A Theoretical Approach for Diagnosing Corneal Ectasia Using Purkinje Image Geometry: An Accessible Method for Low-Cost and Early Detection
Abstract:
Purpose: To propose a theoretical method for diagnosing corneal ectasia, including keratoconus and related conditions, by analyzing Purkinje images, specifically the displacement and distortion of the first Purkinje image (P1). The method aims to provide a cost-effective, accessible diagnostic tool for early detection and monitoring.
Methods: The proposed method utilizes an algorithm that analyzes key metrics, including Euclidean distance, eccentricity, area, and perimeter, derived from Purkinje images. Data augmentation techniques are employed to create a diverse reference dataset simulating varying ectatic conditions. The method is adaptable to multiple platforms, such as attachments to existing fundus cameras or standalone devices and incorporates a composite scoring system to quantify ectasia severity.
Results: Simulated scenarios demonstrate that the method effectively identifies deviations in P1 metrics indicative of ectasia. Higher composite scores correspond to more severe cases, reflecting significant changes in eccentricity, perimeter, and displacement metrics. The robustness of the algorithm is supported by data augmentation, which enhances diagnostic reliability even in the absence of clinical data.
Conclusions: This method provides a promising approach for diagnosing corneal ectasia, offering a cost-effective and adaptable solution. Future integration of artificial intelligence and machine learning is expected to optimize the scoring system, automate the analysis process, and further improve diagnostic accuracy. Translational Relevance: The proposed method bridges the gap between high-end diagnostic tools and accessible, low-cost solutions, particularly for underserved populations. It has the potential to revolutionize corneal ectasia diagnostics by enabling early detection and improving patient outcomes in diverse healthcare settings
