Digital Twin Simulates Liver Health and Disease Progression

The HEPATWIN framework introduces a physiology-informed digital twin designed to simulate the human liver's function and the progression of early-stage diseases over time. Unlike purely data-driven models, HEPATWIN incorporates mechanistic representations of key hepatic processes, including carbohydrate, lipid, and protein metabolism, bilirubin conjugation, bile production, and detoxification. This unified systems-level framework uses patient-specific inputs such as diet, activity, and baseline biomarkers to generate trajectories of clinically observable biomarkers. To ensure these simulations align with real-world disease progression, a stage-transition-driven calibration mechanism is employed, matching simulated outputs with population-level biomarker distributions across various disease stages like NAFLD, fibrosis, and cirrhosis. Validation against the NIDDK NAFLD dataset demonstrated that HEPATWIN produces longitudinal biomarker estimates within clinically acceptable ranges and can forecast trajectories over multiple years. Furthermore, the simulated biomarkers retained sufficient clinical signal to support downstream NASH detection with performance comparable to models using actual laboratory data, highlighting its potential for personalized, non-invasive organ health monitoring and prediction.