Artificial Intelligence in Dental Radiology: A Multi-modality Systematic Review

Cephalometric radiography, orthopantomography (OPG), and cone beam computed tomography (CBCT) are three critical imaging modalities that have significantly benefited from the use of artificial intelligence (AI), machine learning (ML), and deep learning (DL). This study aims to bridge the gap between theoretical advancements and practical applications while presenting a comprehensive bibliometric analysis in the area. AI-based models like Convolutional Neural Networks (CNNs), U-shaped Convolutional Neural Network (U-Net), no-new-U-Net (nnU-Net), transformer-based models, and hybrid models perform landmark localization, tooth classification, three-dimensional segmentation, sex classification, lesion detection, and growth prediction accurately. AI improves caries, cyst, and periodontal anomaly detection in OPGs, reducing manual tracing errors in cephalometric radiographs. DL is used for CBCT applications for performing complex tasks like airway detection, mandibular canal segmentation, alveolar bone segmentation, and implant planning. These technologies have improved diagnostic accuracy, automated workflows, and clinical decision-making capabilities. As AI is gaining momentum in transforming orthodontics, implantology, and oral surgery, solving issues like dataset shortage, model generalizability, and clinical validation is of utmost importance. This review emphasizes on the rapid development of AI in dental imaging and the need for future studies to address the limitations that currently exist to achieve the ethical and widespread use of AI in dental diagnostics.