Architected materials that consist of multiple subelements arranged in particular orders can demonstrate a much broader range of properties than their constituent materials. However, the rational design of these materials generally relies on experts’ prior knowledge and requires painstaking effort. Here, we present a data-efficient method for the high-dimensional multi-property optimization of 3D-printed architected materials utilizing a machine learning (ML) cycle consisting of the finite element method (FEM) and 3D neural networks. Specifically, we apply our method to orthopedic implant design. Compared to uniform designs, our experience-free method designs microscale heterogeneous architectures with a biocompatible elastic modulus and higher strength. Furthermore, inspired by the knowledge learned from the neural networks, we develop machine-human synergy, adapting the ML-designed architecture to fix a macroscale, irregularly shaped animal bone defect. Such adaptation exhibits 20% higher experimental load-bearing capacity than the uniform design. Thus, our method provides a data-efficient paradigm for the fast and intelligent design of architected materials with tailored mechanical, physical, and chemical properties.

Reconciling the dilemma between rapid degradation and overdose toxicity is challenging in biodegradable materials when shifting from bulk to porous materials. Here, we achieve significant bone ingrowth into Zn-based porous scaffolds with 90% porosity via osteoinmunomodulation. At microscale, an alloy incorporating 0.8 wt% Li is employed to create a eutectoid lamellar structure featuring the LiZn₄ and Zn phases. This microstructure optimally balances high strength with immunomodulation effects. At mesoscale, surface pattern with nanoscale roughness facilitates filopodia formation and macrophage spreading. At macroscale, the isotropic minimal surface G unit exhibits a proper degradation rate with more uniform feature compared to the anisotropic BCC unit. In vivo, the G scaffold demonstrates a heightened efficiency in promoting macrophage polarization toward an anti-inflammatory phenotype, subsequently leading to significantly elevated osteogenic markers, increased collagen deposition, and enhanced new bone formation. In vitro, transcriptomic analysis reveals the activation of JAK/STAT pathways in macrophages via up regulating the expression of Il-4, Il-10, subsequently promoting osteogenesis.

Laser powder bed fusion (L-PBF) of magnesium (Mg) alloy porous scaffolds is expected to solve the dual challenges from customized structures and biodegradable functions required for repairing bone defects. However, one of the key technical difficulties lies in the poor L-PBF process performance of Mg, contributed by the high susceptibility to oxidation, vaporization, thermal expansion, and powder attachment etc. This work investigated the influence of L-PBF energy input and scanning strategy on the formation quality of porous scaffolds by using WE43 powder, and characterized the microstructure, mechanical properties, biocompatibility, biodegradation and osteogenic effect of the as-built WE43 porous scaffolds. With the customized energy input and scanning strategy, the relative density of struts reached over 99.5%, and the geometrical error between the designed and the fabricated porosity declined to below 10%. Massive secondary phases including intermetallic precipitates and oxides were observed. The compressive strength (4.37–23.49 MPa) and elastic modulus (154.40–873.02 MPa) were comparable to those of cancellous bone. Good biocompatibility was observed by in vitro cell viability and in vivo implantation. The biodegradation of as-built porous scaffolds promoted the osteogenic effect, but the structural integrity devastated after 12 h by the immersion tests in Hank's solution and after 4 weeks by the implantation in rabbits' femur, indicating an excessively rapid degradation rate.