Skip to main content
NCBI home page
Acta Stomatologica Croatica logoLink to Acta Stomatologica Croatica
. 2019 Dec;53(4):307–317. doi: 10.15644/asc53/4/1

Dental Age Estimation based on Development Dental Atlas Assessment in a Child/Adolescent Population with Systemic Diseases

Cristiana Palmela Pereira 1,✉, Lucianna Maria Russell 1, Maria de Pádua Fernandes 2, Ricardo Henrique Alves da Silva 3, Rui Filipe Vargas de Sousa Santos 4
PMCID: PMC6993474  PMID: 32099256

Abstract

Background

Forensic estimation of chronological age has played an increasingly important role as part of cadaver identification, and also in living individuals due to the phenomenon of immigration and sexual abuse of undocumented trafficked children.

Objective

This research aimed to validate the already used dental mineralization and eruption atlas in normal children and adolescent population in estimating chronological age in a young population, particularly those with special needs, majority of them with systemic diseases.

Participants and Setting

A sample of 163 orthopantomograms from two independent medical institutions was collected from 133 patients aged between 4 and 23 years.

Method

The orthopantomograms were divided into two groups, 95 from patients with systemic pathologies that have repercussions on dental development and 68 with systemic pathologies without dental repercussions. Dental ages were estimated by the London Atlas using the left side and then independently the right side of the maxilla. The intraobserver and interobserver agreements were evaluated. The difference between dental age estimates and the chronological age and its absolute value were calculated and analyzed.

Results

Statistically significant differences were found between estimates and chronological age, revealing a general prevalence for underestimation; except for those under the age of 12. Nevertheless, the underestimation in individuals under the age of 16 was not significant (with an average of less than one month), while the underestimation was significant for persons who were at least 16 years old (with an average over 26 months). Furthermore, for those persons with systemic diseases with dental repercussions a greater error in underestimation was obtained, which indicates that the midpoint values should be reassessed in persons with Down’s syndrome, chromosomal alterations, syndromes and central nervous system disorders.

Conclusions

This atlas can be potentially used as a tool for age estimation in a population with special needs and, also, in a population with systemic diseases, but we suggest further studies with larger international samples to create adequate atlases for all the required scenarios, mainly, diagrams for people with systemic diseases who are over the age of 16.

Key words: Legal Age Estimation, Orthopantomography, Systemic Diseases, Forensic Dentistry

Introduction

In Forensic Sciences the estimation of chronological age has achieved an important role in forensic cases, not only in the process of cadaver identification. Namely, its importance and use in living individuals has also expanded in recent years, particularly due to the increasing phenomenon of immigration and sexual abuse of undocumented trafficked children (1-4).

Chronological aging and biological aging are quite distinct concepts; the first one relates to the time elapsed since birth, and is measured in the units of time, extending to any point in time being used in a research and development monitoring as a legal measure of grouping individuals. The second, apart from other methods, refers to bone age, dental age and sexual age; In this case, the age is being estimated from the analysis of the maturation of one or more tissues or organs, such as bones or teeth, wherein maturation is assessed by the occurrence of an event or a sequence of irreversible events which are subsequently compared to normal standards (5-7).

Dental development has several structural changes throughout life that make them good age indicators, and their mineralization and eruption are less affected by endocrine, environmental and nutritional changes, compared to bone markers, thus being extremely significant and fundamental in estimating age (8-11).

Dental radiographs have been used for dental age estimation, proving to be an important source of information for Forensic Dentistry (12). Several methods, developed by different authors, have used periapical radiographs and/or orthopantomograms (OPGs), which are less expensive and faster compared to other methods (10, 13, 14), but no method is 100% accurate mainly due to the systematic errors inherent to each method, the intra - and inter - observer variability (15) and factors related to ethnicity, socioeconomic level and pathological alterations, leading to situations of overestimation or underestimation (6, 16, 17).

The difficulty arises when it comes to individuals with special needs, be it due to a syndrome or other conditions which may modify the expected dental development (18, 19). Chromosomal disorders and syndromes arising from numerical and structural defects of chromosomes often include manifestations affecting the craniofacial region, and many of these chromosomal and multifactorial disorders have characteristic oral manifestations, such as multiple ageneses and a late eruption of the deciduous dentition and permanent dentition (6, 18, 20-22). The prevalence of births with chromosomal anomalies was 34.8 cases/10000 births, Down’s syndrome being the group of congenital anomalies with a higher prevalence (19.6 cases/10000 births) (23, 24).

