Mapping Mandibular Aging in the White Population: AI-Enhanced Shape Analysis of the Lower Facial Third Bony Changes Based on 3D CT Scans of 280 Patients

BACKGROUND: The facial skeleton undergoes dynamic resorption and remodeling with aging which impacts facial shape. Despite the recognition of these skeletal alterations, sparse research has been conducted to explore facial skeletal aging in-depth. This study aims to assess the aging pattern of the mandible, which forms the lower third of the face and the transition point to the neck, and comprehensively analyze normal population across different genders and age groups. METHODS: A retrospective study was approved by our institutional review board, where head and neck Computed Tomography (CT) scans and demographic data of white ethnicity were imported and analyzed. We divided our cohort into three age groups (20:39, 40:59, and 60:79 years old) and further stratified them by gender. We excluded trauma scans and edentulous individuals or any suspicion of deformities and orthodontics interventions. 3D CT scans were reconstructed by Machine-Learning (ML)-assisted scripts in Mimics software v.25 for each subject. Statistical shape modeling (SSM) was used to create an average 3D mandible for each age group. Of a total of 280 mandibles, we conducted pair-wise analysis where 20:39 SSM was compared to 40:59 and 60:79, and 40:59 was compared to 60:79 for each gender. Contour variation was analyzed using 3D overlays and heatmaps analysis. Furthermore, semi-automated anthropometric landmarks were identified, and linear and angular measurements were calculated. RESULTS: The most pronounced bone resorption was observed across the midline and the inferior border of the mandible. The ramus height showed marked decrease with aging in both genders, aligning with the linear measurements. This resorption is more extensive in females with 3 (± 5.23) mm reduction (p < 0.001) among the 60:79 age group. In contrast, Inter-rami and inter-condylar widths increased with age with a mean 4.42 mm widening, which shows rami divergence with aging (p < 0.01). The widening was more posterior towards the condyles for females and more anterior towards the coronoid process in males. Bigonial width did not show significant changes (p = 0.54). The gonial angle increased with aging in females from 119.3 to 123.2 degrees (p < 0.01). However, the gonial angle for males does not follow a clear pattern, except for the extra bone resorption in the characteristic gonion projection (> 3 mm), giving a smoother transition to the rami with aging. Anteriorly, even with retained anterior teeth, there is a subtle inward collapse of the middle height of the mandibular body, demonstrating high-level resorption across the alveolar side. This finding was particularly notable in females, where it leads to chin protrusion; with an increase of up to 11.1 degrees (p < 0.01) in the alveolar angle. CONCLUSIONS: By leveraging AI-assisted shape analysis, we bridge the gap in aging research between historical knowledge and advanced technology. The mandibular bone resorption intensifies during the 5th and 6th decades, leading to structural changes such as rami divergence, slight alveolar inversion, and chin protrusion, particularly in females. These findings underscore the need for personalized aesthetic interventions as part of facial rejuvenation and reconstruction management based on age and gender.