A comparison of the accuracy of Willems’ and Cameriere’s methods based on panoramic radiography

A comparison of the accuracy of Willems’ and Cameriere’s methods based on panoramic radiography

Accepted Manuscript Title: A comparison of the accuracy of Willems’ and Cameriere’s methods based on panoramic radiography Authors: Neslihan Ozveren (...

NAN Sizes 0 Downloads 0 Views

Accepted Manuscript Title: A comparison of the accuracy of Willems’ and Cameriere’s methods based on panoramic radiography Authors: Neslihan Ozveren (Conceptualization) (Methodology) (Validation) (Formal analysis) (Investigation) (Resources) (Data curation) (Writing - original draft) (Writing - review and editing) (Visualization) (Supervision), Gozde Serindere (Writing - original draft) (Writing - review and editing) (Validation), Pamir Meric (Investigation) (Resources), Roberto Cameriere (Software) (Writing - review and editing) PII: DOI: Article Number:

S0379-0738(19)30324-X https://doi.org/10.1016/j.forsciint.2019.109912 109912

Reference:

FSI 109912

To appear in:

FSI

Received date: Revised date: Accepted date:

8 November 2018 3 May 2019 30 July 2019

Please cite this article as: Ozveren N, Serindere G, Meric P, Cameriere R, A comparison of the accuracy of Willems’ and Cameriere’s methods based on panoramic radiography, Forensic Science International (2019), https://doi.org/10.1016/j.forsciint.2019.109912 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A Comparison of the Accuracy of Willems’ and Cameriere’s Methods Based on Panoramic Radiography

Neslihan Ozveren1 , Gozde Serindere2 , Pamir Meric3 , Roberto Cameriere4

1

Trakya University, Faculty of Dentistry, Department of Pediatric Dentistry, Edirne, Turkey.

2

Mustafa Kemal University, Faculty of Dentistry, Department of Dentomaxillofacial,

ro of

Radiology, Hatay, Turkey. 3

Trakya University, Faculty of Dentistry, Department of Orthodontics, Edirne, Turkey.

4

University of Macerata, Institute of Legal Medicine, Macerata MC, Italy

-p

1 [email protected]

re

2 [email protected] 3 [email protected]

na

Corresponding author:

lP

4 [email protected]

Neslihan ÖZVEREN

ur

Trakya University, Faculty of

Jo

Dentistry, Department of Paediatric Dentistry, Edirne, Turkey. Email: [email protected]

Phone: +90 530 642 75 91

Highlights 

Estimating age plays a significant role in forensic odontology.



Cameriere’s method provided a better percentage of accuracy than Willems’ for Turkey population.



Combined usage of age estimation methods may help to hinder misjudgements based on dental age.

lP

re

-p

ro of

Abstract: Tooth development evaluation is majorly the current approach to identify age and maturity. Dental age (DA) is therefore considerably emphasized in forensic and pediatric dentistry. The aim of this study was to assess the applicability of Willems and Cameriere methods for Turkish children. Panoramic images of 636 Turkish healthy children (319 male; 317 female) aged between 6 and 15 years were assessed. DA of the subjects was determined by using Willems and Cameriere methods. The seven right and left permanent lower teeth were valued using these two methods. The mean chronological age (CA) of the subjects was 11.53 ± 2.25(min:6, max:14.99) years (boys: 11.29 ± 2.34 years, girls: 11.77 ± 2.15 years). Willems method overestimated the CA (p≤0.001), in contrast, Cameriere method underestimated the CA (p≤0.001) for both sexes. In different age groups the mean difference between the CA and DA calculated by Cameriere method(CA-DA) ranged from 0.86 to -0.41 years for boys and from 1.00 to -0.21 years for girls, while the mean difference between the CA and DA calculated by Willems method ranged from -0.26 to -0.71 years for boys and from 0.19 to -0.76 years for girls. Cameriere method provided a better percentage of accuracy (84.6% and 77.3% for males and females, respectively) in the absolute difference values within 1 year compared to Willems method. Cameriere method performed slightly better but, both methods may be used for DA estimation in the Turkish population.

na

Keywords: forensic science, dental age, Cameriere method, Willems method, panoramic radiography

ur

1. Introduction

Jo

Age estimation is one of the fundamental issues in some medical disciplines such as orthodontics, pediatric endocrinology, and especially forensic science. Age estimation methods provide age evaluation for legal situations. Some laws differentiate according to the age of young people [1]. Morphological parameters of teeth on dental radiographs and radiographs of hand/wrist are more confidential than the other methods and age is frequently estimated by these methods [1, 2]. Teeth are the most commonly used structure for age estimation. Various methods are available for DA estimation due to the different classifications for continuously differentiating crown and root in children [3].