There are several methods of age estimation through the use of radiography, including the Atlas of London (25) which has already been applied to several populations, including the Portuguese population (16, 26), in the normal population with no systemic diseases. In this way, the objective of this research is to validate the London Atlas by means of panoramic radiographs, in a population with both physical and mental congenital deficiencies in order to be able to estimate chronological age in a forensic context, mainly in living children and young adults.

The main objective of this research was to estimate the chronological age of a population with special needs at the University of Lisbon in two different institutions, through the application of the London Atlas based on the analysis of orthopantomograms with the scope of a future application in the forensic field. Experimental hypotheses were established, where the null hypothesis revealed the absence of statistically significant differences and the alternative hypothesis revealed the existence of statistically significant differences. Hence, the following experimental hypotheses were formulated:

  1. Difference between chronological age and estimated dental age (assessment of the accuracy of estimates)

  2. Difference in the accuracy of estimates between under and over 16 years of age.

As secondary goals, the following experimental hypotheses were established:

  1. Difference in the accuracy between left side and right side estimates

  2. Difference in the accuracy of estimates between genders

  3. Difference in the accuracy of estimates between patients with systemic diseases with dental repercussions and those without dental repercussions.

Materials and methods

The clinical records for this research from the two institutions were collected from the Pre-Graduation Clinics in Dental Medicine, Oral Hygiene, in the Department of Post-Graduation in Orthodontics and in the Service of Stomatology of Centro Hospitalar Lisboa Norte (HSM), both institutions of the University of Lisbon. The population was selected through a systematic sampling process and it comprised a minimum of 100 subjects of both genders, aged between 2.5 and 23 years, all with an OPG. The experimental protocol was evaluated and approved by the Ethics Committee of Faculty of Dental Medicine University of Lisbon (FMDUL) (Figures 1 and 2).

Figure 1.

Figure 1

Open in a new tab

Sample distribution by institution

Figure 2.

Figure 2

Open in a new tab

Sample distributions by gender

The included dental parameters refer to the presence of healthy teeth that are in the period of mineralization and eruption, since the identification parameters relate to the ages between 2.5 years and 23 years. Conversely, the exclusion dental parameters refer to the presence of extensive dental caries lesions, direct or indirect restorations, trauma or dental fracture, periodontal disease, dental rotations, dental rehabilitation with fixed prosthesis, orthodontic appliance, internal or external dental resorption, calcification, pulp fibrosis or periapical pathology and endodontic treatment. The exclusion parameters refer to the presence of dental overlaps, which interferes with the application of the London Atlas and distortion on the radiograph.

For this study a sample of 163 orthopantomograms (76 female and 87 male) from two institutions, FMDUL (112 OPGs) and HSM (51 OPGs), were collected from 133 patients (60 females and 73 males) aged between 4 and 23 years. Two dental age estimates were computed by the London Atlas for each orthopantomogram, first using only the left side (LS) of the maxilla and then using only the right side (RS) of the maxilla. Those two estimates were performed independently to avoid one estimate influence another. The intraobserver and interobserver agreements were evaluated using the intra-class correlation coefficient (ICC) (27). For intraobserver agreement, a reanalysis of 10% (n=17) of the observations of the total sample was conducted after the period of 3 months. For the interobserver agreement, all estimates were performed independently by two observers. The comparison between the use of the right side and left side of the maxilla for age estimation was performed using the variables Dif_Right_Left (difference between the estimated age based on the RS and the estimated age based on the LS, in months) and Abs_ Dif_Right_Left (absolute value of Dif_Right_Left). The variables Dif_Right_Real (difference in months between the estimated age based on the RS and the chronological age) and Abs_ Dif_Right_Real (absolute value of Dif_Right_Real) were computed to assess the accuracy of the estimates by the London Atlas using the RS. Likewise, for the LS estimates, the variables Dif_Left_Real (difference in months between the estimated age based on the LS and the chronological age) and Abs_Dif_Left_Real (absolute value Dif_Left_Real) were computed. Since the t-test should only be used in variables with normal distribution (which has not proven to be the case in this research due to the rejection of the null hypothesis in the application of the Kolmogorov-Smirnov test in all variables), the Wilcoxon’s non-parametric test was applied to compare the medians.

