Possibilities of Explainable Artificial Intelligence for Glaucoma Detection Using the LIME Method as an Example

Our study explored the experience of applying explainable artificial intelligence methods to the analysis of medical (retinal) images for glaucoma diagnosis. The possibility of applying Local Interpretable Model-Agnostic Explanations (LIME) to explain the results of artificial neural networks in the task of binary classification of retinal images for the presence of glaucoma is studied. A scale-up of the technology and its application to real clinical practice is proposed.