Lateral Cephalometric Radiography: Principles, Common Positioning Errors, and AI-Driven Quality Control

This narrative review provides a contemporary synthesis of lateral cephalometric radiography (LCR), addressing both its foundational principles and the impact of technological integration, with a focus on enhancing diagnostic reliability. A structured literature search (PubMed, up to September 2025) was conducted around five domains: LCR's diagnostic role, acquisition methods, positioning errors, comparisons with cone-beam computed tomography (CBCT), and Artificial Intelligence (AI)-driven quality control. Precise patient positioning-maintaining symmetry and a horizontal Frankfort plane-is paramount, as common errors (tilting, rotation, nodding) introduce quantifiable inaccuracies in key measurements. While digital innovation, particularly deep learning models for automated landmark detection and error flagging, improves the consistency of workflow, current AI tools require validation and human oversight to manage limitations in generalizability. When contextualized against three-dimensional imaging, LCR maintains a favorable balance of diagnostic utility and lower radiation dose, supporting its selective, indication-based use in contemporary practice. Ultimately, this review suggests that adherence to a meticulous acquisition technique remains the cornerstone of reliable LCR analysis, even as AI and digital tools evolve to augment the clinician's role.