Abstract
Introduction
Accurate age estimation plays a crucial role in medicolegal investigations, particularly in determining whether an individual has reached the age of majority for criminal responsibility, which is legally defined as 18 years in Indonesia. Geometric morphometric (GM) analysis of the mandible enables the evaluation of shape variability in two-dimensional (2D) data with potential applications in categorizing individual ages. Purpose: This study aimed to evaluate the applicability of geometric morphometric (GM) analysis of mandibular morphology on panoramic radiographs for classifying individuals as adolescents or adults.
Methods
300 digital panoramic radiographs were obtained from Airlangga University Dental Hospital in Surabaya and divided into adolescent (15.0–17.9 years) and adult (18.0–21.0 years) age groups. Each sample was assigned 27 anatomical landmarks and analyzed using MorphoJ software with generalized Procrustes analysis (GPA) and principal component analysis (PCA), while statistical evaluation included Procrustes analysis of variance (ANOVA) and discriminant function analysis (DFA).
Results
GM analysis revealed statistically significant differences in mandibular morphology between adolescents and adults. However, Procrustes ANOVA did not show significant differences in mandibular size between the age groups. Variable mandibular morphology patterns were identified at the incisor point, mental foramen, gonion, and mandibular notch. The geometric morphometric method successfully identified the mandibular morphologies specific to each group, achieving 67 % and 65 % accuracy for the adult and adolescent groups, respectively.
Conclusion
These findings underscore the potential of GM analysis of mandibular morphology for classifying individuals as adolescents or adults.
Keywords: Age classification, Forensic dentistry, Geometric morphometric, Human rights, Legal identity, Mandibular morphology
Graphical abstract
1. Introduction
The Indonesian National Police reported 584,991 criminal cases in 2023, an increase of 212,026 from the previous year.1 In many cases, age estimation plays a critical role in determining whether an individual has reached the age of majority, which is defined as 18 years under Indonesian law.2 Various approaches, including skeletal, dental, and facial analyses, have been explored to improve the accuracy of age estimation. Dental analysis is widely recognized as one of the most reliable approaches.3 However, its application can be limited in cases involving tooth loss or in the absence of antemortem dental records. In such cases, the mandible serves as a valuable alternative because of its structural robustness and resistance to postmortem degradation.4
The mandible experiences significant morphological changes during growth and development, making it a useful anatomical indicator for age estimation.5 In this regard, panoramic radiography is a standard diagnostic tool that provides reliable linear and angular measurements.6 However, traditional methods for assessing mandibular dimensions often fail to account for complex shape variations. This limitation has prompted the use of GM analysis as a more advanced approach.7
GM enables quantitative shape analysis by assigning landmark coordinates on panoramic radiographs, allowing the evaluation of shape variation patterns without being influenced by scale, position, or orientation changes. Procrustes analysis has been widely applied in this context to visualize morphological variations in detail, particularly within dental and mandibular structures.8,9 A GM study in Malaysian adults found significant shape differences of the mandible between sexes and ages.10 However, no study has been done in Indonesia using this technique for age classification. Therefore, this study aimed to examine mandibular shape variation for classifying adolescents and adults within the Indonesian population.
2. Materials and methods
This study received ethical approval from the Health Research Ethical Clearance Commission of the Dental Hospital, Airlangga University (approval number: 26/UN3.9.3/Etik/PT/2024) and was conducted in accordance with the ethical principles of the Declaration of Helsinki. A total of 300 digital panoramic radiographs of outpatients aged 15.0–21.0 years were retrospectively collected from the radiographic archives of Airlangga University Dental Hospital. The radiographs were divided into two age groups: adolescents (15.0–17.9 years) and adults (18.0–21.0 years), with 150 panoramic radiographs in each group (Table 1).
Table 1.
Sex distribution and age characteristics (mean ± SD, in years) of adolescent and adult groups included in the study.
| Groups | Sex | N | Age |
|---|---|---|---|
| Mean ± SD | |||
| Adolescent | Male | 56 | 16.09 ± 0.81 |
| (15.0–17.9) | Female | 94 | 15.95 ± 0.82 |
| Adult | Male | 41 | 18.76 ± 0.83 |
| (18.0–21.0) | Female | 109 | 19.09 ± 0.79 |
| Total | 300 | ||
Radiographs were obtained based on image clarity and complete visualization of the mandible with complete permanent teeth. The exclusion criteria comprised a history of mandibular fracture, post-fracture surgery, or any major developmental disturbance affecting the mandibular morphology. Major developmental disturbances were defined as congenital anomalies (such as cleft mandible, hemifacial microsomia, condylar hyperplasia, and mandibular hypoplasia), pathological lesions affecting mandibular growth, or extensive dental treatment/extractions that significantly altered mandibular morphology.
