Assessing observer-dependent dental age estimation procedures: intra- and inter-observer reliability across four well...

In forensic contexts, age assessments constitute matters of substantial legal consequence, particularly in proceedings involving children and young adolescents. Dental age estimation (DAE) techniques are widely used for this purpose, especially in cases involving undocumented minors. This study assesses intra- and inter-observer reliability across four well established radiographic systems for dental analysis.: Gleiser and Hunt Modified by Köhler (GHK), Demirjian (DEM), Kullman (KUL), and Cameriere's Third Molar Maturity Index (I3M). A total of 50 panoramic radiographs from individuals aged 14-23.99 years were analyzed by nine qualified forensic experts. The observers assessed the development stages of third molars using the three staging methods (GHK, DEM, KUL) and measured the I3M using Cameriere's metric approach. Primarily, the quantitative assessment for analyzing the agreement was Cohen's Kappa, Gwet's Agreement Coefficient (AC1) and (AC2), and Intraclass Correlation Coefficient (ICC). Statistical analysis revealed high intra-observer reliability for all methods, with coefficient values indicating strong agreement among individual observers. In terms of inter-observer reliability, the I3M achieved the highest agreement (ICC 0.986), followed by DEM (AC2 0.918), GHK (AC2 0.914), and KUL (AC2 0.868). Notably, maxillary third molars consistently showed lower inter-observer agreement than mandibular third molars, particularly when assessed using the DEM and GHK methods. The highest inter-observer agreement in cases where a tooth could not be staged or measured was observed for the KUL method (AC1 0.993), followed by I3M (AC1 0.988), with DEM and GHK, demonstrating equivalent levels of agreement (AC1 0.954). All of the tested methods yielded highly reliable results, especially DEM and GHK. The choice of a staging method should be guided by the specific objectives of each study. Moreover, while the I3M method demonstrated high reliability values, obtaining identical repeated measurements was nearly impossible due to its metric approach.