A Study on Tooth Decay Detection Using a Simple Convolutional Neural Network Model

Tooth decay is a prevalent and severe oral health issue affecting individuals of all ages. Early detection and diagnosis of tooth decay are crucial for successful treatment and preventing further damage to the teeth, thereby reducing the workload for dentists. This study proposes a convolution neural network (CNN) model for early-stage tooth decay detection. The proposed CNN model is simple and includes two Convolutional layers trained with minimal parameters. In this study, some model factors were investigated to improve the accuracy of the proposed CNN model in detecting tooth decay. The results show that a CNN model with a batch size of 16 and using a dropout layer achieved an accuracy of 99.61% in detecting tooth decay, which is higher than other popular models, such as VGG19, Resnet50, and DenseNet12, in accuracy with the same dataset.