It is widely accepted that there are some limitations of tooth eruption as an evaluation method. Tooth eruption is extremely affected by environmental factors such as impacted teeth, tipping, proper space in the jaw, extracted fromthe previous deciduous tooth. However, developmental stages of teeth are more convenient for dental age (DA) estimation because tooth development is less affected by environmental factors [4].

-p

ro of

A DA evaluation method was reported by Demirjian et al. in 1973. This method categorized tooth development into the eight stages based on panoramic radiographic images of the permanent teeth of left mandible. According to this method, the crown and root of tooth progressively mineralize until the apical closure is observed. The stages of teeth mineralization are matched with reference radiographic images reported by Demirjian et al. and expressed with a letter between A and H. Then values corresponding to these letters for the respective teeth are read from two tables prepared for different sexes. Values all seven teeth are summed. Finally, dental age corresponding to the total value is determined from two different tables prepared for the sexes. [5]. Willems et al. modified the Demirjian's method by creating new scoring tables based on their own patient data. Also, the sum of the values obtained from these tables gives directly dental age which makes it more practical than the Demirjian's method. [4]. In 2006, another DA estimation method was proposed by Cameriere et al. This method is based on the measurement of the ratio between the width of the tooth open apex and the length of the tooth [2].

na

lP

re

Willems and Cameriere methods had been used in several studies from different populations. In some studies, only Willems or Cameriere method was used while in some of them, both methods were used and compared. Some studies reported that Cameriere method was more accurate than Willems method [6-9]. However, others reported quite the contrary [10-14]. Genetic and environmental factors were considered to be the reason behind these conflicting results and each population may give different success rates for each method. Accurate dental age estimation and interpretation depends on the chosen methods which in turn relies on literature reports.

ur

The aim of the present study is to detect and compare the accuracy of Willems’ and Cameriere’s methods in a population of Turkish children aged 6-15.

Jo

2. Materials and Methods 2.1. Sample

The data was sourced from the Department of Pediatric Dentistry, Faculty of Dentistry, Trakya University, Edirne, Turkey. Ethical approval with the decision number: 14/07, was obtained from the Research Ethics Committee of the Trakya University Faculty of Medicine. Individuals aged between 6 and 14.99 years with sufficient quality were included in this study. Systemic diseases, previous restorative, endodontic, orthodontic treatment history, dental trauma history, dental anomalies, missing lower teeth (except the third molar), jaw bone pathologies such as cysts or tumors were selected as exclusion criteria. The panoramic

radiographs were all performed with the same radiographic device using PaX-Flex (Vatech, Seoul, Korea) at 50–90 kV and 4–10 mA with 10.1 sn exposure time. Patients’ gender, birthday and the date of their panoramic radiographs were recorded. The chronological ages of the subjects were calculated according to the date of panoramic radiographs. 2.2 .Dental age estimation method

ro of

The Cameriere’s method is clearly explained by the author himself. Shortly, the root apices of the left seven permanent mandibular teeth, excluding the third molar (numbered 1 to 7) were evaluated. The number of teeth with the closed apex (N) was noted. For the teeth that have open root apices; for the teeth that have one root (1 to 5), the distance (A) between the inner part of the open apex was measured, for teeth with two roots (6 and 7) the sum of the distances (A) between the inner surface of the two apex openings was calculated. Normalized scores (X=A/L) for each tooth were obtained by dividing these distances (A) by the tooth length (L) to consider the influence of probable variations associated with magnification and angulation among xrays.The sum of the normalized scores of teeth with open root apices (S= X1+X2+..+X7) was calculated. Consequently, dental age was calculated via the formula:

-p

Age= 8.971 + 0.375.g + 1.631.X5 + 0.674.N -1.034.S – 0.176.S.N

re

“g” variable in the formula was taken as “1” for boys and as “0” for girls [2].

2.3.Reproducibility

lP

Willems’ method iscalculated as for the Demirjian method, but the developmental score for each tooth was obtained from the tables defined by the Willems’ method, and the total development score was equal to the DA [4].

ur

na

Measurements were performed by two observers independently. Evaluation of intra-observer and inter-observer reproducibility was calculated based on a random sample of 50 radiographs after an interval of two weeks. A random sample was created by Microsoft® Excel 2013. DA was subtracted from the CA and consequently, positive results indicate age underestimation, negative results indicate age overestimation. 2.4.Statistical analysis

Jo

The Statistical Package for Social Science (SPSS), Version 20.0 was used for the statistical analysis. P values less than 0.05 were considered as significant. The mean difference between CA and DA was measured for each age and both genders. In addition to the descriptive statistics (mean value, standard deviation, and frequency), a paired t-test was used for analyses. Cohen’s Kappa test was used to verify intraobserver and interobserver agreement for Demirjian’s stages for each tooth, whereas concordance correlation coefficient was used to testintraobserver and interobserver agreement for the sum of the standardized open apices calculated for Cameriere’s method. 3. Results

