Predicting Low-Modulus Biocompatible Titanium Alloys Using Machine Learning

Titanium alloys have been present for decades as the main components production of various orthopedic and dental elements. However, modern times require titanium with a low Young’s modulus, without presence cytotoxic alloying Machine learning was used aim to analyze biocompatible predict composition Ti modulus. A database created using experimental data alloy composition, mechanical thermal properties alloys. The Extra Tree Regression model built modulus By processing 246 alloys, specific heat discovered be most influential parameter that contributes lowering Further, Monte Carlo method future desired properties. Simulation results ten million samples, predefined conditions obtaining lower than 70 GPa, show it is possible obtain several multicomponent consisting five elements: titanium, zirconium, tin, manganese niobium.