A Low-Cost Dental Image Scanner for Angle’s Classification of Malocclusion Using Convolutional Neural Network

A malocclusion is a term used by dentists to indicate a poor bite, implying that your teeth are misaligned. This study introduces an innovative approach to implementing the classification of dental malocclusions using a convolutional neural network architecture developed in MATLAB. The proposed method presents an objective and efficient solution by assessing intraoral pictures obtained from a dental imaging scanner. The CNN model was trained on a dataset of classified intraoral photos corresponding to Class I, Class II, and Class III malocelusion types. The findings show that the dental image scanner has a high potential for diagnosing malocclusions using Angle’s classification method, making it a valuable instrument for dentists for assessment and treatment planning. The system’s capacity to give reliable and accurate assessments could improve patient satisfaction and a more efficient dental treatment procedure.