A comprehensive approach to understand the mechanical behavior of materials involves costly and time-consuming experiments. Recent advances in machine learning and in the field of computational material science could significantly reduce the need for experiments by enabling the prediction of a material’s mechanical behavior. In this paper, a reliable data pipeline consisting of experimentally validated phase field simulations and finite element analysis was created to generate a dataset of dual-phase steel microstructures and mechanical behaviors under different heat treatment conditions. Afterwards, a deep learning-based method was presented, which was the hybridization of two well-known transfer-learning approaches, ResNet50 and VGG16. Hyper parameter optimization (HPO) and fine-tuning were also implemented to train and boost both methods for the hybrid network. By fusing the hybrid model and the feature extractor, the dual-phase steels’ yield stress, ultimate stress, and fracture strain under new treatment conditions were predicted with an error of less than 1%.

This study presents a multi-objective optimization framework for designing cylindrical triply periodic minimal surface (TPMS) lattices tailored for bone implant applications. Using an artificial neural network (ANN) as a surrogate model trained on simulated data, four key properties—ultimate stress (U), energy absorption (EA), surface area-to-volume ratio (SA/VR), and relative density (RD)—were predicted from seven lattice design parameters. To address anatomical variability, a novel implant size-based categorization (small, medium, and large) was introduced, and separate optimization runs were conducted for each group. The optimization was performed via the NSGA-II algorithm to maximize mechanical performance (U and EA) and surface efficiency (SA/VR), while filtering for biologically relevant RD values (20–40%). Separate optimization runs were conducted for small, medium, and large implant size groups. A total of 105 Pareto-optimal designs were identified, with 75 designs retained after RD filtering. SHapley Additive exPlanations (SHAP) analysis revealed the dominant influence of thickness and unit cell size on target properties. Kernel density and boxplot comparisons confirmed distinct performance trends across size groups. The framework effectively balances competing design goals and enables the selection of size-specific lattices. The proposed approach provides a reproducible pathway for optimizing bioarchitectures, with the potential to accelerate the development of lattice-based implants in personalized medicine.

In this study, a micromechanical finite element model is proposed based on experimental data and the rule of the mixture (RoM) in order to predict the tensile behavior of mechanical properties of heat-treated dual-phase medical-grade titanium (Ti–6Al–4V). Tensile tests, micro-hardness measurements, and RoM were used to obtain flow curves for the α and β phases. Scanning electron microscopy (SEM) imaging was used to determine phase fractions and to create representative volume elements (RVEs). Furthermore, the Gurson-Tvergaard-Needleman (GTN) damage model was calibrated using the Taguchi design of experiment (DOE) method in order to predict damage in the microstructure and the results were compared to fracture surface obtained using fractography in order to investigate failure mechanisms. The final micromechanical model could accurately predict stress-strain curves and showed that void formation and coalescence is the primary mechanism of failure. Finally, analyses of the surfaces showed that a fully ductile fracture occurs at the failure point, which agrees with the results of the damage model. The results suggest that the proposed model can predict the failure of heat-treated Ti–6Al–4V bio-alloys.

A Machine Learning (ML)-based surrogate modeling framework is presented for mapping structure–property relationships in architected Ti6Al4V cylindrical TPMS metamaterials subjected to quasi-static compression. A Python–nTop pipeline automatically generated 3456 cylindrical shell lattices (Gyroid, Diamond, Split-P), and ABAQUS/Explicit simulations with a Johnson–Cook failure model for Ti6Al4V quantified their mechanical response. From 3024 valid designs, key mechanical properties targets including elastic modulus (E), yield stress (Y), ultimate strength (U), plateau stress (PL), and energy absorption (EA) were extracted alongside geometric descriptors such as surface area (SA), surface-area-to-volume ratio (SA/VR), and relative density (RD). A multi-output surrogate model (feedforward neural network) trained on the simulated set accurately predicts these properties directly from seven design parameters (thickness; unit cell counts in X, Y, and Z directions; unit cell orientation; height; diameter), enabling rapid property estimation across large design spaces. Topology-dependent trends indicate that Split-P exhibits the highest strength, energy absorption, and total SA, and shows the largest variation in SA/VR; Gyroid exhibits the lowest SA with a moderate SA/VR; and Diamond is the most compliant lattice and maintains a higher SA/VR than Gyroid despite lower SA. RD increases with both SA and SA/VR across all topologies. The framework provides a reusable computational tool for architectured lattices, enabling quick prescreening of implant designs without repeated finite-element analyses.