Design of patient specific dental implant using FE analysis and computational intelligence techniques

Abstract

Genetic algorithm is employed for optimum designing of patient specific dental implants with varying dimension and porosity. It is generally recommended that, the micro strain at the bone implant interface should be around 1500–3000. The porous dental implant needs to be designed in such a way that the micro stain remains within the above range, and a value close to 2500 micro strain is most desired. In this design problem, the most important constraint is that the implant stress should be limited within 350 MPa as titanium alloy was considered as implant material. The above attributes are to be achieved per the varying bone conditions of the patients to design a patient specific prosthesis. This design problem is expressed as an optimization problem using the desirability function, where the data generated by finite element analysis is converted to an artificial neural network model. The output of the neural network model is converted within a range of 0–1 using desirability function, where the maximum value is reached at the most desired micro strain of 2500. This hybrid model of neural network and desirability function is used as the objective function for the optimization problem using genetic algorithm. Another neural network model describing the implant stress is used as the constraint. The optimum solutions achieved from ANN and GA are validated again through finite element method. Without doing stress analysis by FEM, the ANN models are used for measuring the fitness of the members of the population during optimization. This would predict the optimum dimension of dental implant made of Titanium alloy with most favorable porosity percentage for better ossiointegration for a patient per bone condition.