The "Artificial Intelligence for Robust Glaucoma Screening Challenge" (AIROGS) was organized to provide machine learning research groups with a comparable data set for the development of systems for the automatic detection of glaucomas. The goal of the challenge was to develop machine learning systems that can robustly detect glaucomas in realistic clinical scenarios. 

In total, over 110,000 fundus images from more than 500 locations of a diverse population were made available. In the testing phase, teams worldwide submitted their algorithms to compare them on an unseen test dataset.

The Ocumeda Machine Learning Team joined forces with PD Dr. Daniel Truhn and Firas Khader from RWTH Aachen in a team and achieved 2nd place in the challenge with their developed solution.