Editorial: Explainable Artificial Intelligence (XAI) in Systems Neuroscience
Angela Lombardi, João Manuel R. S. Tavares, Sabina Tangaro · Frontiers in Systems Neuroscience · 2021
In the last 10 years, we have experienced exceptional growth in the development of machine-learning-based (ML) algorithms for the analysis of different medical conditions and for developing clinical decision support systems. In particular, the availability of large datasets and the increasing complexity of both hardware and software systems have enabled the emergence of the new multidisciplinary field of computational neuroscience (Teeters et al., 2008). Sophisticated machine learning algorithms can be trained using brain imaging data to classify neurodegenerative disorders, detect neuropsichiatric conditions (Davatzikos, 2019), and perform accurate brain age prediction for the identification of novel functional and structural biomarkers for different diseases (Cole and Franke, 2017). In this Research Topic, we collected several original research works where different XAI techniques were embedded in both ML and DL algorithms for the extraction of reliable biomarkers from neuroimaging datasets for several predictive tasks.