The panoramic radiographs, originally stored in *.jpg format, were converted into *.tps files for GM analysis using tpsUtil software (version 1.83). Anatomical landmarks were digitized using tpsDig2 (version 2.31), and subsequent analyses were performed in MorphoJ (version 1.07a, University of Manchester, UK). A total of 27 anatomical landmarks were identified on each mandible (Table 2, Fig. 1).
Table 2.
Definitions and numbering of mandibular landmarks identified on digital panoramic radiographs, adapted from Zulkifli et al.10.
| No. | Landmarks | Definition |
|---|---|---|
| 1 | Coronion | The most superior point of the coronoid process (right) |
| 2 | Mandibular notch | The most inferior point on the mandibular notch (right) |
| 3 | Condylion medial inferialis | Medial point on the mandibular condyle located in the most curved area and horizontally aligned with landmark 7 (right) |
| 4 | Condylion mediale | The most medial point on the mandibular condyle (right) |
| 5 | Condylion superior | The most superior point on the mandibular condyle (right) |
| 6 | Condylion laterale | The most lateral point on the mandibular condyle (right) |
| 7 | Condylion lateral inferialis | Lateral point on the mandibular condyle located at the most curved area, horizontally aligned with landmark 3 (right) |
| 8 | Inferior alveolar foramen | The most inferior point on the margin of the inferior alveolar foramen (right) |
| 9 | Posterior ramus | Single point located at the posterior border of the ramus, horizontally aligned with landmarks 8 and 27 (right) |
| 10 | Gonion | The most lateral external junction point of the horizontal and ascending rami of the lower jaw (right) |
| 11 | Body mandibular notch | The deepest groove in the ramus of the mandible (right) |
| 12 | Mentale | The most inferior point on the margin of the mandibular mental foramen (right) |
| 13 | Mentale | The most inferior point on the margin of the mandibular mental foramen (left) |
| 14 | Body mandibular notch | The deepest part of the mandible body (left) |
| 15 | Gonion | The most lateral external junction point of the horizontal and ascending rami of the lower jaw (left) |
| 16 | Inferior alveolar foramen | The most inferior point on the margin of the inferior alveolar foramen (left) |
| 17 | Posterior ramus | Single point located at the posterior border of the ramus, horizontally aligned with landmarks 16 and 25 (left) |
| 18 | Condylion lateral inferialis | Lateral point on the mandibular condyle located at the most curved area, horizontally aligned with landmark 22 (left) |
| 19 | Condylion laterale | The most lateral point on the mandibular condyle (left) |
| 20 | Condylion superior | The most superior point on the mandibular condyle (left) |
| 21 | Condylion mediale | The most medial point on the mandibular condyle (left) |
| 22 | Condylion medial inferialis | Medial point on mandibular condyle located at the most curved area, horizontally aligned with landmark 18 (left) |
| 23 | Mandibular notch | The most inferior point on the mandibular notch (left) |
| 24 | Coronion | The most superior point on the coronoid process (left) |
| 25 | Anterior ramus | Point at which the minimum breadth transects the anterior border of the ramus, horizontally aligned with landmarks 16 and 17 (left) |
| 26 | Incisor | The midpoint located between the two mandibular central incisors |
| 27 | Anterior ramus | Point at which the minimum breadth transects the anterior border of the ramus, horizontally aligned with landmarks 8 and 9 (right) |
Fig. 1.
Anatomical landmarks of the mandible on panoramic radiographs, with detailed definitions provided in Table 2.
The two-dimensional (2D) landmark coordinates underwent generalized Procrustes analysis (GPA) to eliminate non-shape variation from the dataset. This process aligned landmark configurations through scaling, translating, and rotating them to a standard coordinate system with a unit centroid size. This normalization enables a biologically meaningful comparison of shape differences that is independent of size, position, and orientation.
Shape variation was then analyzed using principal component analysis (PCA), which grouped and organized samples based on similarity. Morphological differences were visualized through wireframe models and principal component (PC) plots generated in MorphoJ. Classification accuracy was evaluated using discriminant function analysis (DFA) combined with cross-validation, based on PC scores obtained from the GPA/PCA results.
