Deep Dirichlet Uncertainty for Unsupervised Out-of-Distribution Detection of Eye Fundus Photographs in Glaucoma Screening

The development of automatic tools for early glaucoma diagnosis with color fundus photographs can significantly re-duce the impact of this disease. However, current state-of-the-art solutions are not robust to real-world scenarios, providing over-confident predictions for out-of-distribution cases. With this in mind, we propose a model based on the Dirich-let distribution that allows to obtain class-wise probabilities together with an uncertainty estimation without exposure to out-of-distribution cases. We demonstrate our approach on the AIROGS challenge, where we achieve a performance sim-ilar to other participants without requiring additional annotations or artificially generated out-of-distribution labels.