MAECI Italy-China
Bilateral research programme on trustworthy multimodal AI for COVID-19 risk analysis, explainability, and robust validation across imaging and clinical data.
Collaborative Research
MAECI Italy-China
Bilateral research programme on trustworthy multimodal AI for COVID-19 risk analysis, explainability, and robust validation across imaging and clinical data.
Overview
The Italy-China programme focused on trustworthy next-generation precision medicine for COVID-19, combining chest X-ray, chest CT, and clinical information to identify patients at risk of severe outcomes. The technical core of the project was multimodal deep learning for richer joint representations under limited and heterogeneous biomedical data.
Beyond pure prediction, the programme addressed explainability, counterfactual reasoning, and human-understandable concepts to make AI outputs more useful to physicians, patients, and regulators. The collaboration joined ArCo’s work on multimodal learning and explainable AI with complementary expertise in deep networks and CT-driven risk analysis.
Research Directions
- Multimodal signatures for severe COVID-19 outcomes using imaging and clinical variables.
- Explainable AI methods for risk factor analysis and transparent prognostic modelling.
- Bilateral collaboration joining chest X-ray and chest CT expertise with robust validation protocols.
Related Publications
- XGeM: A multi-prompt foundation model for multimodal medical data generation2026foundation models generative AI multimodal learning
- ACGM: Attribute-Centric Graph Modeling Network for Concurrent Missing Tabular Data Imputation and COVID-19 Prognosis2025oncology COVID-19 explainability
- Next-Gen Health: from Multimodal AI to Foundation Models2025foundation models generative AI multimodal learning medical imaging
- Benchmarking foundation models and parameter-efficient fine-tuning for prognosis prediction in medical imaging2025foundation models clinical prediction medical imaging
- Any-to-Any Vision-Language Model for Multimodal X-ray Imaging and Radiological Report Generation2025foundation models medical imaging generative AI
- Can Foundation Models Really Segment Tumors? A Benchmarking Odyssey in Lung CT Imaging2025foundation models medical imaging
- Multi-Dataset Multi-Task Learning for COVID-19 Prognosis2024COVID-19 clinical prediction multimodal learning
- Multi-objective optimization determines when, which and how to fuse deep networks: An application to predict COVID-19 outcomes2023COVID-19 multimodal learning clinical prediction
- AIforCOVID: Predicting the clinical outcomes in patients with COVID-19 applying AI to chest-X-rays. An Italian multicentre study2021COVID-19 medical imaging clinical prediction
- Pareto optimization of deep networks for COVID-19 diagnosis from chest X-rays2021COVID-19 medical imaging clinical prediction
- A Multi-Expert System to Detect COVID-19 Cases in X-ray Images2021COVID-19 medical imaging multimodal learning