SciELO - Scientific Electronic Library Online

 
vol.26 issue2SOME FACTORS ASSOCIATED TO COSMETIC DENTISTRY: A NEW APPROACHRADIOGRAPHIC DESCRIPTION OF TITANIUM AND FIBERGLASS POSTS CEMENTED IN HUMAN PREMOLARS SUBJECTED TO HIGH TEMPERATURES IN VITRO FOR FORENSIC PURPOSES author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Revista Facultad de Odontología Universidad de Antioquia

Print version ISSN 0121-246X

Rev Fac Odontol Univ Antioq vol.26 no.2 Medellín Jan./June 2015

 

ORIGINAL ARTICLES DERIVED FROM RESEARCH

 

IN VITRO EVALUATION OF THE EFFECT OF HYDROFLUORIC ACID ONCENTRATION AND APPLICATION TIME ON ADHESION TO LITHIUM DISILICATE

 

 

Nancy Esperanza Castro Guevara1; Joao Víctor Muñóz Durán2; Luis Alberto López Pérez3; Clementina Infante Contreras4

 

1 Candidate to the MSc in Dentistry, Universidad Nacional de Colombia School of Dentistry. DMD, Orthodontics Specialist. Email address: necastrog@unal.edu.co
2 PhD in Biology. Associate Professor, Department of Biology, Universidad Nacional de Colombia School of Sciences
3 PhD in Statistics. Professor, Department of Statistics, Universidad Nacional de Colombia School of Sciences
4 DMD. Orthodontics Specialist, Statistics Specialist. Professor, Universidad Nacional de Colombia School of Dentistry

 

SUBMITTED: AUGUST 13/2013-ACCEPTED: SEPTEMBER 2/2014

 

Castro NE, Muñoz JV, López LA, Infante C. Variation of craniofacial morphological patterns in Class I, II, and III skeletal relationships. Rev Fac Odontol Univ Antioq 2015; 26(2): 292-313.

 

 


ABSTRACT

INTRODUCTION: the studies on morphological variations of craniofacial components to classify skeletal relationships have traditionally included univariate and multivariate analysis using variables such as distances, angles, and reference planes. However, these methods fail to explain general changes in shape and provide partial localized descriptions of these relationships. Whereas methods using two- or threedimensional (2D or 3D) Geometric Morphometrics (GM) allow a detailed understanding and a more sensitive test of variables. The objective of this study was to identify morphological pattern variations of the Overall Craniofacial Structure (OCS) in skeletal relationships I, II, and III using GM-2D.
METHODS:this was a prospective study using non-probability sampling. It implied taking 272 lateral radiographs of the head of Colombian individuals (140 males/132 females) aged 17 to 25 years, determining intra-examiner error and using F-ANOVA as statistic test. Generalized Procrustes Analysis (GPA) was conducted as well as atypical data detection by Adaptive Quantile. Size variation was analyzed by the Kruskal-Wallis test considering Centroid Size matrix (CS) and conformational differences were analyzed with MANOVA. Craniofacial patterns were identified by Principal Components Analysis (PCA) and K-means/cluster.
RESULTS: the OCS showed conformational differences and a good classification capacity of 89% (Class I), 89% (Class II), and 91% (Class III). Four craniofacial patterns were identified; three of them showed typical skeletal relationships and the other pointed out to a new Class I/II combined group.
CONCLUSIONS:the morphological differences in the four identified patterns were evident; GM allowed an explanatory display of morphological variation patterns, identifying actual sites where changes in size and shape take place.

Key words: geometric morphometrics, anatomical landmark, principal component analysis, Procrustes analysis, morphology, biometrics, discriminant analysis, cephalometrics, cluster analysis.