Intra- and inter-observer reproducibility tests for Willems’ method resulted as almost perfect agreement according to the Landis and Koch guideline (К:0.879 for intraobserver, К:0.819 for interobserver agreement) [15]. Intra- and inter-observer reproducibility tests for Cameriere’s method resulted as excellent agreement according to the Fleiss guideline (r:0.912 to 0.944 at %95 CI for intraobserver, r:0.908 to 0.936 at %95 CI for interobserver agreement, p<0.05 for both analyses ) [16]. 636 Turkish children (319 boys; 317 girls) fulfilled the required criteria (Table 1). The mean CA of the subjects was 11.53 ± 2.25 years (boys: 11.29 ± 2.34 years, girls: 11.77 ± 2.15 years).

ro of

Table 2 gives descriptive statistics for age residuals (CA-DA) for both methods for boys and girls separately. Willems’ method overestimated the CA (p≤0.001), in contrast, Cameriere’s method underestimated the CA (p≤0.001) for both genders. Willems’ method overestimated the DA significantly higher in boys than in girls (p:0.003). Cameriere’s method underestimated the DA significantly higher in girls than in boys (p:0.001).

-p

CA-DA value percentages within ±0.25, ±0.50, ±0.75, ±1.00, ±1.25, ±1.50, ±1.75, ±2.00 and above ±2.00 were represented in Table 3. For Willems’ method, the percentages of accuracy were 74.0% for boys and 67.8% for girls in the absolute difference values within 1 year. The better percentage of accuracy (84.6% and 77.3% for boys and girls, respectively) was obtained from Cameriere’s method in the absolute difference values within 1 year.

lP

re

Table 4-5 and Fig.1 shows that for girls, the Willems’ method yielded insignificant over- and underestimations in comparisonwith CA between 6.00 and 11.99 years of age, and significant overestimation between 12.00 and 14.99 years of age. Similarly, DA was calculated with the Cameriere’s method thatindicated insignificant over- and underestimations in comparison with CA between 6.00 and 10.99 years of age but also indicated significant underestimation between 11.00 and 14.99 years of age.

ur

na

Table 4-5 and Fig.1 shows that for boys for all age groups, the Willems’ method resulted significantly overestimated DAs in comparison with the CA. The number of overestimations was higher in older age groups. DA calculated with the Cameriere’s method showed very close values to CA for ages between8.00 and 8.99, and 10.00 and 10.99. Cameriere’s method showed significant overestimations compared to CA between 6.00 and 7.99 years, and 9.00 and 9.99 years of age but also showed significant underestimation between 11.00 and 14.99 years of age.

Jo

Among different age groups, CA-Cameriere’s DA calculation gave values as low as -0.41 and as high as 0.86 for boys. For girls, these numbers were -0.21 and 1.00 respectively. CAWillems’ DA calculation gave values as low as -0.71 and as high as -0.26 for boys. For girls, these numbers were -0.76 and 0.19 respectively. The least differences between the CA and estimated DA were observed in the 10-10.99 age group both in girls and boys with both Cameriere’s and Willems’ methods. 4. Discussion

The requirement for age estimation becomes more important in forensic dentistry due to the high numbers of immigrants who arrive in a country without admissible identification papers and people with missing or indefinite birth information [2]. Due to the civil war in Syria, which began about 7 years ago, many Syrian people emigrated to Turkey as a neighboring country. According to the report in 2019, 3.613.389 refugees were living in Turkey (General Directorate of Migration Management under the Ministry of Interior, 2019). This number means about 4.5 percent of the total population of Turkey. This situation increases the importance of age estimation in our country.

ro of

According to the current Turkish Criminal Code, 12 and 15 years of ages are important steps for the children's criminal capacity. These ages are also important for imprisonment time for some crimes when the child is victim or perpetrator. Minors who are younger than 12 during the crime and deaf or mute children who are younger than 15 during the crime can’t be arrested [17]. Within this scope, determining the age for a living child of unknown identity can be required when a child is a victim or perpetrator. In these circumstances, age determination methods with high accuracy and certainty are required because of juridical necessities [18].

re

-p

Age estimation using morphological parameters of teeth on dental radiographs are considered the most convenient method for age estimation in term of a cost/benefit ratio. These methods are also reliable and simple but also there is some ethical argument because of the potential side effects of X-rays [2].

lP

This study aimed to contribute to literature answering which dental age calculation methods are applicable to children fromthe Thrace region of Turkey.

ur

na

Dental ages distribution percentages for different age groups according to critical steps in the Turkish criminal code were presented in Table 6. We see that for the age groups of twelve and older Cameriere’s method underestimates the age which may cause mistakenly low punishments when the child is guilty, or high charges when the child is a victim. In our data %37.9 of the children aged between 13 and 13.99 years were mistakenly found to be aged under twelve years by Cameriere’s method. For the age group under twelve Cameriere’s method slightly underestimates, nevertheless %17.9 of the children aged between 11 and 11.99 years were mistakenly found to be aged over twelve years by Cameriere’s method which may cause mistakenly high punishments when the child is guilty, or low charges when the child is a victim.

