Abstract
This study presents a comprehensive framework combining Selective Laser Melting (SLM) of Titanium (Ti64) alloys, finite element simulation, and artificial intelligence (AI) to advance orthopedic implants' design and predictive evaluation. Dense Ti64 specimens are fabricated using ten distinct SLM parameter sets to explore the effects of volumetric energy density (VED) on mechanical behavior, porosity distribution, and microstructural integrity. Optimal VED ranges are identified to balance defect minimization and mechanical performance, with porosity levels strongly influencing tensile strength and Young's modulus. These material properties are then integrated into a mechanobiological bone healing model to simulate fracture repair in tibial mid-diaphyseal regions under intramedullary fixation. A mechano-regulation algorithm incorporating deviatoric strain, fluid velocity, and pore pressure guided tissue differentiation and callus evolution over a 4-month healing period. Variations in implant stiffness-achievable via solid and lattice Ti64 configurations demonstrate distinct influences on endochondral ossification, external callus growth, and overall biomechanical recovery. AI-based regression models are trained on simulation outcomes to accelerate evaluation and enable patient-specific predictions to forecast healing trajectories without repeated finite element analyses. This AI-driven framework facilitates real-time prediction of healing outcomes based on implant design, supporting the development of adaptive and optimized orthopedic implants for clinical translation.