Results

Intraobserver and interobserver agreements

The intra-observer ICC value was 0.956 and 0.977 for the RS and LS (Figure 3), respectively, which indicates an excellent agreement, where the 95% confidence interval ranged between 0.886 and 0.984 (RS) and between 0.938 and 0.991 (LS). In the interobserver reliability, ICC for the RS is 0.923 and for the LS was 0.927 (Figure 4), with a 95 per cent confidence interval ranging between 0.895 and 0.944 (RS) and 0.9 and 0.947 (LS). These values correspond to an almost perfect agreement between the observers, with no problem at the level of the interobserver analysis. It should be noted that in some cases there were a few discrepancies between the observers’estimates. Nevertheless, in most of the cases, 116 (73.9%) on the RS and 121 (77.6%) on the LS, the difference was at most 12 months. Moreover, the mean difference between the estimates from the two observers for the right (as well as for the left) side was 4 months, revealing concordant results between the observers.

Figure 3.

Figure 3

Open in a new tab

Intraobserver agreement

Figure 4.

Figure 4

Open in a new tab

Interobserver agreement

Right side versus left side

The difference between the two estimates was on average 0.07 months. However, in one individual the difference between the two estimates was 36 months; in 7 individuals the difference between the two estimates was 24 months (in 3 the right estimate was higher than the left and the other 4 the opposite); in 40 individuals it was 12 months (in 23 the right estimate was higher than the left and in the remaining 17 the opposite was the case) and in the remaining 115 cases (70.6% of the sample) the estimates were identical. In the application of the Wilcoxon non-parametric test for comparison of medians in related samples, the test p-value was 0.864 and, therefore, there is no evidence that the median of the two estimates is different. This analysis seems to illustrate that there are no significant differences between the two estimates by right versus left sides of the maxilla in a population with systemic diseases.

Accuracy of estimates

The underestimation using the London Atlas, by the left side was on average 11.64 months. The observed maximum underestimation error was 123 months. The observed maximum error of overestimation was 67 months and the average error of estimation (in absolute value) was about 22 months. The p-value of the Wilcoxon's non-parametric test for related samples, that is, comparison of the right side estimates with the real age was equal to 0.000. Therefore, there was evidence that the median is different.

The underestimation using the London Atlas by the left side is on average 11.71 months. The observed maximum underestimation error was 123 months. The maximum error of overestimation was 67 months and the average error of estimation (in absolute value) was about 22.31 months. In the application of the Wilcoxon non-parametric test for comparison of medians in related samples, the value of the p-value was 0.000. Thus, there is evidence that the median of the left side estimates is different from the median of chronological age.

Hence, it seems that the estimates obtained on both sides are biased, i.e., a systematic error was made when using this estimation procedure.

Gender

When splitting the data in gender groups (FG versus MG) there were 87 MG and 76 FG. The results for both genders showed issues in the accuracy of estimates (Figure 2), with significant differences between estimates and chronological ages, although one p-value was slightly higher than 5%. Moreover, no significant differences were found in the non-parametric test to compare MG with the FG: p-values = 0.133 for Dif_Right_Real (MG versus FG); p-value = 0.073 for Dif_Left_Real (MG versus FG); and p-value = 0.563 for Dif_Right_Left (MG versus FG). Therefore, the results reveal no significant differences between MG and FG (Figure 5).

Figure 5.

Figure 5

Open in a new tab

Differences between genders

Age Groups

When splitting the data into age groups, substantial differences were observed between the groups of under the age of 16 (m16) with 93 individuals and at least 16-years of age (M16) with 70 individuals. There was an average error of underestimation of 0.75 months (RS) and 0.37 months (LS) in m16 and 26.10 months (RS) and 26.79 months (LS) on M16. Thus, while m16 reveals no significant differences between estimates and chronological ages, M16 clearly denotes biased estimates (Figure 6) in the application of non-parametric tests. Moreover, the London Atlas has proven to be a good method to identify patients under the age of 16.

Figure 6.

