Background <p>With the increasing use of cone beam computed tomography (CBCT) in dental practice, interest in evaluating the performance of dental age (DA) estimation methods using CBCT in the forensic context has increased. This retrospective study aimed to evaluate and compare the applicability and reliability of three different methods - one based on visual assessment (modified Gustafson) and two based on linear measurements (Kvaal and Drusini) - using CBCT images in a sample of the adult Turkish population.</p> Methods <p>CBCT images of 200 fully erupted mandibular second premolars from 200 patients aged 20–50 years were analysed using the Kvaal, Drusini, and modified Gustafson methods. Paired sample t-test, Wilcoxon signed-rank test, and one-way ANOVA test were used to test for statistical differences between chronological age (CA) and DA. For all three methods, the mean difference between CA and DA was calculated for each age group and both sexes. The accuracy of each technique was determined using the mean absolute error (MAE).</p> Results <p>In the total sample, the mean difference between DA and CA using the Kvaal method was found to be significant at -0.09 years, while the Drusini and modified Gustafson methods showed a higher mean difference (10.81 and 28.63 years, respectively). Pairwise comparisons between the three methods (Kvaal-Drusini, Kvaal-Gustafson, Drusini-Gustafson) revealed statistically significant differences across all age groups and both sexes (<i>p</i> &lt; 0.001).</p> Conclusions <p>Among the three methods evaluated, the Kvaal method was found to be the most reliable method in estimating DA in the total sample and in both sexes, with a lower MAE value. Despite its limitations, the Kvaal method provided more consistent and precise DA estimates than the other two techniques.</p>