Foto del docente

Federico Magnani

PhD Student

Department of Medical and Surgical Sciences

Research

Keywords: Neural Ordinary Differential Equations Physics-Informed Neural Networks Dynamical Systems Network Science Knowledge Graphs (KG) Representation Learning Computer Vision

Dynamical Systems
I'm working towards the integration of neural models into pharmacokynetics, to have personalized therapies. Central challenges of this task are the quantification of the prediction's uncertainty and the interpretability of the algorithm. I'm also involved in a project for estimating the risk of femur fracture, starting from MRI images, in which I would like to employ Physics-Informed Neural Networks.

Network Science
I applied network science to genomic and proteomic data, for identifying signatures related to hematological malignancies; to shotgun sequences of wastewater samples, for studying bacterial populations; to biomedical Knowledge Graphs, for prioritizing promising drug-repurposing candidates.

Computer Vision
I'm contributing with computer vision software to the analysis of whole slide images of the human liver, in the context of host-pathogen interaction; and to that of MRI images, in a variety of settings.

AI Governance and Algorithmic Compliance
In collaboration with the Department of Legal Studies, I contribute to developing good practices and strategies for implementing AI systems compliant by design to the European AI Act. We do so by mutual consultancy and by implementing pilot projects in controlled environments.

Latest news

At the moment no news are available.