Jo

Willems’ method overestimates the age for the age groups of twelve and older nevertheless %17.2 of the children aged between 12 and 12.99 years were mistakenly found to be aged under twelve years by this method which may cause mistakenly low punishments when the child is guilty, or high charges when the child is a victim. Also for ages 13-14.99, Willems’ method gave dental ages over fifteen for a significant proportion of the patients which may cause higher punishments when the child is guilty, or lower charges when the child is a victim. For the age group under twelve Willems’ method overestimates especially for boys so may cause mistakenly high charges when the child is guilty, or low charges when the child is a victim. %42.5 of the children aged between 11 and 11.99 years were mistakenly found to be aged over twelve years.

The results of studies of Turkish and other populations were reviewed in Table 7. Our study showed that Willems’ method overestimates dental age. There are several studies in Europe, Middle East, and India reporting very similar results to our study [6-11]. Akkaya et al. also found that Willems’ method overestimates dental age but reported slightly better mean CA-DA value than our study [12]. Apaydın et al. and Cameriere et al. reported underestimated dental ages with Willems’ method but both studies reported mean CA-DA values which were very close to zero [13, 14].

re

-p

ro of

In the present study, Cameriere’s method provided underestimated dental ages for both sexes. There are several studies in Europe and the Middle East reporting very similar results to our study [6, 8, 19-21]. Wolf et al. also reported underestimated dental ages with Cameriere’s method but the reported mean CA-DA values which were very close to zero [22]. Apaydın et al. and Balla et al. also reported underestimated Cameriere’s dental ages but they reported higher CA-DA values than our study [11,13]. The reason for this higher age residuals may be their inclusion of 15-15.90 age group or exclusion of 6-6.99 age group as distinct from our study. It is known that Cameriere’s method doesn’t estimate age beyond 13.69 years in girls and 14.06 years in boys so studies including age groups over 14 naturally shift to underestimation in some degree [2]. Timmins et al. claimed that Demirjian method can be more useful in age estimation of older adolescents because Demirjian method goes up to 16 years [23]. Even though the expected failure of Cameriere’s method beyond 14 years of age, we included 14-14.99-year-old children in this study to detect the performance of Willems’ method on that age group.

ur

na

lP

Cameriere et al. reported underestimated and overestimated Cameriere’s dental ages in girls and boys respectively. They reported %86.9 percentage of accuracy in the absolute difference values within 1 year in the total population [14]. The same parameter was %81 in our study which was comparable to Cameriere’s results. Galic et al. reported overestimated and underestimated Cameriere’s dental ages in girls and boys respectively. They reported mean CADA values were somewhat closer to zero than our study [7]. The reason for this higher age residuals in our study may be our inclusion of 14-14.99 age group in addition to their study. Studies from Latin America reported mean CA- Cameriere’s DA values which were very close to zero for both sexes [18, 24, 25].

Jo

We report that Cameriere’s method has higher accuracy in boys. The study of El-Bakary et al. and Rivera et al. support us [6, 25]. Some of the previous studies from Europe, Middle East, Turkey, and China assert the contrary [8, 14, 19, 26]. Some others report similar accuracy for both sexes [7, 11, 13, 18, 20-22, 24]. In our study, Cameriere’s method provided a better percentage of accuracy in the absolute difference values within 1 year and also mean CA-DA values than Willems’ method for both sexes. Our study result is consonant with most of the studies from Europe and the Middle East comparing Willems’ and Cameriere’s methods [7, 8, 14, 20]. In contrast, there are some studies reporting better accuracy with Willems’ method for both sexes [11, 13, 27]. El-Bakary et al. reported slightly better accuracy with Willems’ method for girls [6]. In addition to these,

Javadinejad et al. reported that Smith’s method has better accuracy than Demirjian, Willems’, Cameriere’s methods [8].

ro of

In the studies about the accuracy of DA estimation methods in the Turkish population from different areas of the country [12, 28-31], to the best of our knowledge, the frequent result was that Demirjian’s method was not suitable for DA estimation in the Turkish population. Our previous study results supported this idea. In our previous study, we found that the Willems’ method was more accurate for almost all age groups of both genders, and was more suitable than the Demirjian method for children in the Thrace region of Turkey [30]. Because of a few studies about DA estimation using Cameriere’s method in the Turkish subpopulation, we wanted to develop our previous research. With this present study, it was reported that Cameriere’s method provided a better percentage of accuracy in the same region of Turkey. This study is one of the first studies about the accuracy of Cameriere’s method among Turkish children. So, this study will be helpful for further studies.

