PNN-SVM Approach of Ti-Based Powder’s Properties Evaluation for Biomedical Implants Production

The advent of additive technologies has provided a significant breakthrough in the production medical implants. It reduced costs, increased productivity and accuracy implant manufacturing process. However, there are problems associated with assessing defects microstructure, mechanical technological properties alloys, both during their by powder metallurgy process 3D printing. Thus traditional research methods alloys demand considerable human, material, time resources. At same time, artificial intelligence tools create opportunities for intelligent evaluation conformity phase composition, titanium powder’s alloys. provides new possibilities efficient biocompatible implants various functional purposes. models used should be as high possible. In this paper we designed hybrid PNN-SVM (Probabilistic Neural Network-Support Vector Machine) high-precision approach alloy biomedical We have proposed extending dimensionality input data space outputs summation layer modified PNN topology. Subsequent classification based on expanded dataset is performed using SVM. conducted experimental modeling set Ti-6Al-4V Ti-Al-V-Zr. demonstrated increase scheme compared to single classifiers that form it other machine learning methods.