Research
Breast Cancer Screening
AI models are applied to the EPIC dataset to predict women’s risk of developing the disease. The research focuses mainly on modifiable risk factors, exploring model explainability and whether it might offer insights relevant for prevention strategies.
Decoder-Free Supervoxel GNN for Accurate Brain-Tumor Localization in Multi-Modal MRI
This project shows that shifting focus from reconstruction to representation leads to both accurate and interpretable brain-tumor localization.
Federated Learning Approach for Explainable Multilayer Graph Neural Networks
This project shows how explainable AI and federated learning can work hand-in-hand to support clinicians while protecting patients.
The Interplay Between Explainability and Differential Privacy in Federated Healthcare
This project highlights the delicate balance between privacy, accuracy, and explainability in federated healthcare, and takes a step toward AI systems that are both trustworthy and clinically usable.
Publications
Federated GNNs for EEG-Based Stroke Assessment
A Federated Learning Platform as a Service for Advancing Stroke Management in European Clinical Centers
Towards Explainable Graph Neural Networks for Neurological Evaluation on EEG Signals
Feasibility Analysis of Federated Neural Networks for Explainable Detection of Atrial Fibrillation
Brain MRI Screening Tool with Federated Learning
Investigation and perspectives of using Graph Neural Networks to model complex systems : the simulation of the helium II bayonet heat exchanger in the LHC
A Carbon Tracking Model for Federated Learning: Impact of Quantization and Sparsification
Decentralized Federated Learning for Healthcare Networks: A Case Study on Tumor Segmentation
A Niching Augmented Evolutionary Algorithm for the Identification of Functional Dependencies in Complex Technical Infrastructures from Alarm Data
Identification of Critical Components in the Complex Technical Infrastructure of the Large Hadron Collider Using Relief Feature Ranking and Support Vector Machines
A Novel Association Rule Mining Method for the Identification of Rare Functional Dependencies in Complex Technical Infrastructures from Alarm Data
Association Rules Extraction for the Identification of Functional Dependencies in Complex Technical Infrastructures
A Feature Selection-Based Approach for the Identification of Critical Components in Complex Technical Infrastructures: Application to the CERN Large Hadron Collider
Hybrid machine learning method for support brain tumor diagnosis
Data-Driven Extraction of Association Rules of Dependent Abnormal Behaviour Groups
Survey on international standards and best practices for patch management of complex industrial control systems: the critical infrastructure of particle accelerators case study
A smart framework for the availability and reliability assessment and management of accelerators technical facilities
