Filippo Ruffini
Full profile and recent publications of Filippo Ruffini.
PhD
Filippo Ruffini
PhD Candidate @ Umeå University / Università Campus Bio-Medico di Roma
Filippo Ruffini is a PhD student enrolled in the Italian National Program in Artificial Intelligence, jointly affiliated with Università Campus Bio-Medico di Roma (UCBM) and Umeå University (Department of Diagnostics and Intervention). He received his Bachelor's degree in Medical Engineering from the University of Rome Tor Vergata, and his Master's degree in Biomedical Engineering, with honors (summa cum laude). His research focuses on trustworthy multimodal AI for oncology, spanning survival prediction, radiology vision-language models, medical image synthesis, and cross-modal retrieval. He is particularly interested in building models that fuse heterogeneous clinical data, imaging, tabular, and textual, into robust, interpretable representations for personalized cancer care. His work combines large-scale HPC infrastructure with end-to-end deep learning pipelines built on PyTorch and MONAI.
Recent Publications
- Virtual Scanning for NSCLC Histology: Investigating the Discriminatory Power of Synthetic PETarXiv preprint arXiv:2605.02746, 2026
- SHOVIR: A Benchmark for Evaluating Vision Shortcut Learning in Radiology Report GenerationarXiv preprint arXiv:2606.30201, 2026
- Probabilistic NDVI Forecasting from Sparse Satellite Time Series and Weather CovariatesarXiv preprint arXiv:2602.17683, 2026
- Machine Learning Models for Sepsis: From Early Detection to Short- and Long-Term Prognosis2026
- Handling Missing Modalities in Multimodal Survival Prediction for Non-Small Cell Lung Cancer2026
- Cross Modality Image Translation In Medical Imaging Using Generative FrameworksarXiv preprint arXiv:2605.13686, 2026
- Text-to-CT Generation via 3D Latent Diffusion Model with Contrastive Vision-Language PretrainingarXiv preprint arXiv:2506.00633, 2025
-
- Machine Learning for Predicting the Low Risk of Postoperative Pancreatic Fistula After Pancreaticoduodenectomy: Toward a Dynamic and Personalized Postoperative Management Strategy2025
- Doctor-in-the-Loop: An explainable, multi-view deep learning framework for predicting pathological response in non-small cell lung cancer2025