Date:17/06
Hour: 13.00
Student Name & ID: Sena Tütüncü & 255105104
Supervisor: Prof.Dr. Ömer KARAL
Topic: Deep Learning-Based Classification of Malocclusion from 3D Dental STL Models
Link or Room: A516
Abstract:
The growing complexity of orthodontic diagnostics and the increasing availability of digital dental data present significant opportunities for the application of artificial intelligence in clinical dentistry. Conventional diagnostic approaches rely on manual examinations and subjective clinical assessments, which are time-consuming and prone to inter-examiner variability. In particular, the classification of sagittal occlusal disorders, such as overjet and underbite, remains a challenging task demanding considerable clinical expertise. This study proposes a deep learning-based classification framework for the automated detection and categorization of these malocclusion types using three-dimensional (3D) dental models in STL format. The system classifies dental occlusion into three distinct categories: overjet, underbite, and normal occlusion. A convolutional neural network (CNN) architecture is employed to extract discriminative features from the processed 3D dental data and perform accurate classification. The proposed framework aims to reduce diagnostic subjectivity, accelerate clinical workflows, and serve as a reliable decision support tool for orthodontists in the early identification of occlusal disorders.