Dental caries is one of the most common oral health issues that needs to be identified and treated early. Although panoramic X-ray imaging is frequently used to diagnose dental caries, dentists find it difficult and time-consuming to interpret these dental images. This study suggests a unique method based on artificial intelligence approaches for automatically detecting dental caries in panoramic X-ray images. The deep learning architecture Panoramic X-ray Tooth Detection Network (PaXNet) for caries classification and a tooth segmentation module based on genetic algorithms make up the suggested approach. In order to detect and classify carious lesions, PaXNet combines transfer learning and capsule networks. On the other hand, tooth segmentation module precisely extracts individual teeth from the panoramic X-ray images. The experimental results show an accuracy of 95.6%. Gradient-weighted Class Activation Mapping visuals and comparisons with the most advanced models further demonstrate PaXNet's robustness and interpretability. The proposed method has a great deal of promise to assist dentists to identify and treat dental caries at its early stage, which will enhance patient care and results. For practical application, future research will concentrate on improving the approach by adding more clinical data and integrating it into clinical workflows
Datta et al. (Thu,) studied this question.