Evaluation and comparison of smartphone application tracing, web based artificial intelligence tracing and...

Area of Science:

• Dentistry
• Orthodontics
• Radiology

Background:

• Cephalometric analysis is crucial for orthodontic diagnosis and treatment planning.
• Traditional hand tracing methods are time-consuming and prone to variability.
• Emerging digital tools offer potential for improved efficiency and accuracy.

Purpose of the Study:

• To compare the reliability of three cephalometric tracing methods: Smartphone Application Tracing Method CephNinja (SATM), Web Based Artificial Intelligence (AI) Driven Tracing Method WebCeph (WATM), and Conventional Hand Tracing Method (CHTM).

Main Methods:

• 110 lateral cephalometric radiographs were analyzed.
• Four linear and seven angular parameters were measured using SATM, WATM, and CHTM by a single examiner.
• Statistical analysis included Kolmogorov-Smirnov and Shapiro Wilks tests for intra-examiner differences and correlation coefficients for intra- and inter-method reliability.

Main Results:

• Statistically significant differences were found between methods for SNA, SNB, SN-MP angle, U1-SN angle, L1-NB(mm), and E Line-Upper Lip(mm) measurements (p < 0.05).
• All intra-method correlation coefficients exceeded 80%.
• Inter-method agreement varied, with the lowest coefficient between CHTM and SATM (0.170 for U1-NA) and the highest between WATM and SATM (0.884 for SNB).

Conclusions:

• The null hypothesis was rejected, indicating statistically and clinically significant differences among the cephalometric tracing methods.
• Digital methods (WATM, SATM) and conventional methods (CHTM) exhibit varying degrees of reliability and agreement.
• The choice of method can impact cephalometric measurements and subsequent clinical decisions.