Development of AI Tool for Caries Detection from Dental Radiograph

Dental caries (DC), sometimes called tooth decay, is a bacterial illness that demineralizes and destroys the tooth's complex structures, leaving cavities or holes in the teeth. DC may result in abscesses and tooth loss if left untreated. It is strongly advised that DC be diagnosed and treated as soon as possible. Clinical evaluations usually involve a dentist's ocular assessment and then radiograph-supported screening. This project aims to create an AI-supported system for radiograph-based dental caries detection. The suggested method uses a Convolutional Neural Network (CNN) architecture to find the DC region in digital radiographs. The work combines CNN segmentation with picture pre-processing to increase detection accuracy. The suggested tool's efficacy is assessed on a chosen database, and the experimental findings verify that the scheme can successfully identify DC using VGG-UNet with an accuracy of over 98%, surpassing the performance of UNet, SegNet, and UNet+.