Evaluating Artificial Intelligence versus Manual Digital Tracing for Cephalometric Radiograph Analysis in Orthodontics

What this Trial Is About

Researchers are evaluating how well artificial intelligence AI systems perform in analyzing cephalometric radiographs compared to traditional digital manual tracing. This observational study includes 100 standardized lateral cephalometric radiographs from orthodontic patients, aiming to assess the accuracy, reproducibility, and clinical reliability of AI-based methods versus manual tracing by an experienced orthodontist. The study focuses on cephalometric parameters related to sagittal, vertical, dental, and soft tissue relationships to determine if AI can be a reliable tool for orthodontic diagnosis and treatment planning. The study uses Delta-Dent software for manual landmark identification on each radiograph, serving as the reference method. These same radiographs are then analyzed by three fully automated AI-based systems WebCeph, QuantX, and Smartee, without any manual adjustments. No therapeutic interventions or changes to patient care are involved, as the study is retrospective and observational, focusing solely on comparing measurement methods. Participants are patients with existing lateral cephalometric radiographs taken under standardized conditions. The study analyzes parameters such as SNA, SNB, ANB, and others using statistical tests to compare methods and assess agreement and clinical relevance. The research team will measure intra-rater reliability and perform comprehensive statistical evaluations to understand the consistency and accuracy of AI measurements compared to manual tracing. The study is expected to be completed by September 2026.

Comparison of Digital Analysis and Artificial Intelligence for Cephalometric Tracing