-p

Incorrect age estimation may have undesirable consequences. In terms of human rights, age estimation has to be for the best interest of the child. Studies evaluating different populations report different age differences and standart deviations. So comprehensive revision of the published literature for the target population is mandatory and combined usage of different dental age estimation methods may help to hinder misjudgements based on dental ages.

re

The effect of geographic and nutritional factors among different populations and population subgroups requires further researchto identify some correction factors for more accurate CA estimation.

na

lP

In conclusion, we tried to determine which of the two methods is the more suitable and accurate for northwest Turkey population. Although Cameriere’s method provided a better percentage of accuracy, both methods seem to be satisfactory for DA estimation for children in the Thrace region. When dental age assessment is required for criminal cases combined usage of Willems’ and Cameriere’s methods may be wiser.

ur

CRediT author statement

Jo

Neslihan Ozveren: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing - original draft, Writing - review & editing, Visualization, Supervision. Gozde Serindere: Writing- Original draft preparation, Writing – Review & Editing, Validation. Pamir Meric: Investigation, Resources. Roberto Cameriere: Software, Writing – Review & Editing.

ro of

-p

re

lP

na

ur

Jo

REFERENCES

Jo

ur

na

lP

re

-p

ro of

[1] R. Cameriere, L. Ferrante, Age estimation in children by measurement of carpals and epiphyses of radius and ulna and open apices in teeth: a pilot study, Forensic Sci. Int. 174(1) (2008) 60-63. [2] R. Cameriere, L. Ferrante, M. Cingolani, Age estimation in children by measurement of open apices in teeth, Int. J. Legal Med. 120(1) (2006) 49-52. [3] R. Cameriere, D. De Angelis, L. Ferrante, F. Scarpino, M. Cingolani, Age estimation in children by measurement of open apices in teeth: a European formula, Int. J. Legal Med. 121(6) (2007) 449-453. [4] G. Willems, A. Van Olmen, B. Spiessens, C. Carels, Dental age estimation in Belgian children: Demirjian's technique revisited, Journal of Forensic Science 46(4) (2001) 893-895. [5] A. Demirjian, H. Goldstein, J.M. Tanner, A new system of dental age assessment, Hum. Biol. (1973) 211-227. [6] A.A. El-Bakary, S.M. Hammad, F. Mohammed, Dental age estimation in Egyptian children, comparison between two methods, Journal of forensic and legal medicine 17(7) (2010) 363367. [7] I. Galić, M. Vodanović, R. Cameriere, E. Nakaš, E. Galić, E. Selimović, H. Brkić, Accuracy of Cameriere, Haavikko, and Willems radiographic methods on age estimation on Bosnian– Herzegovian children age groups 6–13, Int. J. Legal Med. 125(2) (2011) 315-321. [8] S. Javadinejad, H. Sekhavati, R. Ghafari, A comparison of the accuracy of four age estimation methods based on panoramic radiography of developing teeth, Journal of dental research, dental clinics, dental prospects 9(2) (2015) 72. [9] H.O. Altan, A. Altan, F. Bilgiç, Ö.A. Sözer, İ. Damlar, The applicability of Willems' method for age estimation in southern Turkish children: A preliminary study, Journal of forensic and legal medicine 38 (2016) 24-27. [10] N. Marinkovic, P. Milovanovic, M. Đuric, N. Nedeljkovic, K. Zelic, Dental maturity assessment in Serbian population: A comparison of Cameriere’s European formula and Willems’ method, Forensic Sci. Int. (2018). [11] S.B. Balla, P.V. Baghirath, B.H. Vinay, J.V. Kumar, D.G. Babu, Accuracy of methods of age estimation in predicting dental age of preadolescents in South Indian children, Journal of forensic and legal medicine 43 (2016) 21-25. [12] N. Akkaya, H.Ö. Yilanci, D. Göksülük, Applicability of Demirjian’s four methods and Willems method for age estimation in a sample of Turkish children, Leg. Med. 17(5) (2015) 355-359. [13] B. Apaydin, F. Yasar, Accuracy of the demirjian, willems and cameriere methods of estimating dental age on turkish children, Nigerian journal of clinical practice 21(3) (2018) 257. [14] R. Cameriere, L. Ferrante, H. Liversidge, J. Prieto, H. Brkic, Accuracy of age estimation in children using radiograph of developing teeth, Forensic Sci. Int. 176(2-3) (2008) 173-177. [15] J.R. Landis, G.G. Koch, The measurement of observer agreement for categorical data, Biometrics (1977) 159-174. [16] J.L. Fleiss, B. Levin, M.C. Paik, Statistical methods for rates and proportions, John Wiley & Sons2013. [17] V. İrtiş, Understanding juvenile penal justice in Turkey, The criminalisation of yought, Eds: Bailleau F, Cartuyvels Y. Brussels, Vubpress (2010) 231-55. [18] S. De Luca, S. De Giorgio, A.C. Butti, R. Biagi, M. Cingolani, R. Cameriere, Age estimation in children by measurement of open apices in tooth roots: study of a Mexican sample, Forensic Sci. Int. 221(1-3) (2012) 155. e1-155. e7.

