Machine learning for 3D printed multi-materials tissue-mimicking anatomical models

Polyjet, a material jetting 3D printing technique, has been widely used for the fabrication of patient-specific anatomical models owing to toolless technique and its ability print multiple materials in single part. Although with high dimensional accuracy demonstrated, printed tissue-mimicking properties have not realized. In this study, composite layering design was tune shore hardness compressive modulus Polyjet-printed parts an attempt mimic human tissues. 216 specimens (with 72 combinations parameters) were tested develop library models. An analytical model developed estimate effective laminate. A neural network learn multi-dimensional relationship between parameters mechanical properties. The 5-33-2 size is found be optimum structure mean square error 0.98% modulus, lower than traditional response surface method model. genetic algorithm search space most targeted hardness.