A Scoping Review of Machine Learning in Dental Radiography: Its Current Applications and Relevance in Dentistry

Abstract Background Artificial Intelligence (AI) has rapidly developed over the past decade, with seamless integrations across many industries. In a world where healthcare is more crucial than ever, AI can assist clinicians in identifying and diagnosing dental-related anatomy and pathology. Aims Explain the current AI model designs utilised in dental radiography, map out the emergent themes in the current literature and comment on AI model accuracy in radiographic object recognition and interpretation. Methods Using four databases (PubMed, Web of Science, Scopus and EBSCOHost), a search strategy was employed to identify relevant published literature from January 2012 - September 2022. The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool was used to assess the experimental validity of each study included in this review. For each study included, the data extracted included study source, image type, dataset number, AI architecture, data pre-processing, train/validation/test data split and model performance values. Results 18 studies were included in the Discussion spanning four different categories including dental and maxillofacial radiology, orthodontics, periodontology, and restorative dentistry. Conclusions AI models as demonstrated in this study can identify dental-skeletal landmarks with reasonable accuracy and can be applied in numerous restorative dentistry contexts.