Background and Motivation:
Liver transplantation (LTx) is a life-saving treatment for end-stage liver disease, but a critical shortage of donor organs remains a persistent challenge. The growing number of marginal donors—often older, multimorbid individuals—and the simultaneous increase in multimorbid recipients complicate clinical decision-making. Marginal liver grafts, typically affected by hepatic steatosis, exhibit impaired perfusion, metabolism, and function, increasing the risk of post-transplant complications.
Ischemia-reperfusion injury (IRI) during transplantation significantly impacts graft and patient outcomes. Machine perfusion has emerged as a promising strategy to mitigate these effects, extending preservation times and enabling ex vivo repair. However, there remains considerable uncertainty about the optimal perfusion method—normothermic or hypothermic oxygenated perfusion – and how to tailor these approaches to specific clinical scenarios.

Objective:
Building on the success of SimLivA I, which developed a coupled continuum-biomechanical model for IRI on liver lobule and cell scale, SimLivA II aims to refine this model to better predict IRI and support decision-making in LTx. Specifically, the project will extend the existing model to simulate different machine perfusion strategies and incorporate bile production as an indicator of liver function. SimLivA II seeks to develop a robust simulation-supported scoring system for organ allocation, enabling clinicians to optimize outcomes for marginal grafts.

Methods:
The project employs an iterative co-design approach to integrate modeling, experimental, and clinical perspectives. The project begins by developing finite element meshes based on experimental and clinical samples, enabling detailed simulations of liver tissue. Lobular geometries will be extracted, and zonated quantifications of proteins, damage, and steatosis will be incorporated to create simulation-ready geometries.
To capture the complexity of liver graft perfusion, the existing model will be extended to couple ordinary and partial differential equations across organ, lobule, and cellular scales. Temperature-dependent tissue properties, including deformation, permeability, and solute diffusivity, will also be incorporated to refine the simulation’s realism under varying machine perfusion conditions. A key extension involves adding bile transport and metabolism as a second diffusive pathway, complementing the blood transport mechanism and providing a more comprehensive representation of liver function under stress.
To support clinical decision-making, a surrogate modeling framework will be developed, leveraging a physics-informed, data-driven operator learning approach. This framework will enable near real-time predictions based on the complex multi-scale model. Experimental and clinical validation will be conducted using ex vivo liver perfusion and archived clinical samples, ensuring the model’s accuracy and applicability. The experimental findings will also guide parameterization, focusing on the effects of donor pathology, cold storage, and machine perfusion conditions.
The project culminates in the development of a simulation-supported scoring system that translates model outputs into actionable clinical insights. This scoring system will synthesize existing scoring methodologies with new data derived from the model, enhancing decision-making for marginal liver graft allocation and optimizing patient outcomes.

Impact:
SimLivA II will advance predictive modeling for LTx by mapping machine perfusion strategies into a multi-scale, continuum-biomechanical simulation framework. The project aims to provide clinicians with a robust decision-support tool to optimize the use of marginal grafts, ultimately improving patient outcomes and addressing the organ shortage crisis.