3. Results
The GPA analysis generated a new matrix of procrustes coordinates by superimposing each set of landmarks and standardizing them to a unit centroid size through scaling and rotation. The scatterplot in Fig. 2 illustrates the 27 landmark configurations derived from the GPA, representing mandibular morphology across 300 digital panoramic radiographs. In this plot, the black dots denote individual landmark positions, while the blue dots represent average positions. Principal Component Analysis (PCA) was then used to identify the principal components (PCs) of shape variation within the dataset. The consensus configuration, illustrated as a wireframe in Fig. 3, represents the average mandibular shape.
Fig. 2.
Scatterplot of 27 landmark configurations from 300 panoramic radiographs after Generalized Procrustes Analysis (GPA) superimposition.
Fig. 3.
Wireframe plot of the consensus mandibular configuration showing morphological landmarks (
: mandibular landmarks as listed in Table 2; 
: connecting lines between landmarks).
In GM analysis, an eigenvalue is a key concept related to PCA, which is used to summarize and visualize shape variation in the data. After landmark coordinates are aligned through GPA to remove differences in size, position, and orientation, PCA is applied to the covariance matrix. Eigenvalues quantify how much morphological variation in shape is captured by each PC axis after landmark data undergoes procrustes superimposition. The first PC has the largest eigenvalue and accounts for the most significant proportion of the total shape variation, followed by the subsequent PCs in decreasing order. In this study, the first eight PCs explained 81.8 % of the total variance (Fig. 4) in mandibular shape (PC1 = 30.7 %; PC2 = 18.1 %; PC3 = 10.7 %; PC4 = 6.4 %; PC5 = 5.5 %; PC6 = 5.1 %; PC7 = 3.2 %; PC8 = 2.1 %).
Fig. 4.
Principal Component (PC) analysis of mandibular shape based on 300 panoramic radiographs. The first eight PCs accounted for 81.8 % of the total variance.
Table 3, Table 4 present the variation in wireframe shapes across PC1 to PC3 by age group. The dark blue lines represent distinct changes in different PCs and age groups, whereas the light blue lines indicate the average configuration. Lollipop graphs (Table 5) illustrate the direction and magnitude of landmark displacement, where the rod indicates the shift and the head marks the average reference position. These visualizations highlight the differences in mandibular morphology between adolescents and adults.
Table 3.
Wireframe plots of principal components (PC1–PC3) representing 60 % of the total variance in mandibular morphology within the adolescent group.
| PCA | Wireframe of mandibular morphology |
|---|---|
| PC1 (33 %) |  |
| PC2 (17 %) |  |
| PC3 (10 %) |  |
Dark blue lines represent distinct changes in different PCs; Light blue lines indicate the average configuration.
Table 4.
Wireframe plots of principal components (PC1–PC3) explaining 57 % of the total variance in mandibular morphology within the adult group.
| PCA | Wireframe of mandibular morphology |
|---|---|
| PC1 (29 %) |  |
| PC2 (17 %) |  |
| PC3 (11 %) |  |
Dark blue lines represent distinct changes in different PCs; Light blue lines indicate the average configuration.
Table 5.
Lollipop graphs illustrating landmark displacement along principal components (PC) 1, 2, and 3.
| PCA | Anatomical displacement of landmarks |
|---|---|
| PC1 (30 %) |  |
| PC2 (18 %) |  |
| PC3 (11 %) |  |
Rod indicates the landmark shift; Head marks the average reference position of the landmarks.
Table 6 summarizes the results of the Procrustes ANOVA for centroid size and shape. Although no significant differences were observed in centroid size, statistically significant differences in centroid shape were found between age groups (p < 0.0001). The DFA results showed that 110 panoramic radiographs of adults (73 %) were correctly identified as adults, while 40 (27 %) were misclassified as adolescents. Among the adolescent group, 118 radiographs (79 %) were correctly classified, while 32 (21 %) were misclassified as adults. Cross-validation results demonstrated that 100 adult radiographs (67 %) were correctly identified, whereas 50 (33 %) were misclassified as adolescents. Ninety-eight radiographs (65 %) were accurately classified for adolescents, while 52 (37 %) were misclassified as adults (Table 7).
Table 6.
Results of Procrustes ANOVA for mandibular centroid size and shape.
| Effect | SS | MS | dF | F | p |
|---|---|---|---|---|---|
| Centroid Size | 60958473.841 | 205247.387 | 297 | 4.35 | 0.2052 |
| Shape | 0.702 | 0.00004 | 14850 | 2.07 | <0.0001a |
indicating a significant difference, sum of squares (SS), mean square (MS), degrees of freedom (dF).
