GV Black Inspired Hierarchical Multiclass Classification using Panoramic Radiographic Synthetic Data

The purpose of this research is to demonstrate that using the generated synthetic dataset will produce better results than the original teeth x-rays and thus be more efficient for carrying out hierarchical multiclass classification, this will also allow dental caries classification to be automated in accordance with the GV Black Standards. Five distinct models that were trained and tested using both the new and original datasets are compared. As the first step in each of these tasks, simple object detection will be used to identify each tooth, and then hierarchical classification will be used to achieve the desired results for classifying the cavities so that appropriate treatment can be determined based on the class of caries.