Robust Fovea Localization Based on Symmetry Measure.

Area of Science:

• Ophthalmology
• Medical Image Analysis
• Computer Vision

Background:

• Automatic fovea localization is crucial for diagnosing retinal diseases.
• Lesions and illumination variations pose significant challenges to existing methods.
• Current techniques often rely on complex preprocessing like blood vessel segmentation.

Purpose of the Study:

• To propose a robust and accurate automatic fovea localization method.
• To address the issue of lesion interference in fovea detection.
• To provide a practical and controllable tool for ophthalmologists.

Main Methods:

• Concentric Circular Sectional Symmetry Measure (CCSSM) for symmetry axis detection and region of interest (ROI) determination.
• Index of Convexity and Concavity (ICC) for quantitative lesion assessment within the ROI.
• Weighted gradient accumulation map for refined localization, robust to intensity variations and noise.

Main Results:

• The proposed method achieves high accuracy and robustness without complex preprocessing.
• It effectively mitigates interference from lesions and unbalanced illumination.
• Demonstrated reliable performance across five different datasets.

Conclusions:

• The symmetry-based fovea localization method is innovative, simple, and practical.
• It offers superior accuracy and robustness compared to state-of-the-art techniques.
• The method provides a reliable tool for clinical applications in ophthalmology.