Table 7.
Discriminant function analysis (DFA) results and cross-validation outcomes for adolescent and adult groups.
| DFA |
Cross-validation |
Total | |||
|---|---|---|---|---|---|
| Adult | Adolescent | Adult | Adolescent | ||
| Adult | 110 (73 %) | 40 (27 %) | 100 (67 %) | 50 (33 %) | 150 (100 %) |
| Adolescent | 32 (21 %) | 118 (79 %) | 52 (37 %) | 98 (65 %) | 150 (100 %) |
4. Discussion
This study used GM analysis to examine mandibular morphology and classify age groups of 300 digital panoramic radiographs from the Dental Hospital of Airlangga University in Surabaya, Indonesia. The radiographs were divided into adolescents and adults based on the Indonesian legal age of majority. The analysis showed significant differences in mandibular morphology between the two groups, although mandibular size did not differ considerably. Mandibular size was calculated from the centroid size in GM analysis, which is the square root of the sum of squared distances from all landmarks to the centroid. Conversely, mandibular shape was derived from the geometric configuration of 27 landmarks after Procrustes superimposition.11
Generalized Procrustes analysis (GPA) is the most common method used in landmark-based GM analysis. This approach involves aligning landmark configurations from different specimens to remove variations caused by size, position, and orientation, thus enabling shape comparison. In this study, a Procrustes coordinate matrix was created from GPA by aligning 27 landmarks across all digital panoramic radiographs. Superimposition includes translating the 27 landmarks to the same point, rotating them based on the least-squares criterion to minimize the Procrustes distance, and scaling them to the centroid size to ensure consistent orientation. GPA yields differences in mandibular configurations that are independent of the original size. The data obtained from GPA included centroid size values and Procrustes distances, which were then used for further analysis. The GPA results allowed for the creation of a scatterplot of the morphological landmarks, representing the consensus configuration or average mandibular shape of all individuals. In later analyses, this scatterplot was used to assess the distribution of mandibular landmarks among different groups.12
Variations and changes in the mandibular morphological configuration were visualized via PCA by observing changes in landmark positions in geometric space. PCA involves the eigenanalysis of the covariance matrix derived from the Procrustes coordinates obtained through GPA to generate eigenvalues. The morphological variation in PCA was determined by calculating the PC values for each sample. Based on the PCA results, PC1 explained 30.7 % of the total morphological variations in the panoramic radiographs of the adolescent and adult groups, reflecting primary differences in the mandibular morphology within the dataset.
Mandibular morphological variation in this study can be visualized by connecting anatomical landmarks into a wireframe plot and using lollipop graphs to illustrate shifts in landmark position.7 Notable changes were observed in the incisor, mentale, gonion, body mandibular notch, and condylion regions. In contrast, minimal or no variation was found in the inferior alveolar foramen, posterior ramus, and anterior ramus between the adult and adolescent groups, while the remaining landmarks exhibited moderate variation. The most significant variation (PC1) was observed in the mental foramen, as indicated by the length of the lollipop rods. In younger individuals, the mental foramen is located between the upper and lower mandibular borders, whereas with age, it shifts closer to the upper border. Asrani and Shah (2018) described four variations of the mental foramen visible on panoramic radiographs: continuous, separate, diffuse, and unidentified. In addition to age, the position of the mental foramen may also be influenced by ethnicity, race, sex, and systemic conditions such as diabetes mellitus, hyperparathyroidism, and steroid use.13,14
The gonion point also exhibited variation in the lollipop graphs, showing a tendency to decrease with age. In adults, the gonial angle at this point is located more distally compared to the consensus configuration. Similarly, the coronoid process shifted superiorly and distally, a change likely influenced by masticatory function and microstructures of the associated muscles.15 The condylar landmarks also varied, reflecting upward and downward rotation of the mandible during growth due to vertical development of the posterior region.16 The mandibular incisor showed increasing lingual inclination with age. In contrast, no age-related changes were observed in the inferior alveolar canal, consistent with the findings of Khalid et al. (2020), who reported no significant variations in the position of the canal across age or sex.17
In addition to the gonial point, differences were also identified in the location of the mental foramen. The foramen was located closer to the alveolar crest in adults, whereas it was nearer to the mandibular body's midpoint in adolescents. The mandibular condyle undergoes remodeling in response to continuous functional stimuli from childhood to adulthood. The condyle is the major center of mandible growth, and its final dimensions, shape, and volume are associated with the final relationship between the maxillary base and the mandible.18
In this study, the first three PCs accounted for 60 % of the morphological variations in the adolescent group, compared with 57 % in the adult group. This finding indicates that adolescents exhibited greater variability in mandibular morphology, reflecting the active growth phase during this period. A study by Patcas et al. (2017) reported that the peak of mandibular growth occurs around 12–14 years in girls and 14 years in boys. Significant growth differences were observed between individuals aged 16–18 and those aged 18–20, with more pronounced growth in the younger group.19 Mandibular growth tends to stabilize in individuals aged 18–20 years.16