 

 

INTRODUCTION

Development of the craniofacial complex is the result of the coordinated interaction between morphogenesis and growth of three cranial components: neurocranium, face, and masticatory apparatus.1-3 The amount of variation due to structural changes within and among these cranial components4-6 is relevant and informative concerning the regulation of normal development7-9 as well as its influences on final morphology in the presence of skeletal Classes I, II, and III. 10-12

Similarly, the craniofacial pattern in these skeletal relationships is influenced by the position of the cranial base,1, 13 as Enlow14, 16 explains in his analysis of counterparts. Each of these morphological patterns has a typical set of craniofacial characteristics that predispose to the development of a specific skeletal relationship; thus, dolichocephalic patterns are associated with Class II, while brachiocephalic patterns are associated with Class III.13, 14, 17

The studies on such craniofacial relationships have traditionally used univariate and multivariate statistical methods of association18, 19 by means of cephalometric variables and linear and angular measurements,20-22 known in the literature as Conventional Cephalometric Analysis (CCA). However, this does not explain general changes in shape and provides only a partial localized description of it. CCA presents some difficulty in classifying individuals based on subgroups of points—whose selection is prone to bias, integrating the common referential plane—,23, 24 since it does not allow assessing the combined covariance of all the variables that help identify structures.25

In the 1960s, Enlow14, 15, 16 introduced the analysis of counterparts, including the study of key morphological conditions in each individual, by observing and describing the association of the anatomical components of the various structures, which led him to identify craniofacial patterns.

The 1980s witnessed the development of coordinate-based methods as well as the statistical theory of shape, which allowed exploring and visualizing data analysis supported on exact statistical tests based on resampling procedures. This new approach is known as Geometric Morphometrics (GM), a method that keeps the geometry of landmark settings27 during the entire analysis, which allows representing statistical results in a way that is closest to reality.

With GM, landmarks are transformed into a 2D or 3D plane, forming matrices that represent geometric configurations of the structures under study. This information is processed by the Generalized Procrustes Analysis (GPA),28 which provides a referential geometric setting (consensus)27 as well as the variables that incorporate deviations from each configuration with respect to the consensus: Partial Warps (PW) and Relative Warps (RW), which are not affected by changes in size, position and orientation of configurations in space, and contain all the information to be used in the multivariate statistical analysis (PCA, cluster analysis, and discriminant analysis).29

The general objective of this study was to identify morphological pattern variation of the Overall Craniofacial Structure (OCS) in skeletal Class I, II, and III, using GM 2D. Three specific objectives were set out for this purpose: 1. Characterize differences in size and OCS conformation in skeletal Class I, II, and III, 2. Classify individuals within the groups established for Class I, II, and III, and 3. Describe the patterns of craniofacial conformation in the identified skeletal relationships.

 

METHODS

Following a non-probability sampling prospective study, 272 lateral head radiographs were taken to Colombian individuals (140 males and 132 females) aged 17 to 25 years, who were attending Universidad Nacional de Colombia at Bogotá and required orthodontic treatment, during the periods of 2011-II, 2012-I and 2012-II, distributed by sex and skeletal relationship (table 1) . These individuals had good overall health, permanent full dentition with or without third molars, and parents and grandparents from similar geographical areas. Individuals were excluded if they had history of oral rehabilitation, orthodontic treatment, orthopedics or either esthetic or orthognathic surgery, severe parafunctional habits, and acquired or congenital malformations. Each selected patient was provided with an informed consent and a booklet containing information on the project. This study adhered to Resolution No. 008430 of 1993 of Act 84 of 1989 to comply with Article 8 (Title II, Chapter 1). The study was approved by the Ethics Committee of Universidad Nacional de Colombia School of Dentistry.

Profile radiographs were taken with one X550 model, EX-1 type Veraviewepocs x-rays equipment by J Morita, with Class II laser positioner for patient location. The software used for image processing was I-Dixel by Morita Mfg. Corp. A metric witness was included to obtain 100% images (1:1). The radiographs were taken at a standard 15 cm subject distance from film and 150 cm cone from film. The equipment operated at 80 kw, 3mA and 4s per x-ray.

Radiographs were classified by CCA (the following angles were measured: SNA, SNB, ANB maxillomandibular, incisor-maxillary, and IMPA)30 and GM, considering 4 landmarks (S, N, A, and B) and using the following software: Metronukak for CCA and Matlab for GM.31 The ANB angle was used, converting its values from radians to degrees for classification purposes. Similarly, mathematical subtraction was made between the SNA and SNB angles in order to establish the positive or negative value of this angle. The information collected through these methods was distributed in three groups: Class I ANB (0, 4°), Class II (> 4 °) and Class III (< 0 °) (table 1). A number of 47 individuals changed classification (17.21%) between one method and the other.

