An adaptive bidirectional mapping framework for the customized design of interpenetrating phase composite scaffolds

Structural optimization of artificial bone scaffolds faces trade-offs between mechanical integrity and permeability. To address this, we propose an integrated framework for interpenetrating phase composites (IPCs) using a hybrid TPMS topology (F-RD, IWP, and Neovius). We developed a Dung Beetle Optimizer-enhanced Deep Neural Network (DBO-DNN) for accurate property prediction and a Particle Swarm Optimization algorithm (SCLFPSO) for inverse design. Numerical experiments demonstrate that our framework achieves superior convergence, enabling efficient, on-demand customization of scaffolds. This approach provides a robust, low-cost paradigm for structural-functional design for bone tissue engineering.