Image Classification and Analysis Using Deep Learning: Trends, Insights, and Applications of Convolutional Neural Networks

Dental imaging plays a crucial role in various healthcare applications such as dental planning, disease treatment, and post-mortem human identification in case of a disaster scenario. This paper provides a comprehensive study of the diverse applications of the various dental imaging techniques for the different applications. It focuses on the methodology, application, dataset, merits, demerits, and system evaluation metrics. Further, it provides the implementation of dental radiograph classifications using a deep convolution neural network (DCNN) to depict the texture and shape attributes of dental radiographs. The effectiveness of the DCNN-based dental radiograph classification is estimated on the dental radiography dataset (DRD) based on recall, precision, F1-score, negative predictive value (NPV), selectivity, and accuracy on the dental radiograph dataset. The DCNN provides improved overall accuracy of 91%, recall of 90.7%, selectivity of 91.3%, precision of 91.3%, F1-score of 91.04%, and NPV of 90.7%.