Morphogeometric Analysis

To achieve the first objective, OCS geometry was obtained through landmarks in three steps: 1) selection of 14 Class I, II and III landmarks27 obtained using the following criteria: homology, relative position consistency, proper shape coverage, and repeatability28 (figure 1); OCS was defined using specific variables of Facial Middle Third (FMT), Cranial Base, and Mandible; 2) Collection of information that describes the conformation: 14 landmarks were digitized by the tpsDig2 software. Based on the obtained data, the .tps files were created to configure the landmark corresponding to the described OCS, by means of the tpsUtil software.34 The standard GM procedures were performed, namely GPA and Thin Plate Spline (TPS); the first GPA was done with the MOGwin software,35 which was used to obtain the PW and CS matrices in separate files.27 Completion of these steps ensures the repeatability of the procedure.

The second objective was achieved by evaluating the morphologic OCS of four new individuals a priori classified by ACC, using GM 2D and classifying by Discriminant Function based on Mahalanobis Distances (MD). The four new individuals' correspondence with Class I, II and III skeletal groups was evaluated, analyzing the classificatory power of OCS in these groups.

To achieve the third objective, Principal Components Analysis (PCA) was conducted on the PW matrix of OCS, using the Broken Stick and Jollife Cut-Off criteria36 as indicative of the number of components to use. The diagram of values helped in deciding the number of axes that were used for classification.37

A cluster analysis38 was conducted using the K-means partition algorithm. The groups obtained by this method were validated using discriminant analysis. Visualization and analysis of the morphological changes within each group was done by using the TPS function with the tpsRelw software,39 which allowed describing patterns of craniofacial conformation in the identified skeletal relationships.

Statistical analysis

All the statistical tests were reported considering a significance level lower than 5%. To measure intraobserver error, 14 craniofacial landmarks were used in 12% of the radiographs (n = 32), randomly taken. One of the researchers conducted a repeated digitalization of the landmarks with one week in difference. A repeatability analysis was conducted with the VarWin software35, 40 on the PW matrix, taking ANOVA F as statistic test. Coordinates with a high repeatability percentage were included in the analysis as illustrated in figure 1. The detection of atypical data was conducted on PW matrices, having the adaptive quantile profile proposed by Filzmoser (n = 273)41 as exclusion criterion. Analysis of the sample's normality and homoscedasticity was made by Shapiro Wilks Multivariate test and Box's M test, respectively.42

Sexual dimorphism analysis

The Centroid Size (CS) matrix was used to compare structures size. Size variation due to sexual dimorphism was evaluated with the t-test, comparing average CS between males and females for skeletal Class I, II, and III. The conformation variation due to sexual dimorphism was evaluated with multivariate analysis of hypotheses verification, for possible morphological separations between groups with non-parametric tests on permutations (n = 1000), based on MD.43

Analysis of OCS size and conformation in Classes I, II, and III

Size variation among Classes I, II, and III without considering sex as a variable was explored by means of Box-Plot. It was statistically evaluated with the Kruskall-Wallis test, using the PAS software.42 To evaluate conformation variation, the Manova multivariate variance analysis was used. Nonparametric tests were made on permutations (n = 272) based on the MD's.40, 44 , 45 This test was applied the Bonferroni correction (p < 0,05). The MD's also made it possible to conduct a reclassification, which was done by cross checking. This procedure was carried out with the PAD software.46

Method error. The repeatability analysis showed maximum variation due to the digitalization (intraobserver error), lower than 2.04% in the X axis and 3.43% in the Y axis of landmark 12 (Pogonion), showing an accuracy of 97.9% in its X component and 96.5% in its Y component. Therefore, this landmark was treated as a slider in the GPA. For details on the results, see Castro.33 An atypical case was identified due to its extreme morphological characteristics, and it was excluded from the analysis, which was conducted on (n = 272).

 

RESULTS

The sample of 272 lateral head radiographs of Colombian individuals (140 males and 132 females) aged 17 to 25 years yielded the following findings about sexual dimorphism and OCS size and conformation in Class I, II, and III. Since the data did not prove to result from normal (p< 0,05) or homoscedastic (p< 0.01) distribution, the nonparametric tests were then conducted.

