Filippo Bargagna is a biomedical engineer and machine learning researcher at the University of Pisa. He earned both his Bachelor's (2019) and Master's (2022) degrees in Biomedical Engineering at the University of Pisa, with a Master's thesis on Bayesian deep learning applied to structured clinical data using probabilistic programming. He holds a PhD in Information Engineering (Biomedical AI), carried out jointly between the University of Pisa and Fondazione Toscana Gabriele Monasterio, with a thesis on trustworthy AI in medical imaging. During his doctorate he spent six months as a visiting researcher at the MGH/HST Athinoula A. Martinos Center for Biomedical Imaging (Harvard Medical School / Massachusetts General Hospital), working at the intersection of neuroimaging and probabilistic deep learning.
His research integrates bioimage and signal analysis with probabilistic deep learning, uncertainty quantification, explainable AI, and generative modelling, with applications spanning cardiac imaging, neuroimaging, and peripheral physiological signals. A recurring theme of his work is the separation of model uncertainty from data-intrinsic stochastic components, and the development of methods that remain reliable under conditions of noise, data scarcity, and multi-site heterogeneity.
At the Neuro-Cardiovascular Intelligence Lab, Filippo works on concurrent multimodal bioimage and signal analysis (via statistical and ML/DL based methodologies) for the investigation of brain–heart axis interplay dynamics, with particular attention to the role of the endogenous biological stochastic component underlying these interactions.
