AIRIS

Mechanism-Informed Multimodal Generative AI for Causal and Dynamical Modelling in Biomedical Research

Description

Despite rapid advances in artificial intelligence, today's biomedical AI systems largely rely on statistical correlations rather than an understanding of the biological mechanisms underlying disease. As a result, their predictions are often difficult to interpret and trust, particularly for complex diseases with not yet fully understood causes.

AIRIS addresses this challenge by developing a new generative AI platform that combines multimodal biomedical data with biological knowledge to model disease mechanisms and support scientific discovery. The AI research collaborator will support scientists throughout the entire biomedical research process. By integrating medical imaging, genomic data, laboratory results, patient records and scientific literature, AIRIS will enable researchers to identify previously unknown disease pathways, generate evidence-based hypotheses and accelerate the development of personalised medicine.

Unlike conventional "black-box" AI systems, AIRIS places explainability, transparency and ethical AI at its core. The platform will evaluate every generated hypothesis for plausibility, novelty and supporting evidence before presenting it to researchers. Its capabilities will be demonstrated across five complex disease areas – Pulmonary Fibrosis (PF), Steatotic Liver Disease (SLD), Cardiovascular Disease (CVD), Chronic Kidney Disease (CKD), and Inflammatory Bowel Disease (IBD) – with findings validated through computational and laboratory studies.

Coordinator

Prof. Dr Christos Diou

Athina-Erevnitiko Kentro Kainotomias Stis Technologies Tis Pliroforias, Ton Epikoinonion Kai Tis Gnosis

Email