Machine Learning-Guided Process Optimization for Fatigue Life Enhancement in Additively Manufactured Titanium Dental Implants

Titanium and titanium alloys are widely used as raw materials for dental implants due to their high corrosion resistance and favorable mechanical properties. Among the methods for producing these implants is the additive manufacturing method. One method for investigating the properties of these implants can be based on machine learning.. study presents a machine learning (ML)-driven approach to predict and optimize the fatigue life and build quality of titanium dental implants fabricated through additive manufacturing (AM). By analyzing parameters such as laser power, scan speed, hatch spacing, and preheat temperature, the models assess their influence on porosity, surface roughness, and fatigue performance. A synthetic dataset was generated based on validated AM ranges. Through correlation analysis, feature importance modeling, and classification/regression techniques, we identify optimal processing windows. Random Forest and XGBoost models demonstrated high predictive accuracy, validating the feasibility of data-driven process control for producing high-performance implants with extended fatigue resistance.