Computational models for predicting bone fracture healing: a review of modeling approaches, predictions, and emerging...

The development of computational models for predicting bone fracture healing process holds strong potential to optimize therapeutic management in non-unions and delayed healing, reducing healthcare costs and disability-adjusted life years. The main goal of this study is to provide a thoroughly comparative analysis to computational models already proposed to predict fracture healing, including methodologies, mathematical frameworks, validation techniques, and comparative findings across different studies.

This review analyzes 60 computational models selected through a systematic search in the Scopus database (2000-2025) using targeted keywords and rigorously screened according to predefined inclusion and exclusion criteria for simulating bone fracture healing, focusing on mechanical, biological, mechanobiological, ultrasound, and bioelectronic dynamics, employing FEM and artificial intelligence techniques. The effectiveness of each model in predicting healing progression was assessed by analyzing their computational frameworks, accuracy, and limitations.

Comparative analysis revealed that both mechanical and biological models provide fundamental predictions related to fracture healing (e.g., stress distribution and vascularization), but they often lack the physiological complexity demanded for clinical application. Mechanobiological models accurately predict tissue differentiation by combining mechanical stimuli, with strong qualitative agreement with in vivo histological patterns. Ultrasound models have been effective for non-invasive structural assessment, despite existing limitations due to simplified boundary conditions. Notably, the development of bioelectric models has been demonstrating a highly sensitive approach for assessing fracture healing.

This study highlights that multidomain computational frameworks combining mechanobiological dynamics with dielectric properties hold significant potential to personalize the clinical management of delayed bone healing.