A Theoretical Approach for Diagnosing Keratoconus Using Purkinje Image Geometry: An Accessible Method for Low-Cost and Early Detection

Purpose: To develop and evaluate a theoretical method for diagnosing keratoconus by analyzing the geometric properties of Purkinje images, with a focus on the displacement and distortion of the first Purkinje image (P1). Methods: Purkinje images are captured using near-infrared LEDs integrated into attachments for existing fundus cameras or stand-alone devices. Image processing techniques are employed to measure P1 displacement and distortion. A composite scoring system quantifies keratoconus severity, and data augmentation simulates diverse corneal shapes for robust diagnostic comparisons. Results: Simulated analysis demonstrated that the method effectively differentiates between healthy and keratoconic corneas by assessing key metrics such as Euclidean distance, eccentricity, area, and perimeter of P1. The scoring system consistently reflected keratoconus severity, with higher scores indicating greater deviation from normal corneal profiles. Conclusions: This study presents a promising, low-cost approach to diagnosing keratoconus using Purkinje image analysis. The method's adaptability and potential for early detection make it particularly suitable for low-resource settings. Translational Relevance: This method bridges the gap between basic research and clinical care by providing an affordable and accessible diagnostic tool for early keratoconus detection, especially in underserved regions.