Trustworthy and Explainable Artificial Intelligence for Harmonization and Analysis of Medical Imaging in Neurodegenerative Diseases

The research activity focuses on the development of advanced Artificial Intelligence methods forharmonization, analysis, and interpretation ofheterogeneous medical imaging and clinical data related to neurodegenerative diseases. The work will address preprocessing, domain adaptation, and multimodal learning techniques aimed at reducing variability across acquisition protocols, devices, and clinical centers. Deep learning models for segmentation, classification, and clinical decision support will be developed, with particular emphasis on Explainable AI, Trustworthy AI, robustness, fairness, and uncertainty quantification. The proposed solutions will be validated on real multicenterdatasets to promote the adoption of reliable,transparent, and transferable AI systems for healthcare and clinical research.