This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ORIGINAL ARTICLE
J. Djordjevic
M. Jadallah
A. I. Zhurov
A. M. Toma
S. Richmond
Three-dimensional analysis of facial
shape and symmetry in twins using
laser surface scanning
Authors' affiliation:J. Djordjevic, M. Jadallah, A. I. Zhurov,
All 19 39.39 36.43 44.08 10 24.80 18.62 30.93 <0.001
N, number of within-pair observations; perc., percentile; Pr. dist., Procrustes distance between two landmark configurations within atwin pair; Av. dist., average distance between the two faces of twins; Coincid., the percentage of coincidence between the two faces oftwins (see text and Fig. 2 for further explanation).Eight mixed dizygotic pairs were excluded from the analysis (see text for further explanation) and the data for dizygotic females werenot analysed separately due to only three observations.NS, not statistically significant.†Mann–Whitney U test was performed.
Orthod Craniofac Res 2012 | 7
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
Average faces of MZ and DZ twins of both gen-
ders are presented in Fig. 5. Monozygotic males
tend to have wider face and nose with prominent
upper lip when compared with DZ males, with a
maximum difference of 1.8 mm. On the other
hand, DZ males tend to have more prominent
eyes, upper part of the forehead and central part
of the chin, with a maximum difference of
1.2 mm. The two average male faces coincide
34.1%, mainly in the forehead, supraorbital and
infraorbital ridges, the bridge of the nose and
lower lip. Monozygotic females tend to have
wider lower part of the face, wider nose and more
prominent lips when compared with DZ females,
with a maximum difference of 1.5 mm. Dizygotic
females tend to have more prominent forehead,
the bridge of the nose, malar area and lower part
of the chin, with a maximum difference of
1.1 mm. The two average female faces coincide
61.1%, mainly in the eyes, supraorbital and infra-
orbital ridges, philtrum of the upper lip and lower
part of the cheeks.
Three-dimensional symmetry analysis showed
that there was no statistically significant
difference in the amount of facial symmetry
between MZ and DZ twins (Table 6). When
upper, middle and lower facial thirds were com-
pared, lower facial third was found to be the
most asymmetrical in both twin groups
(Table 7).
Discussion
In this exploratory cross-sectional study, three-
dimensional facial shape and symmetry of 37
twin pairs was investigated using non-invasive
laser surface scanning. Facial analysis was per-
formed using two methods. The first method
was based on geometric morphometrics, which
employs GPA to register (align) sets of landmark
configurations. Although powerful, Procrustes
registration is not an ideal technique to superim-
pose numerous faces, as it relies only on limited
Table 3. Facial shape similarity by facial thirds within monozygotic and dizygotic twin pairs
Twins N Parameter
Upper facial third Middle facial third Lower facial third
N, number of within-pair observations; perc., percentile; MZ, monozygotic; DZ, dizygotic; Av. dist., average distance between the twofaces of twins in a given facial third; Coincid., the percentage of coincidence between the two faces of twins in a given facial third.NS, not statistically significant.Figures in bold indicate facial third which was significantly different from the other two (Mann–Whitney U test was applied).†Kruskal–Wallis test was performed.‡The data for dizygotic females were not analysed separately due to only three observations.§Eight mixed dizygotic pairs were excluded from the analysis (see text for further explanation).
8 | Orthod Craniofac Res 2012
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
number of landmarks and therefore does not
take into account other information available
(17). This is certainly the case when laser scan-
ning devices are used, as these can capture at
least 40 000 points on the facial surface (42).
These points do not have fixed positions and
hence cannot be used as landmarks (17). For
this reason, the second method used in this
study relied on the comparison of facial surfaces.
Two different superimposition techniques were
used and this requires further explanation.
The first method of superimposition (the best-
fit alignment) was performed to compare faces
of twins within each pair. The presumption was
that the faces of twins will be very much alike
and that slight differences can be registered
using this approach. The best-fit method is
based on the iterative closest point algorithm, a
built-in feature of the commercial software.
Essentially, it is a mathematical least-squares
method. Automatic comparison took into
account all pairs of points on the facial surfaces
captured by the devices (several thousands). We
are unaware of any study, which specifically
investigated the reliability of this software tool,
although it has been used before (17, 35, 42).
There is a theoretical possibility that different
piece of software would produce slightly differ-
ent superimposition results. This could be inves-
tigated in future studies.