Figure 6

Open in a new tab

Differences between under and over 16 years of age

Thus, in this study, the success rate for properly classifying children under 16 years of age (those who must be protected) was 96.8% (right and left), and a considerable success rate of 75.7% for properly classified on the RS and 74.3% on the LS for individuals of at least 16 years of age.

When the data were separated into the subgroups of those up to 6 years, between 7 and 12 years, between 13 and 16 years, and at least 17 year-olds, the results of the age estimation were different. For those under 6 years of age (overestimation, on average, of 4.57 months on RS and 2.86 months on LS) and the 7 to 12 years old (overestimation, on average, of 2.02 months on both sides) the results were not biased, i.e. the differences between estimates and chronological ages were not statistically significant. For those between 13-16 years old (underestimation of on average 7.95 months on RS and 6.97 months on LS) and at least 17 years old (underestimation, on average, of 29.23 months on RS and 29.83 months on LS) the results were biased with a statistically significant underestimation error (Figure 7) (Table 1).

Figure 7.

Figure 7

Open in a new tab

Differences between different age groups

Table 1. Distribution of the sample by institution, gender and chronological age.
          HSM           FMDUL
          Gender           Gender
          Male            Female           Male
          Score           Score           Score
           Chronological age           4           0           0           1
          5           1           0           1
          6           0           0           2
          7           1           2           0
          8           5           0           1
          9           0           1           3
          10           2           3           8
           11           3           3           2
           12           1           1           3
           13           7           0           1
           14           2           2           2
           15           1           1           2
           16           2           1           4
           17           1           2           5
           18           1           1           4
           19           1           2           4
           20           0           1           5
           21           0           0           4
           22           2           0           5
           23           0           1           0
Open in a new tab

Disease

A division was made into a group where no repercussions in the mineralization and dental eruption were apparent and a group where dental repercussions were evident. The repercussions group was divided into several sub-groups where we divided the systemic diseases into Down’s syndrome, chromosomic alterations, syndromes and central nervous system (Table 2). Comparing all of the six groups, in the non-parametric hypothesis tests, we reached the following conclusions: p-value = 0.255 (right versus real); p-value = 0.579 (left versus real); p-value = 0.796 (right versus left). When comparing the medians of the three variables under analysis, the results reveal no significant differences between the six groups.

Table 2. Diagnosed Systemic Diseases (DsD).

          Down's Syndrome           Chromosomic Alterations           Syndromes           Central Nervous System           Others           Pathologies with no dental symptoms
        Chromosome 1 deletion         Cornelia de Lange syndrome         Cerebral palsy         Arthrogryposis cognitive deficits         21 Alpha hydroxybase deficiency
        Chromosome 1 alteration         De Charge syndrome         Development delay         Asperger´s         Congenital cardiopathy
Chromosome 4 deficiency         DiGeorge syndrome         Global development delay         Autism         Epilepsy
        Chromosome alteration         Dravet syndrome         Embryopathy         Cognitive impairment         Hearing impairment
        Deletion chromosome 16         Kabuki syndrome         Malformation of central nervous System         Encephalitis         Hyperactivity
        Non-identified chromosomal mutation         KGB syndrome         Mental retardation         Encephalopathy         Metabolic diseases
        Landau Kleffner syndrome         Microcephaly         Global cognitive deficit         Muscular dystrophy
        McCune syndrome         Neural impairment         Poliformative hypotonic syndrome         OTC deficiency
        Prader Willi syndrome         Psychomotor retardation         Without a closed diagnosis         Polymyositis
        Rett syndrome         Slight mental retardation         Psychological immaturity
        Rubiten syndrome         Severe atopic eczema
        Syndrome 47         Skeletal side root syndrome
        Syndrome type 1         Spina bifida
        Williams syndrome
Open in a new tab

Comparing the data from both groups (Table 3), we can verify that the average and the median of the difference between estimated and chronological age in the control group is much lower than in the group with dental repercussions, which implies that the estimates are less accurate in the group with dental repercussions, despite the minimum and the maximum estimation errors are higher in the control group. As for the difference between RS and LS, there are no significant differences. The results of the comparison of these groups were the following: p-value = 0.044 (right versus real); p-value = 0.076 (left versus real); p-value = 0.630 (right versus left). The results show a discrepancy in the estimation of right versus real and left versus real which prevents us from making a full conclusion as to whether exits significant differences in estimates accuracy between these two groups. One of the p-value is higher than the 5% significance; therefore, based only on the p-value, the null hypothesis should not be rejected for left versus real, but is rejected for right versus real. Thus, it seems that we are on the threshold of rejecting and not rejecting the null hypothesis, because it does not detect differences between the left and right estimates, nor between the left estimates and the real values, but detects significant differences between the right estimates and the real values.