Sexual dimorphism

The t-test showed that the average CS in males (1401, 1427 and 1416) was greater than in females (1301, 1296 and 1299) in Class I, II, and III respectively, and sexual dimorphism by size is evident (figure 2). Concerning OCS conformation, sexual dimorphism was significant (p = 0.0001, 0.0002, and 0.0039) respectively for Class I, II, and III, although there was an overlap between sexes as displayed in figure 3. Classification validated by discriminant analysis of Class I individuals yielded 76% in both females and males—these being the highest classification percentages among the three groups—. In Class II it yielded 65 and 76% for females and males respectively, and in Class III 63 and 58%.

OCS size and conformation in Classes I, II, and III

Regarding size comparison between skeletal Classes I, II, and III, the Kruskall-Wallis test did not yield significant CS differences among the three groups: between I and II (p = 0.8806), I and III (p = 0.4963), and II and III (p = 0.4423).

A remarkable result concerning conformation variation among Classes I, II, and III is shown in figure 4: the distances between groups I-II (2.56) and I-III (3.45) may well be half the distance between groups II-III (5.92). This allows us to deduce that groups II and III are quite different in terms of conformation.

Allocating one individual to one of the three previously defined groups, taking into account OCS conformational characteristics, was made by analysis of permutations, obtaining high classification percentages: 93, 95, and 97% respectively for each group —confirmed by cross-checking: 89, 89, and 91%—. The capacity of classification of new individuals was 100% efficient.

Patterns of craniofacial conformation: cluster analysis by K-means on the 4 main components of PW matrices yielded 4 groups: 1. Mainly consisting of Class II and a few class III individuals; 2. Consisting of Class III individuals; 3. Includes Class I and II individuals who were treated as independent groups to analyze similarities and differences between these two clusters, while exploring why Class II individuals are associated with this Class I group, and 4. Composed by Class I individuals (figure 5).

The discriminant analysis confirmed the classification of individuals within the previously established groups, demonstrating a better discrimination in groups 1 and 2, while groups 3 and 4 showed overlap with group 1, as shown in figure 6.

The highest classification percentage occurred in group 2, correctly classifying 63 of 69 individuals, with 91% accuracy; the lowest percentage was found in group 3 (mixed), which only classified 42 of 72 individuals correctly, with 58% accuracy. The results of this classification are summarized in table 2.

In the graphical exploration of OCS, the consensus distribution of the 4 groups in the shape field showed a continuum in which the consensus of group 4 is located between the consensus of groups 1 and 2 (figure 7). GPA of each group's consensus of proved that FMT showed a more retrusive position in group 2 (Class III), while group 1 (Class II) showed a more protrusive position.

The variations in conformation are expressed with the TPS function—an analysis that was performed in each group—. The factorial map associated to the TPS grid explored variations in conformation within each group in the shape field, and it also showed an estimation of the conformation of an individual in different spots of this field. Eight individuals from each group were chosen for this analysis, in order to observe OCS morphological patterns.

Group 1 factorial map (figure 8a) showed a distribution of individuals in the shape field that graphically corroborates this composition: the left quadrants showed a high concentration of individuals with Class II characteristics, while individuals with Class III conformations were located in the lower right quadrant. The grid in figure 8b showed a first variation related to anterior facial height, and the second variation contrasts posterior facial height with FMT prognathism.

The correspondence among these variations is as follows: in RW1, posterior facial height increase is associated with FMT contraction in the sagittal plane, while FMT expansion causes posterior facial height decrease (individuals 32 and 47). The variation shown in RW2 occurs due t o changes in anterior facial height. As this height increases, FMT expands, and as this height decreases, FMT starts to contract (individuals 44 and 35). This correspondence between verticality and sagittality was common to all groups.

The morphological examination of this group also showed that there was global variation in mandibular conformation, with a compressive tendency of the grid in this area.

In group 2, the consensus showed well-defined characteristics; FMT behavior continues experiencing association with both anterior and posterior facial heights, although its projection is in general low for this group, compared with that in group 1.

In group 3a (Class II predominance), FMT behavior continues experiencing association with anterior and posterior facial heights. The differences with group 1, which comprises Class II individuals also, are the grid compressions in this group's lower third. Concerning 3b (Class I predominance), the main difference with group 3a are the grid expansions in the lower third.