The second method of superimposition was
used to compare average faces of MZ and DZ
twins. The faces were aligned on mid-endocan-
thion, the point halfway between the inner cor-
ners of the eyes. As previous research has
shown, this point can be used as a relatively
Table 4. Mean and standard deviation of 21 facial landmarks in x-, y- and z-directions for monozygotic and dizygotic malesafter scaling and Generalized Procrustes registration
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
stable reference during facial growth (42). The
expected level of similarity of facial features
between MZ and DZ twins was less than within
the twin pairs. In this case, the best-fit approach
would tend to decrease the difference between
the average MZ and DZ faces. On the other
hand, similarity that was demonstrated by
superimposition on mid-endocanthion point had
a greater chance of representing a true effect.
Surface-based average faces have already been
applied in orthodontics and maxillofacial surgery
to illustrate facial anomalies, evaluate facial
growth, analyse treatment effects, compare facial
morphologies between genders and among dif-
ferent populations (17, 18, 25, 35).
Both facial analyses (landmark-based and sur-
face-based) revealed greater similarity of facial
surfaces in MZ twins than in DZ twins. This is
in agreement with previous studies (19, 21, 22).
Instead of presenting only descriptive data
(obtained from colour maps), facial shape simi-
larity within twin pairs was further investigated
by dividing the faces into thirds and analysing
facial shape parameters statistically. In MZ
males, the lower facial third was the least simi-
lar, whereas in MZ females, no statistically sig-
nificant difference was determined. In DZ
males, upper facial third had the most similar
shape within twin pairs. These findings can
indicate that the influence of genetic and envi-
ronmental factors and their interaction on soft
tissue shape is not the same in all facial
regions.
The amount of three-dimensional facial sym-
metry measured in MZ and DZ twins was similar
as the one measured in 270 singletons from the
same population (40). The study failed to show
significant differences in facial symmetry
between MZ and DZ twin groups. It can either
be due to small sample size, which prevents
detection of any difference (possible type II
error) or an indication that facial soft tissue
symmetry of healthy individuals is not under
strong genetic control. The latter statement is
supported by previous study (20). More studies
with larger samples are needed to reach a defi-
A
B
Fig. 4. The 95% of the variation in scaled 21 facial landmarks for 74 monozygotic and dizygotic 15-year-old twins. (A) Front (x
and y coordinates) and profile view (y and z coordinates) for male facial landmarks (18 monozygotic males represented by light
grey ellipsoids and 22 dizygotic males represented by dark grey ellipsoids). (B) Front (x and y coordinates) and profile view (y
and z coordinates) for female facial landmarks (20 monozygotic females represented by light grey ellipsoids and 14 dizygotic
females represented by dark grey ellipsoids). The data were aligned on the mid-endocanthion, a point halfway between the inner
canthi of the eyes. For names of landmarks refer to Figure 1.
10 | Orthod Craniofac Res 2012
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
nite conclusion. Lower facial third was found to
be the most asymmetric, contrary to the findings
of previous study on 270 singletons (40), which
showed no statistically significant difference in
the amount of symmetry among upper, middle
and lower facial thirds. There is no consensus in
the literature on the most asymmetric part of
the face (39, 40).
One of the strengths of this study is that facial
analysis was performed in a sample of twins of
the same age, with confirmed zygosity, who were
born and raised in the same geographical region.
Hence, the sample was more homologous than
in previous three-dimensional studies on facial
morphology of twins. The accuracy of the laser
scanning method enables quantification of even
subtle differences in facial morphology and sym-
metry, which has not been feasible previously
using two-dimensional data from photographs
or lateral cephalograms. In order to use the full
potential of this technology, future studies on
heritability of facial features will have to adopt
some method of three-dimensional facial
analysis.
As with any observational study, there are
some limitations. The sample comes from a
longitudinal population-based study, in which
twins constitute approximately 1.3% of the
cohort (24). As facial laser scanning was orga-
nized only during one follow-up clinic, sample
size could not be increased. In addition, many
confounding factors could not be controlled. In
facial analysis, some of these factors might be
related to body mass index, medical conditions,
orthodontic treatment and trauma. In future
studies, these factors should be carefully moni-
tored to ensure the consistency of the results.