Discussion

The obtained intraobserver ICC values are slightly higher than in AlQahtani (0.879) (23) and Cesário et al (0.925) (16) but lower than in Pavlović et al (26) (0.998 and 0.997 for RS and LS, correspondingly). As for Pinchi et al (6) they reported a 93% agreement.

Regarding the interobserver reliability, the obtained values are in concordance with findings of Pinchi et al (6) which have an interobserver agreement of 90%.

One of the main goals of this work was to analyze whether there are noteworthy differences between chronological and estimated age. The Atlas of AlQahtani (25) was selected as the method of estimation since, compared with other atlas methods, it is the one with the best results, as it has showed no bias (p=0.720) and correctly estimated 53% of the cases in the original thesis (16, 25).

In this study, the results reveal that there were some errors in the estimates for both RS and LS. For comparison purposes, the results from the study of Cesário et al (16), Pavlović et al (26) and Pinchi et al (6) were used. The performance test showed a statistically significant difference between age estimation and chronological age. It seems that the estimates obtained on both sides are biased, i.e., a systematic error was made (Right: author p=0.000; Pavlović p=0.104; Left: author p=0.000; Pavlović p=0.052). Thus, these results are not consistent with Pavlović et al (26), knowing that the main difference between these studies was the population at hand, i.e. patients with syndromes. Furthermore, the error was consistent between sides and no significant difference between age estimation using the RS or the LS (author p=0.864; Pavlović p=0.066) was perceived. For the author the maximum underestimation was 123 months, (Pavlović 91 months), while the maximum overestimation was 67 months (Pavlović 79 months) (26).

Another main goal was to analyze if there were any differences between those under or at least 16 years of age (m16 vs M16). In both age groups there was a tendency to underestimate in this study, although AlQahtani (25) and Cesário et al (16) came to the opposite conclusion. The average underestimation error in the age group m16 was much lower than in the opposite age group M16 (0.75 months right and 0.37 months left for m16 and 26.10 months right and 26.79 months left for M16), although the difference between sides was almost negligible (6). Furthermore, the results obtained revealed that there was no evidence that the median of the estimates is different from the median of chronological age in the under 16 years of age group (right: author p=0.532; Pavlović p=0.000; left: author p=0.859; Pavlović p=0.000) while in the M16 group the difference seems to be significant (right: author p=0.000; Pavlović p=0.105; left: author p=0.000; Pavlović p=0.161). Hence, authors obtained opposite results for these age groups. This can be explained by the fact that populations examined are distinct, and when patients with (more likely to have) slower tooth development are included, estimates should further underestimate age. It is therefore expected that when there is an overestimation, this problem is lower, while if there is an underestimation, the problem is even more serious. Due to the results obtained in the m16 and M16 groups, we have decided to perform a separate analysis wherein the sample was divided into different age groups. Out of these, in the under 6 and 7 to 12 year- old groups, the estimates were unbiased with a trend towards overestimation. However, the age groups of 13 to 16 years of age and above 17 years of age, the estimates were biased with a trend towards underestimation.

Conclusions

From the results obtained in this study the following conclusions, related to the formulated experimental hypotheses, can be drawn: Chronological versus estimated age: there were statistically significant differences, revealing that the estimates using the Atlas of AlQahtani are biased in the population with special needs. The difference between under (m16) and over (M16) 16 years of age: the estimation error in M16 is significant and much higher than in m16 where the estimation errors are not significant. Moreover, the use of the obtained estimates to identify individuals younger than 16 years old reveals a good accuracy. Thus, the Atlas of AlQahtani can be used as a method in age estimation for legal purposes, but with caution towards the 16-year threshold. There were no statistically significant differences in the accuracy of the estimates obtained using the left side and the right side. The choice of side is, therefore, irrelevant. There were no statistically significant differences in the accuracy of the estimates between genders. It seems, therefore, that the same atlas can be used for both genders. When the sample was divided into with and without dental repercussions: those persons with systemic diseases with dental manifestations had a much higher error in underestimation. Therefore, the midpoint in these cases should be clearly enlarged for persons with Down’s syndrome, chromosomic alterations, syndromes and central nervous system disorders. As for those without dental manifestations, the midpoint should also be reassessed in some ages, but the correction will not be significant.