Jo

ur

na

lP

re

-p

ro of

[19] A. Gulsahi, R.E. Tirali, S.B. Cehreli, S. De Luca, L. Ferrante, R. Cameriere, The reliability of Cameriere's method in Turkish children: a preliminary report, Forensic Sci. Int. 249 (2015) 319. e1-319. e5. [20] T. Lauc, E. Nakaš, M. Latić-Dautović, V. Džemidžić, A. Tiro, I. Rupić, M. Kostić, I. Galić, Dental Age in Orthodontic Patients with Different Skeletal Patterns, BioMed research international 2017 (2017). [21] M. Latić-Dautović, E. Nakaš, A. Jelešković, J. Cavrić, I. Galić, Cameriere’s European formula for age estimation: A study on the children in Bosnia and Herzegovina. [22] T.G. Wolf, B. Briseño-Marroquín, A. Callaway, M. Patyna, V.T. Müller, I. Willershausen, V. Ehlers, B. Willershausen, Dental age assessment in 6-to 14-year old German children: comparison of Cameriere and Demirjian methods, BMC oral health 16(1) (2016) 120. [23] K. Timmins, H. Liversidge, M. Farella, P. Herbison, J. Kieser, The usefulness of dental and cervical maturation stages in New Zealand children for disaster victim identification, Forensic science, medicine, and pathology 8(2) (2012) 101-108. [24] M.M. Fernandes, R.L.R. Tinoco, D.P.P. de Braganca, S.H.R. de Lima, L.F. Junior, E.D. Junior, Age estimation by measurements of developing teeth: accuracy of Cameriere’s method on a Brazilian sample, J. Forensic Sci. 56(6) (2011) 1616-1619. [25] M. Rivera, S. De Luca, L. Aguilar, L.A.V. Palacio, I. Galić, R. Cameriere, Measurement of open apices in tooth roots in Colombian children as a tool for human identification in asylum and criminal proceedings, Journal of forensic and legal medicine 48 (2017) 9-14. [26] Y.-c. Guo, C.-x. Yan, X.-w. Lin, H. Zhou, J.-p. Li, F. Pan, Z.-y. Zhang, L. Wei, Z. Tang, T. Chen, Age estimation in northern Chinese children by measurement of open apices in tooth roots, Int. J. Legal Med. 129(1) (2015) 179-186. [27] V. Pinchi, N. Gian-Aristide, F. Pradella, G. Vitale, D. Rugo, M. Nieri, Comparison of the applicability of four odontological methods for age estimation of the 14 years legal threshold in a sample of Italian adolescents, J. Forensic Odontostomatol. 30(2) (2012). [28] M. Celikoglu, K. Cantekin, I. Ceylan, Dental age assessment: the applicability of Demirjian method in eastern Turkish children, J. Forensic Sci. 56(s1) (2011). [29] B. Nur, A. Kusgoz, M. Bayram, M. Celikoglu, M. Nur, S. Kayipmaz, S. Yildirim, Validity of Demirjian and Nolla methods for dental age estimation for Northeastern Turkish children aged 5–16 years old, Medicina oral, patologia oral y cirugia bucal 17(5) (2012) e871. [30] N. Ozveren, G. Serindere, Comparison of the applicability of Demirjian and Willems methods for dental age estimation in children from the Thrace region, Turkey, Forensic Sci. Int. 285 (2018) 38-43. [31] E.S. Tunc, A.E. Koyuturk, Dental age assessment using Demirjian's method on northern Turkish children, Forensic Sci. Int. 175(1) (2008) 23-26.

Jo

ur

na

lP

re

-p

ro of

FIGURE 1 Mean CA-DA results with %95 Confidence intervals(CI) in ascending age groups are represented for Cameriere’s and Willems’ methods.

