In this presentation, we introduce a novel two-scale computational approach tailored to model cancellous bone remodelling. This approach offers a novel perspective on computational biomechanics by efficiently integrating meso- and macro-scale considerations. At the macro-scale, conventional principles governing one-scale continuum bone remodelling are upheld, focusing on established kinematics and kinetics. However, we depart from traditional methods by refraining from phenomenologically postulating constitutive behavior at this level. Instead, we derive it from the meso-scale.

At the meso-scale, our approach leverages computational efficiency by idealizing trabecular architecture as a truss network, dynamically adapting trabecular cross-sectional areas in response to mechanical loading. The synergy between meso- and macro-scale dynamics is achieved through sophisticated up- and down-scaling techniques.

Computational experiments demonstrate the effectiveness and computational efficiency of our proposed two-scale approach. Notably, it seamlessly captures anisotropic properties stemming from the irregular trabecular architecture at the meso-scale. Moreover, it provides a unique opportunity to directly explore various trabecular structures, acting as a virtual "magnifying glass" for detailed analysis.

The presentation will discuss the potential for further advancements, including the exploration of more complex trabecular architectures and the integration of micro-scale bone cellular activities. Through this study, we aim to inspire advancements in biomechanical research and enhance our understanding of cancellous bone remodelling processes for clinical applications.

Recent advancements in 3D printing (3DP) technologies, particularly light-based ones, have augmented the production of functionally graded porous structures (FGPSs) tailored for tissue engineering (TE) applications. However, selecting optimal architectures for specific tissue scaffolds involves time-consuming trial-and-error processes. Therefore, there is an urgent need to integrate in silico-aided procedures with 3DP methods to streamline and optimise this workflow.

This study focuses on exploiting such procedures by developing an in silico tool for the design, mechanical validation, and optimisation of light-based 3D-printed porous structures designed for bone TE scaffolds. The architectures were designed with full interconnectivity, varied pore sizes, layer pore size gradients, and porosities. Employing low-force stereolithography 3DP for its superior resolution and cost-effectiveness, quasi-static uniaxial compression tests were conducted for mechanical property assessment in low-speed analyses. Finite element models were then employed to validate mechanical and fluid flow properties, while morphological analysis aided in identifying printing accuracy and limitations. Finally, an artificial neural network (ANN) was trained to determine the correlation between the scaffolds’ architecture and their mechanical properties.

Firstly, it was possible to control the printability of structures, understanding printing limitations arising from material properties (e.g., the viscosity of the liquid resin and its stiffness) and design methodology. Secondly, the developed finite element (FE) model enabled the correlation of scaffolds’ architecture with their mechanical properties, predicting the elastic region of the stress-strain curve. Finally, the trained ANN facilitated the creation of functionally graded TE scaffolds with desired stiffness by selecting optimal structural properties such as pore size and pore size gradient.

The obtained data are a starting point for establishing an automated workflow to generate ad-hoc FGPS based on specifications for mechanical properties. Further works will include using the ANN to predict structural non-linear properties under large deformations (e.g., absorbed energy).

Worldwide, an estimated 4M bone grafts are performed each year. While many products are available that are used for bone regeneration for small defects or non load-bearing defect filling, non-union grafts and grafts for treatment of bone metastasis of prostate cancer are the most challenging. Severe skeletal failures result in prostate cancer bone metastasis patients and these are often the reason for the morbidity of prostate cancer bone metastasis. We report the use of a novel interlocked Bone morphogenic protein 2 and 7 (BMP2 and BMP7), coated scaffold block-assembly for applications for non-union defects. Increased surface area allowed by the interlocking interfaces enhance bioactivity while maintaining appropriate mechanical integrity. Extracellular matrix (ECM) formation is enhances in the interlocked assembly. A longer-term study up to 63 days with BMP coated scaffolds indicates a 120% increase in the elastic modulus obtained using nanomechanical evaluation. Interestingly a significant increase in the bone-related protein and osteogenesis-related Wnt-factors was observed in BMP coated scaffolds. It is to be noted that the BMP release rate in the media is about 17-days while the influence on ECM formation lasts much longer, indicating the critical role of BMPs on initial stage osteogenesis. Additional studies report effective use of BMPs in influencing fundamental osteogenic pathways while adverse effects on metastasized cancer. The novel interlocked BMP loaded scaffolds present a unique approach for tuning osteogenesis and thus impacting mechanical behavior of bone.

The observation proposed by Julius Wolff - called Wolff's law - can be described as a structural adaptation of the bone to the external forces. Thus the trabecular bone remodeling process numerical simulation has to include a very important aspect of the external load, namely the variable loads has to be taken into account. For the simulation purposes it means that the numerical tool must be able to simulate multiple load cases and the geometric form of the bone must correspond to these loads. The presented trabecular bone remodeling numerical tool enables multiple load case simulation taking into account the postulates regarding the evolution of the trabecular bone. The trabecular bone remodeling regulatory model applied to an actual three-dimensional trabecular structure requires the preparation of an appropriate numerical approach. Since the local change on the structural surface leads to global minimization of the strain energy for the whole structure, the fulfillment of both postulates requires energy distribution analysis on the structural surface. Thus, the most important role in such an approach must be played by a very efficient finite element mesh generator for structural computations as well as an efficient computational environment. In both cases, it becomes necessary to use parallel processing. So, on one hand it will be possible to repeat virtually the observations recorded on the micro-CT scans, and on the other hand, to better adjust the continuous models of the trabecular bone remodeling phenomenon. The developed software promises to simulate significant fragments of trabecular bone tissue, and as equipment develops, also structures covering the entire bone. Mesh generator performance is no longer a limiting factor nor is computing power. Technically the numerical system is .Net C# project designed with Inversion of Control paradigm design pattern that provides pluggable and extensible platform.