Surrogate model assisted design optimization of composite bone plates to achieve selective stress shielding

ABSTRACT

Optimal composite bone plate is designed for Periprosthetic Femoral Fracture (PFF) fixation using genetic algorithm (GA) by hybrid and varying fiber orientation. The bone plate should allow higher displacement in axial direction and minimum movement in shear direction. Composite material is considered for the bone plate, the most important constraint in this design problem is that the axial movement and shear movement are limited to 2 and 0.5 mm, respectively. The simulated data from the finite element analysis (FEA) are used for creating the artificial neural network (ANN) surrogate models for the bone displacements in both the directions, which act as the objective functions for the GA-driven optimization in the multi-objective fashion. FEA is used to validate the solutions derived through the design optimization process. The FE simulations have shown only 10% deviation in the simulation output when compared to the achieved optimum solutions.

KEYWORDS:

• Composite
• bone plate
• femur
• periprosthetic fracture
• multi-objective optimization
• movements
• finite element analysis
• artificial neural network
• genetic algorithm

Disclosure statement

No potential conflict of interest was reported by the authors.