Recent Advances in Dental Panoramic X-Ray Synthesis and Its Clinical Applications

Abstract

X-ray imaging plays a pivotal role in clinical diagnostics by facilitating tasks ranging from organ segmentation to disease detection. With the integration of artificial intelligence (AI), the capabilities of X-ray analysis have been greatly enhanced. However, acquiring a comprehensive dataset for training AI models remains challenging due to various factors including patient privacy, radiation exposure, and cost constraints. To overcome this hurdle, researchers have turned to create synthetic X-ray datasets. In dental imaging, the synthesis of Panoramic X-rays (PXs) poses unique challenges owing to the intricate nature of the oral cavity and the inherent diversity among patients. Existing efforts predominantly rely on Cone Beam Computed Tomography (CBCT) or phantoms for such tasks. Synthesizing PX can be beneficial when the PX imaging cannot be performed immediately or when dentists need to find a one-to-one correspondence between PX and CBCT. This review paper comprehensively examines the methodologies employed for synthesizing dental PXs, specifically, automated synthesis, elucidating the methods utilized and the associated advantages, limitations, and potential clinical applications in routine practice. By shedding light on these approaches, this review aims to provide valuable insights for researchers and practitioners in the dental imaging domain.

Keywords

• CBCT
• CT
• deep learning
• maximum intensity projection
• multi-planar reformatting
• panoramic X-ray