Uncertainty optimization of dental implant based on finite element method, global sensitivity analysis and support vector regression

In this work, an uncertainty optimization approach for dental implant is proposed to reduce the stress at implant–bone interface. Finite element method is utilized to calculate the stress at implant–bone interface, and support vector regression is used to replace finite element method to ease the computational cost. Deterministic optimization based on support vector regression is conducted, which demonstrates that the method using support vector regression replacing finite element method in dental implant optimization is efficient and reliable. Global sensitivity analysis based on support vector regression is used to assign different uncertainties (manufacturing errors) to different design variables to save the manufacturing cost. Two popular uncertainty optimization methods, k-sigma method and interval method, are used for the uncertainty optimization of dental implant. The results indicate that the stress at implant–bone interface is reduced greatly considering the uncertainties in design variables with the manufacturing cost increasing a little. This approach can be promoted to other types of bio-implants.