Mandibular morphology changes throughout life, primarily due to bone remodeling driven by osteoblast and osteoclast activity. These processes result in variations in mandibular size and shape over time.6,20 GM analysis allows these variations to be statistically assessed by examining spatial landmark configurations and their covariation with other variables. In this study, 27 anatomical landmarks were digitized from panoramic radiographs, which provide a comprehensive overview of maxillomandibular morphology and are widely used in dental practice.21, 22, 23 Digitized film-based panoramic images may also be used for GM analysis, as they are generally comparable to digital radiographs for most applications. However, slight differences in visualization quality may occur. Therefore, while digitized images are suitable for morphometric studies, caution is needed to ensure that the digitization process does not introduce artifacts that could affect accuracy.24 MorphoJ, the software employed in this study, effectively visualizes morphological variation through wireframe models, although it does not directly provide quantitative measurements of these differences.25
GM analysis offers a significant advantage over traditional linear measurements by capturing complex shape information with greater statistical power. Landmarks can be mapped in two or three dimensions, enabling detailed visualization of morphological characteristics. This study applied GM analysis as a novel approach to age group determination, specifically distinguishing adolescents from adults in the Indonesian population. GM analysis presents a valuable supplementary approach for age classification, particularly in forensic age estimation. Moreover, GM methods offer the potential to establish new standards for the rapid and accurate identification of human skeletal remains in forensic contexts. Multivariate descriptors of mandibular shape and size derived from landmark configurations in 2D or 3D space may further support age estimation and sex determination in human skeletal remains.12 A study by Franklin et al. (2008) reported that GM analysis of the mandible can predict age in the subadult skeleton with accuracy comparable to dentition-based methods.26 Furthermore, this approach has demonstrated high accuracy (84 %) in sex determination using digital panoramic radiographs.27
Age estimation is important in legal and humanitarian contexts, as it verifies evidence presented in court and protects the rights of children and adolescents. Reliance solely on visual assessments may lead to inaccurate age determinations, significantly influencing legal outcomes, especially in criminal cases involving minors.28 Therefore, our study aimed to investigate the age-related progression of mandibular morphology using GM methods, particularly for legal applications in Indonesia. The findings indicate that GM analysis of the mandible can reliably classify individuals into age groups, thereby assisting law enforcement in determining legal age categories through forensic odontology using archived antemortem panoramic radiographs. These results highlight the potential of GM methods to distinguish individuals around the critical 18-year threshold defined by Indonesian law.
The common method for estimating the age of majority involves analyzing the development of third molars, particularly the mandibular third molars, due to their late development stages, which extend into early adulthood.29, 30, 31 Studies have shown that the development of mandibular third molars can be used to estimate whether an individual is above or below 18 with reasonable accuracy.32,33 Combining the analysis of mandibular morphology and third molar development provides a robust method for estimating the age of majority. While third molar development stages and mandibular morphology offer a reliable indicator, the accuracy can be enhanced by considering sexual dimorphism, population-specific data, and advanced imaging techniques.34, 35, 36
The limitation of this study is that it did not account for variables affecting mandibular growth and morphological variation, such as ethnicity, sex, or other external factors. Age-related changes in mandibular morphology are known to vary by sex, mainly due to the influence of growth hormone in stimulating bone elongation and epiphyseal maturation. Males typically experience delayed peaks in growth hormone secretion, leading to a longer period of mandibular growth than females.37,38 Ethnicity has also been shown to impact cranial growth patterns in young adults.39 Apaydin and Ozbey (2020) further indicated that ethnicity can be a significant determinant of skeletal morphology, partly affected by environmental factors like sunlight exposure.40
5. Conclusions
This study revealed clear differences in mandibular morphology between adolescent and adult groups based on geometric morphometric (GM) analysis of digital panoramic radiographs. The variation was more prominent in adolescents, aligning with ongoing mandibular growth during this stage. Key landmarks, including the incisor, mental foramen, gonion, and mandibular notch, significantly contributed to the observed morphological differences, reflecting their roles in mandibular development. However, factors such as sex and ethnicity, which may also affect mandibular morphology, should be considered in future research to improve the accuracy and applicability of these findings.
