Justraigs Competition Insights: An Ensemble of Methods for Automated Referable Glaucoma Diagnosis

Glaucoma is a leading cause of irreversible vision loss and blindness worldwide. Early detection and treatment are essential for preventing visual impairment, and population-based glaucoma screening plays a key role in facilitating early identification. This paper presents a technical report on a solution for the Justified Referral in AI Glaucoma Screening (JustRAIGS) challenge. The proposed approach tackles a dual classification task, combining binary and multi-label classification using large input images. To manage data imbalance, techniques such as soft labeling, balanced weighted sampling, and image augmentations were employed. For post-processing, the solution optimizes thresholds and utilizes test-time augmentations to enhance prediction accuracy. This comprehensive approach aims to improve the efficiency and effectiveness of glaucoma screening, ultimately contributing to better patient outcomes and advancing the field of ophthalmology.Source code: github.com/danpresil/JustRAIGS-IEEE-ISBI-2024.