 

DISCUSSION

GM was used taking into account that CCA depends on referential planes and the interpretation of results varies according to the direction of such planes.23, 25 GM offers the possibility of performing multivariate analysis, improving the observation of results and offering additional morphological aspects that are not usually captured by CCA.28

Concerning the subjects who switched groups during the initial classification process, they showed dentoalveolar and vertical compensations that, despite being registered by ACC, do not provide conclusive classification. As for exploration of sexual dimorphism in OCS conformation, there were significant differences in terms of the size of males and females. These findings were reported in previous studies.45, 47 An important aspect described in the literature is that the shape variables that best differentiate the sexes are mainly those related to facial width and cranial vault13—aspects that cannot be evaluated using lateral views—.

There was a continuous spectrum of craniofacial patterns that could be described as from dolichofacial to brachiofacial, in which vertical facial variation is the result of the interaction of craniofacial components according to ontogenetic settings and functional and structural conditions that directly influence these anterior and posterior vertical variations. These results are consistent with those reported by Enlow15, 16, 26 and Bastir.11

The influence of anterior and posterior vertical height on the sagittal behavior of FMT and mandible, as well as their relationships in the conformation of skeletal Classes II and III identified in this study are consistent with the findings by Enlow26 and Akimoto et al48 (figure 9).

(figura 9)

shows the sagittal behavior with vertical heights variation in a): Middle Cranial Fossa (a) produces a protrusive effect on FMT (b) produces a lower location of it with respect to the condyle, causing a downward rotation of the mandibular ramus (c) this favors a retrusive effect in the mandibular body (d) a similar effect on the mandible occurs when the posterior height of the nasomaxilar complex increases (e), and in b): the vertical-most Middle Cranial Fossa (g) causes a higher and more retrusive position of the FMT (h) causing an upward and forward rotation of the mandibular ramus (j) producing a protrusive effect on the mandibular body (k), when the FMT posterior height decreases, it causes a similar effect on mandible (m).

The discriminant analysis identified overlapping between Class I and Class II groups. Enlow26 reports similar results, in which Class I and Class II individuals share similar characteristics, such as middle cranial fossa inclination and posterior maxillary height, which are the basis for defining Class II patterns or tendencies to such patterns in Class I individuals.

Interactions among the craniofacial components explored in this study are mainly related with changes in shape rather than with relative position, agreeing with results reported by Singh and Harvati.49

The identification of four groups with defined skeletal characteristics, based on the analysis of OSC conformation, suggest that, while CCA offers important information on craniofacial morphology, it cannot extract all information from shape.

 

CONCLUSIONS

  • OCS showed significant conformational differences among skeletal relationships I, II, and III.
  • There was continuous variation among skeletal relationships I, II, and III.
  • Conformational differences were identified in four groups, suggesting explanatory patterns. GM facilitates the interpretation of quantitative and qualitative results in conformational variations.
  • Class I and Class II groups share conformation al characteristics that are difficult to identify with CCA. MG allows a more accurate analysis of such characteristics.

Application and recommendation of the study

This study offers the possibility of researching the conformation of the various craniofacial areas in 2D lateral images, usually present in large samples for the morphogeometric study of facial growth, the response of growth in children and adolescents undergoing orthodontic treatment, and pre and postsurgery evaluation. The methodology proposed in this article is recommended in diagnostic procedures.

 

CONFLICTS OF INTEREST

The authors declare not having conflicts of interest.

 

REFERENCIAS

1. O'Higgins P, Bastir M, Kupczik K. Shaping the human face. Int Congr 2006; 1296: 55-73.         [ Links ]

2. Bastir M, Rosas A. Hierarchical nature of morphological integration and modularity in the human posterior face. Am J Phys Anthropol 2005; 128(1): 26-34.         [ Links ]

3. Szabo-Rogers HL, Smithers LE, Yakob W, Liu KJ. New directions in craniofacial morphogenesis. Dev Biol 2010; 341(1): 84-94.         [ Links ]

4. Bookstein FL. The geometry of craniofacial growth invariants. Am J Orthod 1983; 83(3): 221-234.         [ Links ]

5.Sardi ML, Ramirez Rozzi FV. A cross-sectional study of human craniofacial growth. Ann Hum Bio 2005; 32(3): 390-396.         [ Links ]

6. Pucciarelli HM, Ramirez Rozzi FV, Muñe MC, Sardi ML. Variation of functional cranial components in six Anthropoidea species. Zoology (Jena) 2006; 109(3): 231- 243.         [ Links ]