Patient's/Guardian's consent
Not applicable.
Ethical clearance
Ethics approval was obtained from the Health Research Ethical Clearance Commission of the Dental Hospital, Airlangga University (approval number: 26/UN3.9.3/Etik/PT/2024), dated July 16, 2024.
Authors’ contribution
AK, LDPL, and AC worked on conceptualization, methodology, data curation, and formal analysis. AK, LDPL, BNR, MIM, and QOAP wrote the original manuscript, reviewed and edited the manuscript. AK, AA, RAR, and AM worked on methodology, reviewed, and edited the manuscript. AK and AC supervised the study. All authors have critically reviewed and approved the final manuscript.
AI disclosure
AI tools were employed to assist in language editing, grammar correction, and refining sentence structure to improve clarity and readability. The intellectual content, study design, data interpretation, and conclusions presented in this manuscript are solely the work of the authors. All critical thinking, scientific analysis, and decision-making were performed by the authors, and no generative AI was used for data generation, analysis, or drawing scientific conclusions.
Source of funding
This research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
None.
References
- 1.Direktorat Statistik Ketahanan Sosial . fifteenth ed. Badan Pusat Statistik; Jakarta: 2024. Statistik Kriminal 2024. [Google Scholar]
- 2.Cahyaningsih D.T., Sulistyono A., Purwadi H. In: Proceedings of the International Conference on Law, Economics, and Health (ICLEH 2022) Kusumaningrum A.E., Wardani W.I., Pranoto E., Pujiyanto R., editors. Atlantis Press; 2023. Harmonization of the legal age in the context of private law as a means of achieving legal certainty and justice in Indonesia; pp. 341–354. [DOI] [Google Scholar]
- 3.Priya E. Methods of skeletal age estimation used by forensic anthropologists in adults: a review. Forensic Res Criminol Int J. 2017;4(2):41–51. doi: 10.15406/frcij.2017.04.00104. [DOI] [Google Scholar]
- 4.Kastamoni Y., Dursun A., Sanli O.C., Yuzbasioglu N., Albay N. Morphometric development of the mandible in fetal cadavers. J Craniofac Surg. 2020;31(7):2036–2039. doi: 10.1097/SCS.0000000000006670. [DOI] [PubMed] [Google Scholar]
- 5.Motawei S.M., Helaly A.M., Aboelmaaty W.M., Elmahdy K., Shabka O.A., Liu H. Length of the ramus of the mandible as an indicator of chronological age and sex: a study in a group of Egyptians. Forensic Sci Int. 2020;2 doi: 10.1016/j.fsir.2020.100066. [DOI] [Google Scholar]
- 6.Shah P.H., Venkatesh R., More C., Vaishnavee V. Age- and sex-related mandibular dimensional changes: a radiomorphometric analysis on panoramic radiographs. Indian J Dent Res. 2020;31(1):113–117. doi: 10.4103/ijdr.IJDR_327_18. [DOI] [PubMed] [Google Scholar]
- 7.Mitteroecker P., Schaefer K. Thirty years of geometric morphometrics: achievements, challenges, and the ongoing quest for biological meaningfulness. Am J Biol Anthropol. 2022;178(S74):181–210. doi: 10.1002/ajpa.24531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Johnson A., Singh S., Thomas A., Chauhan N. Geometric morphometric analysis for sex determination using lateral cephalograms in Indian population: a preliminary study. J Oral Maxillofac Pathol. 2021;25(2):364–367. doi: 10.4103/0973029X.325242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Desai P., Awatiger M.M., Angadi P. Geometrics morphometrics in craniofacial skeletal age estimation - a systematic review. J Forensic Odontostomatol. 2023;41(1):57–64. [PMC free article] [PubMed] [Google Scholar]
- 10.Zulkifli N.A.F., Mohd Saaid N.A.S., Alias A., et al. Age estimation from mandibles in Malay: a 2D geometric morphometric analysis. J Taibah Univ Med Sci. 2023;18(6):1435–1445. doi: 10.1016/j.jtumed.2023.05.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Webster M., Sheets H.D. A practical introduction to landmark based geometric morphometrics. Paleontol Soc Pap. 2010;16:163–188. doi: 10.1017/s1089332600001868. [DOI] [Google Scholar]
