A generative artificial intelligence platform is being developed by the international AIRIS collaboration to help scientists integrate complex biomedical data, explore how chronic diseases develop and produce testable hypotheses that could point towards new treatment approaches, with input from UCL on lung and liver disease
Understanding why patients with the same diagnosis follow different disease trajectories remains a major research challenge
Existing generative artificial intelligence (AI) systems can identify statistical relationships in large datasets but may not adequately capture the underlying biological mechanisms driving disease progression, thus limiting the interpretability and reliability of their outputs
The four-year Mechanism-Informed Multimodal Generative AI for Causal and Dynamical Modelling in Biomedical Research (AIRIS) project aims to address this by developing an AI research collaborator that builds and reasons with mechanistic models of disease rather than relying solely on statistical patterns
The €16.9m Horizon Europe-funded consortium began in June 2026 and brings together 21 research and industry partners across Europe, Canada and the US
Five chronic disease areas whose underlying biological mechanisms remain incompletely understood will be used to evaluate the platform: pulmonary fibrosis, steatotic liver disease (SLD), cardiovascular disease, chronic kidney disease and inflammatory bowel disease
Exploring chronic disease knowledge and data
AIRIS will integrate multimodal biological and clinical data, including medical imaging, laboratory tests, genomic information and patient records
Designed as a research platform rather than a clinical decision-making tool, it will support multiple stages of the research process, from harmonising fragmented datasets and identifying potential mechanistic pathways to generating hypotheses for further investigation
The project also plans to use simulated ‘virtual cells’ to link molecular and cellular processes with patient-level outcomes, helping researchers to investigate disease mechanisms across different biological scales
Its outputs will be evaluated across the five disease areas, with key findings independently tested through laboratory and computational studies to determine whether this approach can help to explain differences in disease trajectories, treatment responses and potential therapeutic opportunities
Proposed chronic disease hypotheses will be assessed for plausibility, novelty and supporting evidence before being refined in collaboration with human researchers
Project coordinator Dr Christos Diou, associate professor of AI and machine learning at Harokopio University of Athens in Greece, said: ‘Moving beyond today’s AI systems, our new tool will act as a virtual collaborator capable of helping scientists access and harmonise data, explore disease mechanisms, generate and test hypotheses and design rigorous studies.’
UK expertise in lung disease and SLD
Researchers at University College London (UCL) will contribute expertise in computational imaging, SLD and lung disease to the AIRIS project
Professor Joseph Jacob, a professorial research fellow in respiratory medicine, and colleagues at the Satsuma Lab within the UCL Hawkes Institute, will quantify anatomical structures in lung tissue across imaging scales, from clinical computed tomography to microscale imaging
‘The imaging information we derive will be linked to molecular analyses using generative AI tools to better understand how disease develops and progresses mechanistically across a range of chronic lung diseases,’ Professor Jacob explained
‘Central tenets of the project include transparency, robustness, explainability, bias detection and mitigation, and technical oversight of the AI systems to align them with patient and clinical values and health priorities.’
In the case of SLD, Professor Emmanuel Tsochatzis, professor of hepatology at the UCL Division of Medicine, said: ‘Our team has conducted extensive research over many years to improve the early detection and risk stratification of people with the condition, contributing to meaningful changes in clinical practice
‘By bringing our clinical and research expertise to this collaboration, we aim to generate important new insights into how [SLD] develops and progresses over time, ultimately helping to improve the identification and care of people at greatest risk.’
Want news like this straight to your inbox?
Sign up for our bulletins
