Length: 116 Pages
• Weight: 0.37 lbs.
• Dimensions: 9.0" x 0.3" x 6.0"