An Intelligent System for Dental Disease Detection Using Smart R-CNN Technique

Abstract

Artificial intelligence (AI) in the medical system has more with the development of research in the field of deep learning, demand has increased.   However, only a few Deep Learning models can deliver solutions in real-time applications. In these type of real time medical systems, the primary concern is data availability and the same occurred for dental dataset. In this paper the most optimized and accurate method is discussed. Dataset is classified and labeled into 5 different datasets. We propose a faster neural network with a Densenet model in this research. Densenet is a useful tool in deep learning for overcoming obstacles such as picture segmentation, accurate categorization of images with high levels of recognition, and powerful optimization algorithms that speed up convergence speeds. lowers the local computation on each subsequent outer iteration. We discovered from the current system that when there is a large dataset, the Densenet model exhibits good classification in original picture classification issues. We have identified three different diseases associated with various types of cavities, such as frontal teeth cavities, inner cavity teeth, and oral cancer, which will determine the difference between diseased and healthy teeth in teeth disease detection using this network construction framework on real clinical data at various levels. In order to incorporate the built ML model into the webapp, it is being stored and translated into JSON file format. An interface for diagnosing dental problems will be provided by the web application.