Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
Application of Convolutional Neural Networks for Dentistry Occlusion Classification
Ist Teil von
Wireless personal communications, 2024, Vol.136 (3), p.1749-1767
Ort / Verlag
New York: Springer US
Erscheinungsjahr
2024
Link zum Volltext
Quelle
Alma/SFX Local Collection
Beschreibungen/Notizen
The study in the field of periodontics and etiology focuses on the crucial task of classifying occlusion classes in dentition through the application of deep learning algorithms. The occlusion patterns between upper and lower jaws play a pivotal role in understanding and treating dental conditions such as periodontitis, Pierre Robin syndrome, and maxilla fractures.The extent of asymmetrical overlap between the upper and lower jaw forms various classes of occlusion. Hence, the classification of occlusion becomes an essential prerequisite for the successful treatment of many dentistry related diseases like oral cancer, gingival recession, and tooth erosion.The research employed a dataset comprising 200 dental images extracted from Stereolithography (STL) files using an Intraoral scanner, presenting 2D representations of dental structures. Various deep learning architectures, including LeNet, AlexNet, Inception, and DenseNet, were utilized for the classification task. The Inception model emerged as the most accurate, achieving an 84.39% accuracy rate due to its non-sequential architecture, followed closely by DenseNet at 84.10%, LeNet at 82.39%, and AlexNet at 78.43%. Therefore, these accuracy results indicated a relative trend as Inception > DenseNet > LeNet > AlexNet.The study suggests the potential application of the automated classification system, particularly based on the Inception model, by clinicians due to its high accuracy, effectiveness, and efficiency in processing time. This technological advancement holds promise for significantly contributing to treatment planning and surgeries in dental practice.