Background/Objectives: The purpose of this study is twofold: first, to construct and evaluate a deep-learning model for automated age estimation from lateral cephalograms spanning early childhood to older adulthood; and second, to determine whether sex-specific training improves predictive accuracy. Methods: This retrospective study examined 600 lateral cephalograms (ages 4–63 years; 300 female, 300 male). The images were randomly divided into five cross-validation folds, stratified by sex and age. An ImageNet-pretrained DenseNet-121 was employed for age regression. Three networks were trained: mixed-sex, female-only, and male-only. Performance was evaluated using mean absolute error (MAE) and the coefficient of determination (R²). Grad-CAM heatmaps quantified the contributions of six craniofacial regions. Duplicate patients were excluded to minimize sampling bias. Results: The mixed-sex model achieved an MAE of 2.50 ± 0.27 years, an R² of 0.84 ± 0.04, the female-only model achieved an MAE of 3.04 ± 0.37 years and an R² of 0.82 ± 0.04, and the male-only model achieved an MAE of 2.29 ± 0.27 years and an R² of 0.83 ± 0.04. Grad-CAM revealed dominant activations over the frontal bone in the mixed-sex model; the occipital bone and cervical soft tissue in the female model; and the parietal bone in the male model. Conclusions: A DenseNet-121-based analysis of lateral cephalograms can provide a clinically relevant age estimation with an error margin of approximately ±2.5 years. Using male-only model slightly improves performance metrics, and careful attention to training data distribution is crucial for broad applicability. Our findings suggest a potential contribution to forensic age estimation, growth and development research, and support for unidentified deceased individuals when dental records are unavailable.

Runt-related transcription factor 2 (RUNX2) is essential for skeletogenesis, and mutations in its gene cause cleidocranial dysplasia (CCD), an autosomal dominant skeletal disorder. The evolutionarily conserved 128-amino acid Runt homology domain (RHD) of human RUNX2 is essential for DNA binding and heterodimerization, and serves as a mutation hotspot associated with severe CCD phenotypes. To elucidate the functional impact of pathogenic RHD mutations in vivo, we generated 2 novel mouse lines: one carrying a missense mutation, c.695G>A, p.Arg232Gln (p.R232Q), corresponding to the human RUNX2 c.674G>A, p.Arg225Gln (p.R225Q), and the other harboring a frameshift mutation, c.697_698delGA, p.Glu233Thrfs*9 (p.E233Tfs*9), causing a premature stop codon. Homozygous Runx2R232Q/R232Q and Runx2E233Tfs*9/E233Tfs*9 mice lacked membranous ossification, whereas heterozygous Runx2R232Q/+ and Runx2E233Tfs*9/+ mice displayed typical CCD-like skeletal features, including an open anterior fontanelle and clavicle hypoplasia. Unexpectedly, heterozygotes carrying pathogenic RHD mutations developed small root-like protrusions, mostly 1 but rarely 2, at the pulp chamber floor of 3-rooted maxillary first molars during furcation, revealing a previously unrecognized dental phenotype. Dual luciferase assays showed that p.R232Q almost completely lost transactivation of the osteocalcin enhancer/promoter. Immunostaining showed that wild-type Runx2 was robustly expressed in osteoblasts and hypertrophic chondrocytes during bone formation, while the p.R232Q mutant Runx2 in Runx2R232Q/R232Q mice exhibited reduced expression in hypertrophic chondrocytes and partially impaired nuclear localization. These abnormalities led to defective osteoblast differentiation and chondrocyte maturation. Thus, our mutant mouse model provides a valuable in vivo platform to study CCD pathogenesis, mechanisms of tooth root furcation, and therapeutic interventions targeting dysfunctional RHD.

Runt-related transcription factor 2 (RUNX2) is essential for skeletogenesis, and mutations in its gene cause cleidocranial dysplasia (CCD), an autosomal dominant skeletal disorder. The evolutionarily conserved 128-amino acid Runt homology domain (RHD) of human RUNX2 is essential for DNA binding and heterodimerization, and serves as a mutation hotspot associated with severe CCD phenotypes. To elucidate the functional impact of pathogenic RHD mutations in vivo , we generated two novel mouse lines: one carrying a missense mutation, c.695G>A (p.R232Q) ( Runx2 ^m/+ ), corresponding to the human RUNX2 c.674G>A (p.R225Q), and the other harboring a frameshift mutation, c.697_698delGA (p.E233TfsTer9) ( Runx2 ^112/+ ), causing a premature stop codon. Homozygous Runx2 ^m/m and Runx2 ^112/112 mice lacked membranous ossification, whereas heterozygous Runx2 ^m/+ and Runx2 ^112/+ mice displayed typical CCD-like skeletal features, including an open anterior fontanelle and clavicle hypoplasia. Unexpectedly, heterozygotes carrying pathogenic mutations in RHD developed an accessory root-like protrusion at the furcation of three-rooted maxillary first molars, representing a previously unrecognized dental phenotype during root development. Dual luciferase assays revealed impaired transactivation of the p.R232Q mutant Runx2 on the osteocalcin enhancer/promoter. Wild-type Runx2 was robustly expressed in osteoblasts and hypertrophic chondrocytes during bone formation, but the mutant Runx2 exhibited reduced expression in hypertrophic chondrocytes and partially impaired nuclear localization, resulting in arrested osteoblast and chondrocyte maturation. Our mutant mouse model provides a valuable in vivo platform to study CCD pathogenesis, mechanisms of tooth root furcation, and therapeutic interventions targeting dysfunctional RHD.