Background: The aim is to classify dentition using a novel texture-based automated convolutional neural network (CNN) for forensic and prosthetic applications. Methods: Natural human teeth (n = 600) were classified, cleaned, and inspected for exclusion criteria. The teeth were scanned with an intraoral scanner and identified using a texture-based CNN in three steps. First, through preprocessing, teeth images were segmented by extracting the front-facing region of the teeth. Then, texture features were extracted from the segmented teeth images using the discrete wavelet transform (DWT) method. Finally, deep learning-based enhanced CNN models were used to identify these images. Several experiments were conducted using five different CNN models with various batch sizes and epochs, with and without augmented data. Results: Based on experiments with five different CNN models, the highest accuracy achieved was 0.8 and the precision was 0.8 with a loss value of 0.9, a batch size of 32, and 250 epochs. A comparison of deep learning models with different parameters showed varied accuracy between the different classes of teeth. Conclusion: The accuracy of the point-based CNN method was promising. This texture-identification method will pave the way for many forensic and prosthodontic applications and will potentially help improve the precision of dental biometrics.
Background/purpose With the advancement of an over aging society, the average number of remaining teeth has increased. However, these remaining teeth do not always have sufficient alveolar bone support, and sometimes fabricated connected crowns are applied. This study evaluated the influence of crown material, crown thickness, and alveolar bone resorption on the stress distribution within the abutment teeth of connected crowns. Materials and methods Using structural analysis software, a premolar crown model was fabricated. Three kinds of crown materials, two types of crown thickness, two types of post and core systems, and two levels of alveolar bone were assumed and evaluated for the stress distribution within the abutment teeth. Results The higher material properties crown was, the more stress was concentrated at the marginal area. The composite resin core showed larger stress values around the marginal area, and the metal core showed larger stress values at the tip of the post. Alveolar bone resorption progressed, the marginal area stress value increased. Conclusion The low elastic modulus crown material polyetheretherketone (PEEK) prevented stress concentrations at the marginal area of the crown and dentine, even with alveolar bone resorption. However, the amount of bone resorption has a great influence on the stress distribution around the tip of the post compared to the type of crown material.
To analyze and compare the emergence angle (EA) using two measurement methods, conventional and modified (EA-GPT and EA-R), the EAs of all-natural teeth were evaluated and classified to derive a suitable and predictable clinically applicable measurement method. Methods: Natural human teeth (n=600) were classified, cleaned, and thoroughly inspected. Teeth were scanned using an intraoral scanner. The scanned data were analyzed using three-dimensional analysis software for both methods with several points per surface. A Bland-Altman analysis was used for statistical analysis and a heat map and a nonparametric density plot to assess the repetition and distribution. An XGBoost regression model was used for prediction. Results:The EA-R method showed significantly different values compared to the EA-GPT method, representing an increase of 17.5-20.7% for the proximal surfaces. An insignificant difference between the two methods was observed for other surfaces. Different teeth classes showed variation in the normal range, thereby resulting in a new classification of the EA for all-natural teeth based on the interquartile range. The machine learning gradient boosting model predicted conventional data with an average mean absolute error of 0.9. Conclusions: Variations in the natural teeth EA and measurement methods, suggest a new classification for EA. The established artificial intelligence method demonstrated robust performance, which could aid in implementing EA measurement in prosthetic designs.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.