TABLES TABLE 1 The distribution of the subjects by age and sex. Sex Girls

Boys

Total

18

29

7-7.99

12

23

35

8-8.99

18

20

38

9-9.99

22

27

49

10-10.99 43

47

90

11-11.99 38

40

78

12-12.99 61

55

116

13-13.99 65

47

112

14-14.99 47

42

89

Total

319

636

Jo

ur

na

lP

317

-p

11

re

6-6.99

ro of

Age

TABLE 2 Mean, minimum and maximum difference values between the chronological age and the dental age (CA-DA) using the Cameriere et al. and the Willems et al. methods and p values of the differences in both sexes. CA-Cameriere’s DA

P

Min

Max

Girls

0.37(0.70)

-2.25

2.24

Boys

0.18(0.71)

-2.03

2.45

p

Mean

Min

Max

<0.001

-0.30(1.00)

-3.38 2.41

<0.001

<0.001

-0.52(0.81)

-3.22 2.05

<0.001

Jo

ur

na

lP

re

-p

ro of

Mean(SD)

CA-Willems’ DA

TABLE 3 The percentage of accuracy of dental age estimation using the Cameriere et al. and the Willems et al. methods within variated difference values from chronological age in the test group in both sexes. Willems’ Method

Girls %

Boys %

Girls %

Boys %

±0.25

24.3

28.5

21.1

22.3

±0.50

49.2

52.7

40.4

43.6

±0.75

65.6

72.1

56.5

61.4

±1.00

77.3

84.6

67.8

74.0

±1.25

81.1

91.5

78.5

±1.50

94.3

95.3

83.9

±1.75

97.5

97.8

89.0

±2.00

99.4

98.4

>±2.00

0.6

1.1

-p

80.9 89.3 91.5

93.7

95.0

6.3

5.0

re

lP na ur Jo

ro of

Cameriere’s method

TABLE 4 The differences between the mean chronologic age and calculated dental age using Cameriere’s method, and p values of the differences for different age groups for boys and girls. Sex

N

Mean(SD) Chronological Cameriere’s Age Dental Age

9-9.99

Boys

18

6.60(0.30)

7.02(0.33)

Girls

12

7.70(0.21)

7.42(0.51)

0.28(0.45)

0.053

Boys

23

7.57(0.29)

7.82(0.37)

-0.25(0.40)

0.005

Girls

18

8.57(0.26)

8.32(0.66)

0.25(0.61)

0.099

Boys

20

8.43(0.26)

8.48(0.51)

-0.06(0.41)

0.547

Girls

22

9.46(0.33)

9.67(0.72)

-0.21(0.71)

0.168

Boys

27

9.49(0.30)

9.84(0.63)

-0.35(0.54)

0.002

Girls

43

10.45(0.30)

10.46(0.61)

-0.02(0.60)

0.870

Boys

47

10.62(0.28)

10.61(0.49)

0.01(0.52)

0.918

12-12.99

13-13.99

0.153

-0.41(0.39)

<0.001

Girls

38

11.53(0.28)

11.21(0.90)

0.32(0.80)

0.017

Boys

40

11.48(0.30)

11.23(0.69)

0.24(0.71)

0.038

Girls

61

12.46(0.30)

12.21(0.72)

0.26(0.67)

0.004

Boys

55

12.44(0.29)

12.21(0.81)

0.24(0.77)

0.028

Girls

65

13.51(0.28)

12.90(0.59)

0.62(0.56)

<0.001

Boys

47

13.46(0.27)

12.96(0.82)

0.50(0.74)

<0.001

Jo

11-11.99

-0.17(0.36)

-p

6.78(0.39)

ur

10-10.99

6.61(0.27)

re

8-8.99

11

lP

7-7.99

Age Difference

Girls

na

6-6.99

P Value

ro of

Age

Girls

47

14.43(0.29)

13.43(0.47)

1.00(0.45)

<0.001

Boys

42

14.48(0.28)

13.62(0.53)

0.86(0.49)

<0.001

lP

re

-p

ro of

14-14.99

7-7.99

8-8.99

N

Mean(SD)

P Value

Chronological Age

Willems’ Dental Age

Age Difference

11

6.61(0.27)

6.52(0.57)

0.09(0.54)

0.610

Boys

18

6.60(0.30)

6.98(0.47)

-0.38(0.51)

0.006

Girls

12

7.70(0.21)

7.58(0.72)

0.13(0.65)

0.519

Boys

23

7.57(0.29)

7.96(0.58)

-0.39(0.59)

0.004

Girls

18

8.57(0.26)

8.38(0.73)

0.19(0.67)

0.245

Boys

20

8.43(0.26)

9.02(0.63)

-0.59(0.52)

<0.001

Girls

Jo

6-6.99

Sex

ur

Age

na

TABLE 5 The differences between the mean chronologic age and calculated dental age using Willems’ method, and p values of the differences for different age groups for boys and girls.