Hence, there was a general prevalence for underestimation, except for the age groups of under the age of 6 and 7 to 12. We also confirm the fact that it is easier to classify and estimate age in patients under 16 than over 16 years of age. As these types of subjects tend to move moderately during the radiological examination, it would be of great value that the OPGs are of high quality. If possible, the use of CBCT, which provides better imaging, is recommended. Regarding the sample size, some of the groups analyzed had a very low number of subjects; therefore the conclusions that could be drawn were suggestive and preliminary. We suggest further research with larger international samples to create adequate atlases for all the required scenarios, mainly, charts for persons with systemic diseases aged approximately 16.

Acknowledgments

Fundação Nacional para a Ciência e a Tecnologia, Portugal (FCT) under the project UID/MAT/00006/2019.

Footnotes

Conflict of interest: The authors report no conflict of interest

References

  • 1.Pereira CP, Escobar CP, Santos JC. Age Estimation of Unaccompanied Minors: A Portuguese Overview. Annals of Forensic Research and Analysis. 2015;2(1):1012. [Google Scholar]
  • 2.Oliveira CR, Gomes N. Indicadores de Integração de Imigrantes - Relatório estatístico Anual. [serial on the Internet]. 2017. Available from: http://www.om.acm.gov.pt/documents/58428/383402
  • 3.Ribeiro MJ, Estrela J, Machado R. Relatório de Imigração, Fronteiras e Asilo. [serial on the Internet]. 2017. Available from: https://sefstat.sef.pt
  • 4.Sykes L, Bhayat A, Bernitz H. The Effects of the Refugee Crisis on Age Estimation Analysis over the Past 10 Years: A 16-Country Survey. Int J Environ Res Public Health. 2017. Jun 13;14(6) 10.3390/ijerph14060630 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Anastácio AC, Serras C, Vargas de Sousa SRF, Palmela Pereira C. Validation of Cameriere’s medical-legal age estimation method using seconds premolars in a Portuguese population. J Forensic Leg Med. 2018. Nov;60:30–4. 10.1016/j.jflm.2018.09.005 [DOI] [PubMed] [Google Scholar]
  • 6.Pinchi V, Bugelli V, Vitale G, Pradella F, Farese L, Focardi M. Dental age estimation in children with chromosomal syndromes. J Forensic Odontostomatol. 2018. May 30;36(1):44–52. [PMC free article] [PubMed] [Google Scholar]
  • 7.Galić I, Nakaš E, Prohić S, Selimović E, Obradović B, Petrovečki M. Dental Age Estimation Among Children Aged 5–14 Years Using the Demirjian Method in Bosnia-Herzegovina. Acta Stomatol Croat. 2010;44(1):17–25. [Google Scholar]
  • 8.Garn SM, Lewis AB, Polachek DL. Variability of Tooth Formation. J Dent Res. 1959. Jan-Feb;38(1):135–48. 10.1177/00220345590380010601 [DOI] [PubMed] [Google Scholar]
  • 9.Garn SM, Lewis AB, Kerewsky RS. Genetic, Nutritional, and Maturational Correlates of Dental Development. J Dent Res. 1965. Jan-Feb;44 SUPPL:228–42. 10.1177/00220345650440011901 [DOI] [PubMed] [Google Scholar]
  • 10.Schmeling A, Geserick G, Reisinger W, Olze A. Age estimation. Forensic Sci Int. 2007. Jan 17;165(2-3):178–81. 10.1016/j.forsciint.2006.05.016 [DOI] [PubMed] [Google Scholar]
  • 11.Elamin F, Liversidge HM. Malnutrition Has No Effect on the Timing of Human Tooth Formation. PLoS One. 2013. Aug 30;8(8):e72274. 10.1371/journal.pone.0072274 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Bagić IC, Sever N, Brkić H, Kern J. Dental Age Estimation in Children Using Orthopantomograms. Acta Stomatol Croat. 2008;42(1):11–8. [Google Scholar]