7. Todd JT, Mark LS. Issues related to the prediction of craniofacial growth. Am J Orthod 1981; 79(1): 63-80.         [ Links ]

8. Moss ML, Skalak R, Dasgupta G, Vilmann H. Space, time, and space-time in craniofacial growth. Am J Orthod 1980; 77(6): 591-612.         [ Links ]

9. Buschang PH, Nass GG, Walker GF. Principal components of craniofacial growth for white Philadelphia males and females between 6 and 22 years of age. Am J Orthod 1982; 82(6): 508-512.         [ Links ]

10. Korkhaus G. Disturbances in the development of the upper jaw and the middle face. Am J Orthod 1957; 43(11): 848- 868.         [ Links ]

11. Bastir M, Rosas A. Correlated variation between the lateral basicranium and the face: a geometric morphometric study in different human groups. Arch Oral Biol 2006; 51(9): 814-824.         [ Links ]

12. Guyer EC, Ellis EE 3rd, McNamara JA Jr, Behrents RG. Components of class III malocclusion in juveniles and adolescents. Angle Orthod 1986; 56(1): 7-30.         [ Links ]

13. Bastir M, Sobral PG, Kuroe K, Rosas A. Human craniofacial sphericity: A simultaneous analysis of frontal and lateral cephalograms of a Japanese population using geometric morphometrics and partial least squares analysis. Arch Oral Bio 2008; 53(4): 295-303.         [ Links ]

14. Enlow DH, McNamara JA Jr. The neurocranial basis for facial form and pattern. Angle Orthod 1973; 43(3): 256- 270.         [ Links ]

15. Enlow DH, Kuroda T, Lewis AB. The morphological and morphogenetic basis for craniofacial form and pattern. Angle Orthod 1971; 41(3): 161-188.         [ Links ]

16. Enlow DH, Kuroda T, Lewis AB. Intrinsic craniofacial compensations. Angle Orthod 1971; 41(4): 271-285.         [ Links ]

17. Bastir M, Rosas A, Stringer C, Cuétara JM, Kruszynski R, Weber GW et al. Effects of brain and facial size on basicranial form in human and primate evolution. J Hum Evol 2010; 58(5): 424-431.         [ Links ]

18. Allen D, Rebellato J, Sheats R, Ceron AM. Skeletal and dental contributions to posterior crossbites. Angle Orthod 2003; 73(5): 515-524.         [ Links ]

19. Lavelle CLB. A study of craniofacial form. Angle Orthod 1979; 49(1): 65-72.         [ Links ]

20. McNamara JA Jr. A method of cephalometric evaluation. Am J Orthod 1984; 86(6): 449-469.         [ Links ]

21. Ricketts RM, Bench RW, Hilgers JJ, Schulhof R. An overview of computerized cephalometrics. Am J Orthod 1972; 61(1): 1-28.         [ Links ]

22. Steiner CC. Cephalometrics In clinical practice. Angle Orthod 1959; 29(1): 8-29.         [ Links ]

23. Moyers RE, Bookstein FL. The inappropriateness of conventional cephalometrics. Am J Orthod 1979; 75: 599- 617.         [ Links ]

24. McIntyre GT, Mossey PA. Size and shape measurement in contemporary cephalometrics. Eur J Orthod 2003; 25(3): 231-242.         [ Links ]

25. Halazonetis DJ. Morphometrics for cephalometric diagnosis. Am J Orthod Dentofacial Orthop 2004; 125(5): 571-581.         [ Links ]

26. Enlow DH, Pfister C, Richardson E, Kuroda T. An analysis of Black and Caucasian craniofacial patterns. Angle Orthod 1982; 52(4): 279-287.         [ Links ]

27. Van der Molen S, González-José R. Introducción a la morfometría geométrica curso teórico-práctico: Barcelona: Universidad de Barcelona, Centro Nacional Patagónico. CENPAT-CONICET; 2007.         [ Links ]

28. Toro M, Manriquez G, Suazo G. Morfometría geométrica y el estudio de las formas biológicas: de la morfología descriptiva a la morfología cuantitativa. Int J Morphol 2010; 28(4): 977-990.         [ Links ]

