Exploring the role of Convolutional Neural Networks (CNN) in dental radiography segmentation: A comprehensive Systematic Literature Review

In dentistry, there is a growing need for accurate diagnostic tools, particularly advanced imaging techniques such as Computed Tomography (CT), Cone Beam Computed Tomography (CBCT), Magnetic Resonance Imaging (MRI), ultrasound, and traditional intraoral periapical X-rays. Deep Learning (DL) has emerged as a pivotal tool, facilitating automated segmentation that is crucial for extracting essential diagnostic data. This integration of cutting-edge technology addresses the urgent need for effective management of dental conditions. If left undetected, these conditions can significantly affect human health. Deep Learning (DL) has an impressive track record in various domains, including dentistry, underscoring its potential to revolutionize early detection and treatment of oral health issues. Convolutional Neural Networks (CNNs) have demonstrated significant results in diagnosis and prediction, representing an emerging field of multidisciplinary research. The goals of this study were to provide a concise overview of the state of the art, standardize ongoing debates, and establish baselines for future research. The methodology employed in this study involved a Systematic Literature Review (SLR) to identify and select relevant studies that specifically investigated deep learning techniques for dental imaging analysis. This study elucidates a methodological approach that includes systematic data collection, statistical analysis, and outcome dissemination. By incorporating 45 studies, we identified the selection criteria and research objectives, addressing significant gaps in the existing literature. These studies will assist clinicians in examining dental conditions and classifying dental structures, such as detecting cavities and identifying different types of teeth. The model performance was evaluated by addressing the identified gaps using a variety of metrics that were outlined and explained. This study demonstrated the effectiveness of using CNNs to analyze images and serves as an effective tool for detecting dental pathologies. Despite acknowledging some limitations, CNNs used for segmenting and categorizing teeth demonstrated their highest level of performance overall.