• Open Daily: 10am - 10pm
    Alley-side Pickup: 10am - 7pm

    3038 Hennepin Ave Minneapolis, MN
    612-822-4611

Open Daily: 10am - 10pm | Alley-side Pickup: 10am - 7pm
3038 Hennepin Ave Minneapolis, MN
612-822-4611
Detecting Keratoconus and Non-Keratoconus Cases

Detecting Keratoconus and Non-Keratoconus Cases

Paperback

Technology & Engineering

ISBN10: 6209724450
ISBN13: 9786209724459
Publisher: LAP Lambert Academic Publishing
Published: Apr 6 2026
Pages: 116
Weight: 0.36
Height: 0.28 Width: 6.00 Depth: 9.00
Language: English
The cornea, which provides over two-thirds of the eye's focusing power, is highly sensitive to even minor curvature changes that can significantly affect vision. Keratoconus is a progressive condition characterized by corneal thinning, leading to myopia and irregular astigmatism, making early detection essential for effective management. This book introduces a specialized medical system for early keratoconus detection using a newly developed dataset, the Real Clinical Dataset for Keratoconus (RCDK), which includes multiple corneal maps such as axial/sagittal curvature, thickness, and elevation. Advanced preprocessing techniques are applied to enhance image quality. An Artificial Intelligence approach based on Deep Learning models, is used to analyze these maps. Each model processes a specific map, and a fusion technique combines their outputs to improve diagnostic accuracy. Using 704 images, the system achieved high individual accuracies ranging from 90% to 97.14%, while the fusion method reached 100% accuracy. The system is implemented on Raspberry Pi 4 with Python and a user-friendly interface.

Also from

Mohammed Saeed, Mustafa

Also in

Technology & Engineering