29. Mitteroecker P, Gunz P. Advances in geometrics morphometrics. Evol Biol 2009 36(2): 235-247.         [ Links ]

30. MetroNukak. Registro: 13-14-374. ed. p. Software multiplataforma de asistencia para el análisis radiológico.         [ Links ]

31. Rahman T, Valdman J. Fast MATLAB assembly of FEM matrices in 2D and 3D: Nodal elements. Appl Math Comput 2013; 219(13): 7151-7158.         [ Links ]

32. Cakirer B, Dean D, Palomo JM, Hans MG. Orthognathic surgery outcome analysis: 3-dimensional lanmark geometric morphometric. Int J Adult Orthodon Orthognath Surg 2002; 17(2): 116-132.         [ Links ]

33. Castro N. Modelo de Identificación de patrones del tercio medio facial en clase I, II y III Esquelética: un análisis morfogeométrico. [Trabajo de grado Magister en Odontología] Bogotá: Universidad Nacional de Colombia; 2013.         [ Links ]

34. Rohlf FJ. tpsDig, digitize landmarks and outlines. 2.16. Department of ecology and evolution. Stony Brook: State University; 2010.         [ Links ]

35. Dujardin J. The MOG software. Version 2, June 1991 ed. Unité de Recherches 062-Unité Mixte de Recherches UMR9926, Institut de Recherches pour le Développement (IRD, France); 2003.         [ Links ]

36. Peres-Neto PR, Jackson DA, Somers KM. How many principal components? Stopping rules for determining the number of non-trivial axes revisited. Comput Stat Data Anal 2005; 49(4): 974-997.         [ Links ]

37. Infante C, López L. Uso de técnicas multivariadas para clasificación de estructuras óseas craneanas. Una aplicación en medicina forense [Trabajo de grado Especialización en Estadística]. Bogotá: Universidad Nacional de Colombia; 2003.         [ Links ]

38. Ozdemir ST, Ercan I, Ozkaya G, Cankur NS, Erdal YS. Geometric morphometrie study and cluster analysis of late Byzantine and modern human. Coll Antropol 2010; 34(2): 493-499.         [ Links ]

39. Rohlf J. tpsRelw, 1.39. Department of ecology and evolution. Stony Brook: State University; 2008.         [ Links ]

40. Henry A, Thongsripong P, Fonseca-Gonzalez I, Jaramillo- Ocampo N, Dujardin JP. Wing shape of dengue vectors from around the world. Infect Genet Evo 2010; 10(2): 207- 214.         [ Links ]

41. Filzmoser P. Identification of multivariate outliers: a performance study. Aust J Stat 2005; 34(2): 127-138.         [ Links ]

42. Hammer Ø, Harper DAT, Ryan PD. PAST: paleontological statistics software package for education and data analysis. Palaeontol Electron 2001; 4(1): 1-9.         [ Links ]

43. Kimmerle EH, Ross A, Slice D. Sexual dimorphism in America: Geometric morphometric analysis of the craniofacial region. J Forensic Sci 2008; 53(1): 54-57.         [ Links ]

44. Díaz M. Estadística mutivariada: inferencia y métodos. Bogotá: Universidad Nacional de Colombia, Departamento de Estadística, Facultad de Ciencias; 2007.         [ Links ]

45. Schaefer K, Mitteroecker P, Gunz P, Bernhard M, Bookstein FL. Craniofacial sexual dimorphism patterns and allometry among extant hominids. Ann Anat 2004; 186(5-6): 471-478.         [ Links ]

46. Dujardin J. PAD software. Version 2, June 1991 ed. Unité de Recherches 062- Unité Mixte de Recherches UMR9926, Institut de Recherches pour le Développement (IRD, France); 2002.         [ Links ]

47. Franklin D, Freedman L, Milne N, Oxnard CE. A geometric morphometric study of sexual dimorphism in the crania of indigenous southern africans. S Afr J Anim Sci 2006; 102(5-6): 229-238.         [ Links ]

48. Akimoto S, Kubota M, Sato S. Increase in vertical dimension and maxillo-mandibular growth in a longitudinal growth sample. J Stomat Occ Med 2010; 3: 15-19.         [ Links ]

49. Singh N, Harvati K, Hublin JJ, Klingenberg CP. Morphological evolution through integration: A quantitative study of cranial integration in Homo, Pan, Gorilla and Pongo. J Hum Evol 2012; 62(1): 155-164.         [ Links ]