It has been argued that twin results need to be
interpreted with great caution, and that other
family relations should be taken into account,
A
B
Fig. 5. Average faces for monozygotic and dizygotic 15-year-old twins. (A) Average face for 18 monozygotic males (left), average
face for 22 dizygotic males (centre) and the colour map and the histogram (right), which show the difference in facial shape of
the two male twin groups. (B) Average face for 20 monozygotic females (left), average face for 14 dizygotic females (centre) and
the colour map and the histogram (right), which show the difference in facial shape of the two female twin groups. Average faces
were aligned on mid-endocanthion, and monozygotic average face in both genders was taken as the reference. Shades of blue
represent negative (minimum �2 mm), and shades of red positive differences (maximum 2 mm) in facial shape. Grey area repre-
sents the parts of two facial surfaces, which coincide within 0.5 mm (34.1% in males and 62.1% in females).
Orthod Craniofac Res 2012 | 11
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
especially comparisons between parents and
children (14). An ongoing project aims at
exploring this aspect of the problem, and the
faces of approximately 1300 fathers of ALSPAC
children are currently being laser scanned.
Genome-wide associations will have a major
Table 5. Mean and standard deviation of 21 facial landmarks in x-, y- and z-directions for monozygotic and dizygotic femalesafter scaling and Generalized Procrustes registration
N, number of individuals; perc., percentile; Av. dist., average distance between the original and mirror faces of one twin; Coincid., thepercentage of coincidence between the original and mirror faces of one twin (within 0.5 mm of tolerance); NS, not statistically signifi-cant.*Mann–Whitney U test was performed.
12 | Orthod Craniofac Res 2012
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
role in identifying genetic variants responsible
for normal facial variations. It is believed that
three-dimensional imaging will be of great help
in the quest for further knowledge on genetic
and environmental determinants of facial fea-
tures.
Conclusions
From this study on three-dimensional facial
shape and symmetry of 15-year-old white Brit-
ish twins, the following conclusions can be
drawn:
1. Landmark-based and surface-based three-
dimensional facial analyses can reveal
within- and between-pair differences in facial
soft tissue shapes of MZ and DZ twins.
2. Genetic factors play an important role in
three-dimensional soft tissue shape and the
relative contribution of genetic and
environmental factors is not the same for
the upper, middle and lower facial thirds.
3. Zygosity does not seem to influence the
amount of three-dimensional symmetry of
facial soft tissues. Lower facial third is the
most asymmetrical part of the face in both
MZ and DZ twins.
Clinical relevance
The twin method has been extensively used in
orthodontics in the last few decades to investigate
the relative contribution of genetic and environ-
mental factors to the shape of the craniofacial
complex. Three-dimensional imaging systems
provide a possibility to obtain and analyse facial
soft tissue morphology non-invasively, accurately
and reliably. In this study, laser surface scanning
was used to compare facial shape and symmetry of
MZ and DZ twins. Suggested three-dimensional
facial analyses can reveal differences in facial mor-
phology, which has not been possible previously
using two-dimensional data. This can support fur-
ther research in craniofacial genetics.
Table 7. Three-dimensional symmetry by facial thirds in monozygotic and dizygotic twins
Twins N Parameter
Upper facial third Middle facial third Lower facial third
N, number of individuals; perc., percentile; Av. dist., average distance between the original and mirror faces of one twin; Coincid., thepercentage of coincidence between the original and mirror faces of one twin (within 0.5 mm of tolerance).Figures in bold indicate facial third which was significantly different from the other two (Mann–Whitney U test was applied).NS, not statistically significant.†Kruskal-Wallis test was performed.
Orthod Craniofac Res 2012 | 13
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
Acknowledgements: We are extremely grateful to all
the families who took part in this study, the midwives
for their help in recruiting them and the whole ALSPAC
study team, which includes interviewers, computer and
laboratory technicians, clerical workers, research scien-
tists, volunteers, managers, receptionists and nurses.
The UK Medical Research Council, the Wellcome Trust
(Grant ref: 092731) and the University of Bristol provide
core support for ALSPAC.
References1. Emery AE, Mueller RF. Elements of
Medical Genetics. Edinburgh: Chur-
chill Livingstone; 1988.
2. http://www.genome.gov (accessed
on 26th February 2012)
3. Freimer N, Sabatti C. The human
phenome project. Nat Genet
2003;34:15–21.
4. Peng J, Deng H, Cao CF, Ishikawa
M. Craniofacial morphology in
Chinese female twins: a semi-longi-
tudinal cephalometric study. Eur J
Orthod 2005;27:556–61.
5. Carels C, Van Cauwenberghe N,
Savoye I, Willems G, Loos R, Derom
C et al. A quantitative genetic study
of cephalometric variables in twins.