  • 13.Kvaal SI, Kolltveilr KM, Thomesen IO, Solheim T. Age estimation of adults from dental radiographs. Forensic Sci Int. 1995. Jul 28;74(3):175–85. 10.1016/0379-0738(95)01760-G [DOI] [PubMed] [Google Scholar]
  • 14.Macha M, Lamba B, Avula JSS, Muthineni S, Margana PGJS, Chitoori P. Estimation of Correlation between Chronological Age, Skeletal Age and Dental Age in Children - A Cross-sectional Study. J Clin Diagn Res. 2017. Sep;11(9):ZC01–04. 10.7860/JCDR/2017/25175.10537 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Popović ZB, Thomas JD. Assessing observer variability: a user’s guide. Cardiovasc Diagn Ther. 2017. Jun;7(3):317–24. 10.21037/cdt.2017.03.12 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Cesário C, Santos R, Pestana D, Palmela Pereira C. Medico-Legal Age Estimation in a Sub-adult Portuguese Population: Validation of Atlas Schour and Massler and London. Journal of Civil & Legal Sciences. 2017;05:196. [Google Scholar]
  • 17.Willems G. A review of the most commonly used dental age estimation techniques. J Forensic Odontostomatol. 2001. Jun;19(1):9–17. [PubMed] [Google Scholar]
  • 18.Seagriff-Curtin P, Pugliese S, Romer M. Dental considerations for individuals with Down syndrome. N Y State Dent J. 2006. Mar-Apr;72(2):33–5. [PubMed] [Google Scholar]
  • 19.Diz P, Limeres J, Salgado AF, Tomás I, Delgado LF, Vázquez E, et al. Correlation between dental maturation and chronological age in patients with cerebral palsy, mental retardation, and Down syndrome. Res Dev Disabil. 2011. Mar-Apr;32(2):808–17. 10.1016/j.ridd.2010.10.019 [DOI] [PubMed] [Google Scholar]
  • 20.Suri S, Tompson BD, Atenafu E. Prevalence and patterns of permanent tooth agenesis in Down syndrome and their association with craniofacial morphology. Angle Orthod. 2011. Mar;81(2):260–9. 10.2319/070910-391.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.de Moraes ME, de Moraes LC, Cardoso M, Ursi W, Lopes SL. Age assessment based on dental calcification in individuals with Down syndrome. Res Dev Disabil. 2013. Nov;34(11):4274–9. 10.1016/j.ridd.2013.09.005 [DOI] [PubMed] [Google Scholar]
  • 22.Patil S, Rao RS, Majumdar B. Chromosomal and Multifactorial Genetic Disorders with Oral Manifestations. J Int Oral Health. 2014;6(5):118–25. [PMC free article] [PubMed] [Google Scholar]
  • 23.Cuoghi OA, Topolski F, Perciliano de Faria L, Occhiena CM, Ferreira ND, et al. Prevalence of Dental Anomalies in Permanent Dentition of Brazilian Individuals with Down Syndrome. J Int Oral Health. 2014. Sep;6(5):118–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Braz P, Machado A, Dias CM. Ministério da Saúde - Registo Nacional de Anomalias Congénitas: INSA-IP Lisboa. [serial on the Internet] 2014/2015. Available from: http://repositorio.insa.pt/bitstream/10400.18/5382/5
  • 25.AlQahtani SJ. The London Atlas : developing an atlas of tooth development and testing its quality and performance measures. [serial on the Internet] 2012. Available from: http://qmro.qmul.ac.uk/jspui/handle/123456789/3363
  • 26.Pavlović S, Pereira CP, Santos R. Age estimation in Portuguese population: The application of the London atlas of tooth development and eruption. Forensic Sci Int. 2017. Mar;272:97–103. 10.1016/j.forsciint.2017.01.011 [DOI] [PubMed] [Google Scholar]
  • 27.Fleiss JL. The design and analysis of clinical experiments. Wiley, 1999. pp. 228–242. [Google Scholar]

Articles from Acta Stomatologica Croatica are provided here courtesy of University of Zagreb: School of Dental Medicine

RESOURCES