Machine Learning for Additive Manufacturing of 3D Printed Multi‐Material Tissue Mimicking Anatomical Structures

ABSTRACT A new machine learning‐based framework for additive manufacturing optimization in 3D‐printed multi‐material tissue‐mimicking anatomical structures is presented in this study. The objective is to enhance the precision and functionality of printed models by optimizing material selection and printing parameters to replicate key mechanical properties of human tissues, such as elasticity, tensile strength, compressive modulus, and Shore hardness. A comprehensive dataset of material properties is utilized as input to a Sparse Spectral Graph Convolutional Network (SGCN), which captures complex relationships between materials and anatomical structures to predict optimal material combinations and printing parameters. Since SGCN lacks inherent parameter optimization capabilities, the Artificial Gorilla Troops Optimization (GTO) technique is employed to fine‐tune the model's weight parameters, improving predictive accuracy. Python is used to implement the framework, and performance metrics like accuracy, precision, recall, and Root Mean Square Error (RMSE) are used to assess it. The proposed SGCN‐GTO framework achieves an accuracy of 99.2%, precision of 98.5%, recall of 97.8%, and a low RMSE of 0.034, significantly outperforming existing models such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Artificial Neural Networks (ANNs), which achieved average accuracies below 90%. This research provides a promising step toward advancing the clinical applicability of 3D‐printed anatomical structures for surgical planning, medical education, and personalized healthcare by delivering models with superior mechanical fidelity and optimized production efficiency.