A review of the evolution of in-silico physiological models in bone remodelling: from cell population dynamics to micro-multiphysics agent-based approaches
File(s)
Accepted version
Author(s)
Altaleb, joudi
Hansen, Ulrich
Abel, Richard
Mudway, Ian
Type
Journal Article
Abstract
Bone remodelling is a mechanically regulated, multiscale process that maintains skeletal integrity and is disrupted in common metabolic bone diseases. Experiments, however, cannot easily resolve the coupled cellular, biochemical, and mechanical feedback that drive long-term outcomes. In this Review, we examine five in-silico modelling classes used to study bone remodelling: bone cell population dynamics models, bone tissue dynamics models, finite element models, AI models, and three-dimensional micro-multiphysics agent-based models. We compare these approaches through biological fidelity, computational tractability, and translational readiness. FEM excels at macro-scale mechanics but depends on explicit material laws. Cell population models remain dominant for long-horizon and pharmacological simulations because they are interpretable and efficient, although their non-spatial structure limits prediction of microarchitecture. Tissue dynamics models capture curvature-controlled growth efficiently but often rely on simplified mechanobiology. Micro-multiphysics agent-based models provide the strongest link between cell behaviour, mechanics, and microarchitecture, but face major challenges in parameter identifiability, scalable computation, and validation. AI models can bypass explicit material laws and accelerate other methods through surrogate modelling, but risk non-physical predictions without appropriate constraints. We conclude that the field’s next step is to develop integrated, testable workflows that improve calibration and validation while making assumptions explicit. In this context, AI is best viewed as an enabling tool for surrogate modelling, inverse problems, and data-assimilation workflows, rather than as a replacement for mechanistic understanding.
Date Acceptance
2026-07-01
Citation
The Innovation Life
ISSN
2959-8761
Publisher
The Innovation Life
Journal / Book Title
The Innovation Life
Copyright Statement
Copyright This paper is embargoed until publication. Once published the Version of Record (VoR) will be available on immediate open access.
License URL
Publication Status
Accepted