Clin Orthod Res 2001;4:130–40.
6. Savoye I, Loos R, Carels C, Derom C,
Vlietinck R. A genetic study of ante-
roposterior and vertical facial pro-
portions using model-fitting. Angle
Orthod 1998;68:467–70.
7. Manfredi C, Martina R, Grossi GB,
Giuliani M. Heritability of 39 ortho-
dontic cephalometric parameters on
MZ, DZ twins and MN paired single-
tons. Am J Orthod Dentofacial Ort-
hop 1997;111:44–51.
8. Lobb WK. Craniofacial morphology
and occlusal variation in monozy-
gotic and dizygotic twins. Angle
Orthod 1987;57:219–33.
9. Lundstr€om A, McWilliam JS. A com-
parison of vertical and horizontal
variables with regard to heritability.
Eur J Orthod 1987;9:104–8.
10. Alkhudhairi TD, AlKofide EA. Cepha-
lometric craniofacial features in
Saudi parents and their offspring.
Angle Orthod 2010;80:1010–17.
11. Sherwood RJ, Duran DL, Demerath
EW, Czerwinski SA, Siervogel RM,
Towne B. Quantitative genetics of
modern human cranial variation.
J Hum Evol 2008;54:909–14.
12. Baydas� B, Erdem A, Yavuz I, Ceylan
I. Heritability of facial proportions
and soft-tissue profile characteristics
in Turkish Anatolian siblings. Am J
Orthod Dentofacial Orthop
2007;131:504–9.
13. Gelg€or IE, Karaman AI, Zekic� E. The
use of parental data to evaluate soft
tissues in an Anatolian Turkish
population according to Holdaway
soft tissue norms. Am J Orthod
Dentofacial Orthop 2006;129:330
e1–9.
14. Johannsdottir B, Thorarinsson F,
Thordarson A, Magnusson TE. Heri-
tability of craniofacial characteristics
between parents and offspring esti-
mated from lateral cephalograms.
Am J Orthod Dentofacial Orthop
2005;127:200–7.
15. Paternoster L, Zhurov AI, Toma AM,
Kemp JP, St Pourcain B, Timpson NJ
et al. Genome-wide association
study of three-dimensional facial
morphology identifies a variant in
PAX3 associated with nasion posi-
tion. Am J Hum Genet 2012;90:478–
85.
16. Kohn LAP. The role of genetics in
craniofacial morphology and growth.
Annu Rev Anthropol 1991;20:261–78.
17. Kau CH, Richmond S. Three-Dimen-
sional Imaging for Orthodontics and
Maxillofacial Surgery. Ames, IA:
Wiley Blackwell; 2010.
18. Kau CH, Richmond S, Incrapera A,
English J, Xia JJ. Three-dimensional
surface acquisition systems for the
study of facial morphology and their
application to maxillofacial surgery.
Int J Med Robotics Comput Assist
Surg 2007;3:97–110.
19. Burke PH. Intrapair facial differ-
ences in twins. Acta Genet Med Ge-
mellol 1989;38:37–47.
20. Burke PH, Healy MJR. A serial study
of normal facial asymmetry in
monozygotic twins. Ann Hum Biol
1993;20:527–34.
21. Naini FB, Moss JP. Three-dimen-
sional assessment of the relative
contribution of genetics and environ-
ment to various facial parameters
with the twin method. Am J Orthod
Dentofacial Orthop 2004;126:655–65.
22. Moss JP. The use of three-dimen-
sional imaging in orthodontics. Eur J
Orthod 2006;28:416–25.
23. Golding J, Pembrey M, Jones R, the
Alspac Study Team. ALSPAC – The
Avon Longitudinal Study of Parents
and Children. I. Study methodology.
Paediatr Perinat Epidemiol
2001;15:74–87.
24. Boyd A, Golding J, Macleod J, Lawlor
DA, Fraser A, Henderson J et al.
Cohort profile: the ‘children of the
90s’–the index offspring of the Avon
Longitudinal Study of Parents and
Children. Int J Epidemiol 16 Apr
2012; (epub ahead of print) doi:
10.1093/ije/dys064
25. Toma AM, Zhurov AI, Playle R,
Marshall D, Rosin PL, Richmond S.
The assessment of facial variation in
4747 British school children. Eur J
Orthod 2012;34:655–64.