11-11.99

12-12.99

13-13.99

9.46(0.33)

9.39(0.71)

0.07(0.67)

0.658

Boys

27

9.49(0.30)

9.89(0.48)

-0.40(0.50)

<0.001

Girls

43

10.45(0.30)

10.42(0.90)

0.02(0.80)

0.850

Boys

47

10.62(0.28)

10.88(0.77)

-0.26(0.77)

0.028

Girls

38

11.53(0.28)

11.71(1.19)

-0.18(1.09)

0.329

Boys

40

11.48(0.30)

11.94(0.90)

-0.46(0.92)

0.003

Girls

61

12.46(0.30)

12.92(1.12)

-0.46(1.06)

0.001

Boys

55

12.44(0.29)

13.15(0.91)

-0.71(0.88)

<0.001

Girls

65

13.51(0.28)

14.04(1.21)

Boys

47

13.46(0.27)

14.07(1.06)

Girls

47

14.43(0.29)

15.19(1.02)

Boys

42

14.48(0.28)

15.13(0.94)

Jo

ur

na

lP

re

14-14.99

22

ro of

10-10.99

Girls

-0.53(1.15)

<0.001

-0.61(0.98)

<0.001

-0.76(0.95)

<0.001

-p

9-9.99

-0.65(0.88)

<0.001

TABLE 6 Dental ages distribution percentages for different age gorups according to critical steps in Turkish criminal code. Chronological age

DA≥12

DA≥15

DA<12

Cameriere

Willems Cameriere Willems Cameriere Willems

<9

102

-

-

%100

%100

-

-

9-9.99

49

%2

%2

%98

%98

-

-

10-10.99

90

%2.2

%7.8

%97.8

%92.2

-

-

11-11.99

78

%17.9

%42.3

%82.1

%57.7

-

-

12-12.99

116

%62.1

%82.8

%37.9

%17.2

-

%2.5

13-13.99

112

%92

%95.5

%8.0

%4.5

-

%21.4

14-14.99

89

%100

%100

-

-

-

%62.9

Jo

ur

na

lP

re

-p

ro of

N

TABLE 7 Results of previous reports using Willems’ or/and Cameriere’s methods are represented. Authors

Region(number of subjects)

Included Method of Sex Mean CAs DA differences estimation between the CA and DA (CA-DA)

Altan et al.(Altan et al. 2016)

Turkish(756)

5-14.99

Akkaya et al.(Akkaya et al. 2015)

Turkish(799)

Apaydin et al.(Apaydin and Yasar 2018)

Turkish(330)

Willems

re 8-14.99

Galic et al.(Galić et al. 2011)

Bosnian– 6-13.99 Herzegovian(1089)

ur

na

lP

Turkish(603)

Jo

Willems

Cameriere

Gulsahi et al.(Gulsahi et al. 2015)

Latic Dautovic et al.(LatićDautović et al.)

-0.26

B

-0.44

G

-0.157

B

-0.076

G

0.062

-p

5-15.90

G

ro of

2.2-16

Willems

BosnianHerzegovian(560)

Javadinejad et Iranian(537) al.(Javadinejad et al. 2015)

Cameriere

Willems

Cameriere

8-14.9

3.9014.50

Cameriere

Willems

Cameriere

B

0.056

G

0.550

B

0.603

G

0.24

B

0.47

G

-0.25

B

-0.42

G

-0.10

B

0.02

G

0.14

B

0.17

G

-0.31

B

-0.43

G

0.11

B

0.27

Croatian(776)

Lauc et al. (Lauc et al. 2017)

7-15

Willems

Cameriere

Marinkovic et al.(Marinkovic et al. 2018)

Serbian(423)

Cameriere et al.(Cameriere et al. 2008)

Italian, Spain and Croatian(756)

5-15.99

5-14.99

Willems

Willems

El-Bakary et al. (El-Bakary et al. 2010)

Egyptian(286)

5.8315.75

lP

Mexican(502)

Rivera et al.(Rivera et al. 2017)

Colombian(526)

Fernandes et al.(Fernandes et al. 2011)

Brasilian(160)

Balla et al.(Balla et al. 2016)

Indian(150)

ur

na

De Luca et al.(De Luca et al. 2012)

Jo

6-14.99

5-14.99

6-14.9

5-14.99

7-14.99

Cameriere

5-15.99

-0.57

G

0.17

B

0.19

G

-0.58

B

-0.63

G

0.081

B

0.036

G

0.073

B

-0.247

G

0.08

B

0.07

G

-0.06

B

-0.27

Cameriere

G

0.65

B

0.17

G

-0.10

B

0.00

G

0.25

B

-0.08

G

0.12

B

-0.06

G

-0.19

B

-0.41

G

0.54

B

0.63

G

0.03

Willems

Cameriere

Cameriere

Cameriere

Willems

Cameriere

Chinese(229)

B

-p

German(479)

re

Wolf et al. (Wolf et al. 2016)

-0.48

ro of

Cameriere

G

Cameriere

B

0.43

Jo

ur

na

lP

re

-p

ro of

Guo et al.(Guo et al. 2015)