Use Case 1 – Fibrosis progression in MASLD patients

In Use Case 1, the focus is on developing a mechanistic virtual twin of liver fibrosis progression in MASLD, designed to capture the complex multi-scale biological processes driving the disease. This model integrates knowledge at the cellular and tissue levels with clinical signals derived from imaging and blood biomarkers, thereby bridging the gap between experimental mechanistic insights and real-world patient data. By simulating disease trajectories, the mechanistic model provides a framework to explore how fibrosis develops and to identify potential intervention points.

Artificial intelligence plays a key complementary role by enabling the acceleration and refinement of these mechanistic simulations. AI-based methods are trained on outputs from biophysical models to approximate complex simulations, allowing for faster predictions compatible with clinical decision-making. The interplay of mechanistic modelling and AI not only supports the identification of patients at high risk of rapid fibrosis progression but also enables in silico testing of therapeutic strategies—such as statins, SGLT2 inhibitors, or metformin—to assess their impact on slowing disease progression.

Because many MASLD patients never reach advanced liver disease but remain at significant risk of cardiovascular complications, the model adopts a multi-organ perspective. By extending the liver fibrosis model with AI-driven links to cardiovascular outcomes, the virtual twin evolves into a holistic patient representation, supporting both preventive strategies and individualized treatment planning.