We proposed a convolutional neural network for vertex classification on 3-dimensional dental meshes, and used it to detect teeth margins. An expanding layer was constructed to collect statistic values of neighbor vertex features and compute new features for each vertex with convolutional neural networks. An end-to-end neural network was proposed to take vertex features, including coordinates, curvatures and distance, as input and output each vertex classification label. Several network structures with different parameters of expanding layers and a base line network without expanding layers were designed and trained by 1156 dental meshes. The accuracy, recall and precision were validated on 145 dental meshes to rate the best network structures, which were finally tested on another 144 dental meshes. All networks with our expanding layers performed better than baseline, and the best one achieved an accuracy of 0.877 both on validation dataset and test dataset.
Purpose
This in vitro study aims to explore the effects of selective laser melting (SLM) process parameters on the accuracy of the intaglio surface of cobalt–chromium alloy (Co–Cr), commercially pure titanium (CP Ti) and titanium alloy (Ti–6Al–4V) maxillary removable partial denture (RPD) frameworks and optimize these process parameters.
Design/methodology/approach
Maxillary RPD framework specimens designed on a benchmark model were built. The process parameters, including contour scan speed and laser power, infill scan speed and laser power, hatch space, build orientation and metallic powder type, were arranged through the Taguchi design. Three-dimensional deviations of the clasps area, connector area and overall area of maxillary RPD frameworks were analyzed by using root mean square (RMS) as a metric. One-way analyses of variance with the above RMSs as the dependent variable were carried out (α = 0.05).
Findings
Maxillary RPD frameworks built horizontally had a more accurate intaglio surface than those built at other orientation angles; CP Ti or Ti–6Al–4V maxillary RPD frameworks had a more accurate intaglio surface than Co–Cr ones; the Maxillary RPD framework built with a higher infill scan speed and lower infill laser power had the more accurate intaglio surface than the one built with other levels of these two process parameters.
Originality/value
A novel benchmark model for evaluating the accuracy of the intaglio surface of maxillary RPD frameworks manufactured by SLM is proposed. The accuracy of the intaglio surface of maxillary RPD frameworks can be improved by adjusting SLM process parameters. The optimal setting of process parameters concerning the accuracy of the intaglio surface of maxillary RPD frameworks was given.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.