- 12.Manthey L., Ousley S.D. In: Statistics and Probability in Forensic Anthropology. Obertová Z., Stewart A., Cattaneo C., editors. Academic Press; 2020. Geometric morphometrics; pp. 289–298. [DOI] [Google Scholar]
- 13.Asrani V.K., Shah J.S. Mental foramen: a predictor of age and gender and guide for various procedures. J Forensic Sci Med. 2018;4(2):76–84. doi: 10.4103/jfsm.jfsm_2_18. [DOI] [Google Scholar]
- 14.Yunus M., Achmad M.H., Wulansari D.P., Haeruddin R. Characteristics vertical position of mental foramen reviewed by panoramic radiography as a guide in predicting individual's age on forensic odontology examination. Pesqui Bras Odontopediatria Clin Integr. 2018;18(1) doi: 10.4034/PBOCI.2018.181.63. [DOI] [Google Scholar]
- 15.Kathoju M., Guttikonda V.R. Age estimation using mandibular ramus and gonial angle using digital orthopantamogram. Int J Forensic Odontol. 2021;6(1):27–31. doi: 10.4103/ijfo.ijfo_5_21. [DOI] [Google Scholar]
- 16.Sharma P., Arora A., Valiathan A. Age changes of jaws and soft tissue profile. Sci World J. 2014;2014(1) doi: 10.1155/2014/301501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Khalid M., Ahmed A., Memo A.R., Manzoor F., Salman S. Anatomical location of inferior alveolar canal in different age groups in local population. Ann Pak Inst Med Sci. 2020;16(1):32–35. doi: 10.48036/apims.v16i1.35. [DOI] [Google Scholar]
- 18.Nota A., Caruso S., Ehsani S., Baldini A., Tecco S. Three dimensional volumetric analysis of mandibular condyle changes in growing subjects: a retrospective cross-sectional study. Cranio J Craniomandib Sleep Pract. 2020;38(5):320–326. doi: 10.1080/08869634.2018.1537088. [DOI] [PubMed] [Google Scholar]
- 19.Patcas R., Wiedemeier D.B., Markic G., Beit P., Keller H. Evidence of secular trend in mandibular pubertal growth. Eur J Orthod. 2017;39(6):680–685. doi: 10.1093/ejo/cjx027. [DOI] [PubMed] [Google Scholar]
- 20.Remy F., Godio-Raboutet Y., Captier G., et al. Morphometric characterization of the very young child mandibular growth pattern: what happen before and after the deciduous dentition development? Am J Phys Anthropol. 2019;170(4):496–506. doi: 10.1002/ajpa.23933. [DOI] [PubMed] [Google Scholar]
- 21.Agbolade O., Nazri A., Yaakob R., Ghani A.A., Cheah Y.K. Morphometric approach to 3D soft-tissue craniofacial analysis and classification of ethnicity, sex, and age. PLoS One. 2020;15(4) doi: 10.1371/journal.pone.0228402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Avelar L.E.T., Corradi L.M. Mandible: evaluation of its morphology, aging process and sexual dimorphism for aesthetic treatment purpose. J Dermatol Cosmetol. 2021;5(3):42–46. doi: 10.15406/jdc.2021.05.00182. [DOI] [Google Scholar]
- 23.Yazdanian M., Karami S., Tahmasebi E., et al. Dental radiographic/digital radiography technology along with biological agents in human identification. Scanning. 2022;2022(1) doi: 10.1155/2022/5265912. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Molander B., Gröndahl H.G., Ekestubbe A. Quality of film-based and digital panoramic radiography. Dentomaxillofacial Radiol. 2004;33(1):32–36. doi: 10.1259/dmfr/17777906. [DOI] [PubMed] [Google Scholar]
- 25.Klingenberg C.P. MorphoJ: an integrated software package for geometric morphometrics. Mol Ecol Resour. 2011;11(2):353–357. doi: 10.1111/j.1755-0998.2010.02924.x. [DOI] [PubMed] [Google Scholar]
- 26.Franklin D., Cardini A., O'Higgins P., Oxnard C.E., Dadour I. Mandibular morphology as an indicator of human subadult age: geometric morphometric approaches. Forensic Sci Med Pathol. 2008;4(2):91–99. doi: 10.1007/s12024-007-9015-7. [DOI] [PubMed] [Google Scholar]