26. Kau CH, Knox J, Zhurov AI, Rich-
mond S. The validity and reliability
of a portable 3-dimensional laser
scanner for field studies. In: Giuliani
R, Galliani E, editors. 7th European
Craniofacial Congress. Bologna:
Monduzzi Editore – International
Proceedings Division; 2004.
pp. 41–5.
27. Kau CH, Richmond S, Zhurov AI,
Bouwman S, Scheer R. Feasibility of
measuring 3D facial morphology in
children. Orthod Craniofac Res
2005;7:1–7.
28. Kau CH, Zhurov AI, Knox J, Chest-
nutt I, Hartles FR, Playle R et al.
Reliability of measuring facial
morphology using a 3-dimensional
laser scanning system. Am J
Orthod Dentofacial Orthop
2005;128:424–30.
29. Kau CH, Richmond S, Savio C,
Mallorie C. Measuring adult facial
morphology in three dimensions.
Angle Orthod 2006;76:773–8.
30. Kovacs L, Zimmermann A, Brock-
mann G, Baurecht H, Schwenzer-
Zimmerer K, Papadopulos NA. Accu-
racy and precision of the three-
dimensional assessment of the facial
surface using a 3-D laser scanner.
IEEE Trans Med Imaging 2006;25:
742–54.
14 | Orthod Craniofac Res 2012
Djordjevic et al. 3D analysis of facial shape and symmetry in twins
31. Kusnoto B, Evans CA. Reliability of
a 3D surface laser scanner for
orthodontic applications. Am J
Orthod Dentofacial Orthop
2002;122:342–8.
32. Zhurov AI, Kau CH, Richmond S.
Computer methods for measuring
3D facial morphology. In: Middleton J,
Shrive N, Jones M, editors. Proceed-
ings of the 6th International Sympo-
sium on Computer Methods in
Biomechanics & Biomedical Engi-
neering. Cardiff: First Numerics Ltd;
2005. (CD ROM, paper 151D).
33. Kau CH, Hartles FR, Knox J, Zhurov
AI, Richmond S. Natural head pos-
ture for measuring three-dimen-
sional soft tissue morphology. In:
Middleton J, Shrive N, Jones M,
editors. Proceedings of the 6th Inter-
national Symposium on Computer
Methods in Biomechanics & Biome-
dical Engineering. Cardiff: First
Numerics Ltd; 2005. (CD ROM,
paper 149D).
34. Kau CH, Hartles FR, Knox J, Zhurov
AI, Richmond S. Measuring facial
morphology in young subjects. In:
Middleton J, Shrive N, Jones M,
editors. Proceedings of the 6th Inter-
national Symposium on Computer
Methods in Biomechanics & Biome-
dical Engineering. Cardiff: First
Numerics Ltd; 2005. (CD ROM,
paper 150D).
35. Toma AM, Zhurov A, Playle R, Rich-
mond S. A three-dimensional look
for facial differences between males
and females in a British-Caucasian
sample aged 15 1/2 years old.
Orthod Craniofac Res 2008;11:180–5.
36. Farkas LG, editor. Anthropometry of
the Head and Face. New York: Raven
Press; 1994.
37. Bookstein FL. Morphometric Tools
for Landmark Data. Cambridge:
Cambridge University Press; 1991.
38. Zelditch ML, Swiderski DL, Sheets
HD, Fink WL. Geometric Morpho-
metrics for Biologists: A Primer. New
York: Elsevier academic press; 2004.
39. Djordjevic J, Pirttiniemi P, Harila V,
Heikkinen T, Toma AM, Zhurov AI
et al. Three-dimensional longitudi-
nal assessment of facial symmetry in
adolescents. Eur J Orthod 2011; Feb
7 (epub ahead of print) doi: 10.1093/
ejo/cjr006.
40. Djordjevic J, Toma AM, Zhurov AI,
Richmond S. Three-dimensional
quantification of facial symmetry in
adolescents using laser surface scan-
ning. Eur J Orthod 2011; Jul 27
(epub ahead of print) doi: 10.1093/
ejo/cjr091.
41. Toma AM, Zhurov A, Playle R, Ong
E, Richmond S. Reproducibility of
facial soft tissue landmarks on 3D
laser-scanned facial images. Orthod
Craniofac Res 2009;12:33–42.
42. Huang GJ, Richmond S, Vig KWL.
Evidence-Based Orthodontics. Ames,
IA: Wiley-Blackwell; 2011.
Orthod Craniofac Res 2012 | 15
Djordjevic et al. 3D analysis of facial shape and symmetry in twins