Multi-objective optimization of an origami inspired super-expandable scaffold for distraction osteogenesis

Abstract:

The extended bone consolidation period in distraction osteogenesis (DO) is a known clinical challenge. Tissue-engineered scaffolds (TIS) have shown promise in enhancing bone regeneration, but conventional designs are not suitable for the large deformations required in DO. This study presents an origami-inspired super-expandable scaffold (OISES) tailored for DO. A Bayesian machine learning platform was used to predict key structural and mechanical properties influencing bone regeneration, followed by multi-objective optimization. Selected designs were fabricated and tested both virtually and experimentally. The optimized OISES configurations maintain porosity near 80 % and high surface area-to-volume ratios—parameters associated with osteogenesis—while exhibiting large recoverable deformations. Simulations revealed consistent mechanical behavior and strain localization, along with a deformation-dependent increase in permeability and decrease in wall shear stress during scaffold expansion. This transition supports a shift from early mechanical stimulation to enhanced fluid and nutrient transport, mirroring the evolving biological needs during early-stage bone healing. These findings suggest that OISES scaffolds can dynamically adapt to the mechanical and biological demands of DO, offering a new class of tunable scaffolds optimized for patient-specific treatments.

Keywords:

Origami-inspired scaffold; Multi-objective optimization; Distraction osteogenesis; Fluid–structure interaction; Bayesian machine learning.