- 27.Shakya T., Maharjan A., Pradhan L. Morphometric analysis of mandibular ramus for sex determination on orthopantomogram. J Nepal Health Res Counc. 2022;20(1):65–71. doi: 10.33314/jnhrc.v20i01.3822. [DOI] [PubMed] [Google Scholar]
- 28.Damascena N.P., Lima S.V.M.A., Santiago B.M., et al. Accuracy of geometric morphometrics for age estimation using frontal face photographs of children and adolescents: a promising method for forensic practice. J Forensic Leg Med. 2024;106 doi: 10.1016/j.jflm.2024.102734. [DOI] [PubMed] [Google Scholar]
- 29.Franco A., Murray J., Heng D., et al. Binary decisions of artificial intelligence to classify third molar development around the legal age thresholds of 14, 16 and 18 years. Sci Rep. 2024;14(1):4668. doi: 10.1038/s41598-024-55497-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Naik S.B., Patil S.N., Kamble S.D., Mowade T., Motghare P. Reliability of third molar development for age estimation by radiographic examination (demirjian's method) J Clin Diagn Res. 2014;8(5):ZC25–ZC28. doi: 10.7860/JCDR/2014/8160.4361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Olze A., Bilang D., Schmidt S., Wernecke K.D., Geserick G., Schmeling A. Validation of common classification systems for assessing the mineralization of third molars. Int J Leg Med. 2005;119(1):22–26. doi: 10.1007/s00414-004-0489-5. [DOI] [PubMed] [Google Scholar]
- 32.Alshihri A.M., Kruger E., Tennant M. Western Saudi adolescent age estimation utilising third molar development. Eur J Dermatol. 2014;8(3):296–301. doi: 10.4103/1305-7456.137628. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Klingberg G., Benchimol D., Berlin H., et al. How old are you? A systematic review investigating the relationship between age and mandibular third molar maturity. PLoS One. 2023;18(5) doi: 10.1371/journal.pone.0285252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Nóbrega JBM da, Protasio A.P.L., Ribeiro I.L.A., Valença A.M.G., Santiago B.M., Cameriere R. Validation of the third molar maturation index to estimate the age of criminal responsibility in Northeastern Brazil. Forensic Sci Int. 2019;304 doi: 10.1016/j.forsciint.2019.109917. [DOI] [PubMed] [Google Scholar]
- 35.Espina-Fereira A., Ortega A.I., Barrios F.A., Maldonado Y.J., Fereira J.L. Metric and angular variables of the mandibular ramus on panoramic radiographs, as indicators for chronologic age. Invest Clin. 2007;48(4):403–418. [PubMed] [Google Scholar]
- 36.Dua N., Gupta R., Jain M., Sharma A., Singh H. Effect of age, gender & dental status on mandibular morphology and Articular eminence inclination – panoramic study. J Punjab Acad Forensic Med Toxicol. 2024;24(1):70–76. doi: 10.4103/ccd.ccd_704_17. [DOI] [Google Scholar]
- 37.Lubis H.F., Simanjuntak N.U. The relationship between maxillary and mandibular lengths of ethnic Bataks of chronological age 9–15 years. Dent J. 2022;55(2):88–92. doi: 10.20473/j.djmkg.v55.i2.p88-92. [DOI] [Google Scholar]
- 38.Toneva D.H., Nikolova S.Y., Fileva N.F., Zlatareva D.K. Size and shape of human mandible: sex differences and influence of age on sex estimation accuracy. Leg Med. 2023;65 doi: 10.1016/j.legalmed.2023.102322. [DOI] [PubMed] [Google Scholar]
- 39.Hamdan S.N., Rahmat R.A., Razak F.A., Abd Kadir K.A., Abdullah E.R., Ibrahim N. Morphometric variations of three sub-adult populations in Malaysia using multi-slice computed tomography data. Folia Morphol (Warsz) 2024 Nov 15 doi: 10.5603/fm.100846. [DOI] [PubMed] [Google Scholar]
- 40.Apaydin B.K., Ozbey H. Radiographic evaluation of mandibular morphometric differences in prepubertal and postpubertal individuals in a Turkish subpopulation. J Res Dent Sci. 2020;11(1):16–20. doi: 10.4103/srmjrds.srmjrds_77_19. [DOI] [Google Scholar]