ACCELERATED DESIGN AND INTEGRITY ASSESSMENT OF ADDITIVELY MANUFACTURED METALLIC STENTS USING MACHINE-LEARNING MODELS
Aida NonnGrand Ballroom E
In this work, we investigate the potential of laser powder bed fusion (L-PBF) to meet the stringent requirements imposed on metallic stents for the treatment of aortic dissection. Here, we use microstructure-based modeling to describe the mechanical properties of L-PBF 316L stainless steel. The derived structure-property relationships then serve as a database for training machine learning (ML) models, such as convolutional networks (CNN) and graphical neural networks (GNN). Based on the established modeling framework, we are able to predict the deformation and fracture behavior of 316L stents and identify the improved stent design in an efficient manner.