Experimental and numerical investigation of 3D-Printed bone plates under four-point bending load utilizing machine learning techniques.

The fused deposition modeling (FDM) technique is widely used to produce components for various applications and has the potential to revolutionize orthopedic research through the production of custom-fit and readily available biomedical implants. The properties of FDM-produced implants are significantly influenced by processing parameters, with layer thickness being a crucial parameter. This study investigated the effect of layer thickness on the flexural properties of Polylactic Acid (PLA) bone plate implants produced by the FDM technique. Experimental results showed that the flexural strength is inversely proportional to the layer thickness due to the variation of voids in the specimens. A 3D finite element (FE) model was developed using Abaqus/Explicit software by incorporating the Gurson-Tvergaard (GT) porous plasticity model to predict the elastoplastic and damage behavior of specimens with different layer thicknesses. The characterization of the elastoplastic and GT parameters was done using a tensile test and by the calibration of a machine learning algorithm. It was shown that the FE model was able to predict the flexural behavior of 3D-printed solid plates with a maximum error of 6.13% in the maximum load. The optimal layer height was found to be 0.1 mm, providing both high flexural strength and adequate bending stiffness.