Title: PhagoStat a scalable and interpretable end to end framework for efficient quantification of cell phagocytosis in neurodegenerative disease studies

This study demonstrates that incorporating explainability into AI does not necessarily lead to performance degradation, but offers promising capabilities, particularly for optimizing AI models. This makes them more streamlined, faster, efficient and, above all, environmentally friendly, with a reduced ecological footprint.

Authors: Mehdi Ounissi (co-financed by SCAI), Morwena Latouche, Daniel Racoceanu.

They also released PhagoStat, a comprehensive open-source pipeline as well as a unique microglial cell phagocytosis dataset for immune system characterization in neurodegenerative disease research, both capable of consistently supporting advances futures in this field, by promoting the development of efficient interpretable algorithms dedicated to the critical characterization of neurodegenerative diseases.