Odontology (2017) 105:229–236

DOI 10.1007/s10266-016-0244-7

ORIGINAL ARTICLE

Mathematical beta function formulation for maxillary arch form

prediction in normal occlusion population
Morteza Mina1 • Ali Borzabadi-Farahani2,3 • Azita Tehranchi4,5 •

Mahtab Nouri5,6 • Farnaz Younessian5,7

Received: 20 September 2015 / Accepted: 2 April 2016 / Published online: 11 May 2016
Ó The Society of The Nippon Dental University 2016

Abstract The aim of this study was to assess the dental arch the proposed formula was assessed on 10 randomly selected
curvature in subjects with normal occlusion in an Iranian dental casts. The mean (SD) of the maxillary and mandibular
population and propose a beta function formula to predict IMW and IMD were 57.92 (4.75), 54.19 (5.31), and 31.59
maxillary arch form using the mandibular intermolar widths (2.90) and 28.10 (2.59) mm, respectively. There was no
(IMW) and intermolar depths (IMD). The materials used gender dimorphism (P [ 0.05) for both variables (IMW,
were study casts of 54 adolescents with normal occlusion and IMD). There was a strong positive association (n = 10,
mean age of 14.1 years (25 males, 29 females, age range Pearson r = 0.98, P \ 0.05) between the measured (actual)
12–16 years). Curve-fitting analyses were carried out and the maxillary arch length and proposed arch length derived from
curves passing through the facial-axis point of the canines, generated formula. The goodness of fit (whole arch) for the
premolars, first molars, and the incisal edges of the anterior proposed beta function formula, using adjusted r square
teeth were studied using a 3D laser scanner. Using the measure and root mean square in 10 patients averaged 0.97
measured IMW and IMD of the dental arches at the maxillary and 1.49 mm, respectively. The corresponding figures for
and mandibular first molar region, a beta function formula the maxillary anterior arch (canine to canine) were 0.90 and
proposed for predicting maxillary arch form. The accuracy of 0.92 mm, respectively. The proposed beta function formula
used for predicting maxillary arch form based on two
mandibular measures (IMW, IMD) was found to have a high
& Ali Borzabadi-Farahani
faraortho@yahoo.com
accuracy for maxillary arch prediction in the Iranian popu-
lation and may be used as a guide to fabricate customized
1
Department of Orthodontics, School of Dentistry, North arch wires or as an aid in maxillary reconstructive surgery.
Khorasan University of Medical Sciences, Bojnourd, Iran
2
Warwick Medical School, University of Warwick, Coventry, Keywords Dental arch form  Orthodontics  Normal
UK occlusion  Beta function
3
Orthodontics, Department of Clinical Sciences and
Translational Medicine, University of Rome ‘‘Tor Vergata’’,
Rome, Italy
Introduction
4
Preventive Dentistry Research Center, Research Institute of
Dental Sciences, School of Dentistry, Shahid Beheshti
University of Medical Sciences, Tehran, Iran
The definition of ideal arch form improves the under-
5
standing of malocclusion and assist clinicians in producing
Department of Orthodontics, School of Dentistry, Shahid
Beheshti University of Medical Sciences, Tehran, Iran
orthodontic outcomes that are consistent with the natural
6
laws of biologic muscular variation [1]. The prediction or
Dentofacial Deformities Research Center, Research Institute
preservation of the dental arch form is important in
of Dental Sciences, School of Dentistry, Shahid Beheshti
University of Medical Sciences, Tehran, Iran orthodontic treatment, as for instance archform should not
7 change greatly during orthodontic treatment, affecting
Dental Research Center, Research Institute of Dental
Sciences, School of Dentistry, Shahid Beheshti University of long-term occlusal stability [2, 3]. Arch form can be cat-
Medical Sciences, Tehran, Iran egorized into simple qualitative terms such as elliptic,

123
230
parabolic, and U-shaped [4], which were inadequate to
accurately define the dental arch form [4]. Investigators
have, therefore, shifted toward more quantitative approa-
ches including some linear measurements in describing the
ideal arch form. Arch form can refers characteristics of the
dental arch such as symmetry, roundness, the radius of
curvature of the labial segment, intercanine width, and the
intermolar width.
There have been attempt to identify the best mathe-
matical equation that describes dental arch curvatures. For
instance, a mathematical curve has been suggested for
describing ideal dental arch curvature [5]. In order to
increase the accuracy of dental arch descriptions, mathe-
matical formulas that have been used to describe dental
arch curvatures can include conic sections [6], catenary
curves [7], cubic spline curves [6], second- to eighth-de-
gree polynomials [8], mixed models [6], and the beta
function [9].
It has become clear that the models defined by only one
parameter cannot describe the dental arch form accurately
[4]. It has been shown that the human dental arch form is
accurately represented mathematically by the beta function
formulation using measures of the same arch, which pro-
vides an excellent generalized equation of the maxillary
and mandibular arch shapes for each of the Angle classi-
fications of occlusion [9]. This is because it is based on two
measures of the dental arch and it produces a symmetric
arch.
Determination of ideal individualized upper arch form
can be a valuable clinical guideline, particularly in patients
with congenital anomalies and dentofacial deformities with
malformed arch forms, such as those who need maxillary
surgical reconstruction and parts of the maxilla is missing,
due to accidents or facial clefts, or patients with a very
underdeveloped maxilla [10]. In orthognathic patients, arch
coordination is perceived as the major goal of presurgical
orthodontic phase, which can be evaluated mainly by
impression and stage recording in several clinical
appointments, and any guide to facilitate these adaptations
can be considered a significant step toward better clinical
services.
Using 3D computerized images instead of plaster
orthodontic models has several advantages such as accu-
racy, efficiency, and facility of measuring of tooth and arch
sizes [11]. We conducted a 3D study to assess the arch
dimensions in 12–16 year-old Iranian adolescents with
normal occlusion. To our knowledge, information on the
fitness of mathematical models describing arch forms, such
as the beta function in normal occlusions is limited, and in
particular, applicability of the beta function in an Iranian
population has not been tested. We therefore, using the data
from subjects with normal population formulated a
regression beta function formula for maxillary dental arch,
123
Odontology (2017) 105:229–236
using the two mandibular measures [intermolar widths
(IMW) and intermolar depths (IMD)]. Further, the accu-
racy of its prediction was evaluated by means of root mean
square to the actual points as well as measuring arc length
using integration [12].
The null hypothesis for this study was that no gender
differences existed between the intermolar widths (IMW)
and intermolar depths (IMD) of maxillary and mandibular
arches of subjects with normal occlusion.
Materials and methods
This cross-sectional study was conducted on 54 adolescents
with mean age 14.1 years (25 males, 29 females, age range
of 12–16 years) and normal occlusion. The patients were
randomly selected from the schools of the city of Qazvin
(Iran) with the following inclusion criteria:
– Balanced and symmetric face.
– Class I occlusion with bilateral Class I canine and
molar relationship.
– Maximum interocclusal contact of all permanent teeth
from the first molar on one side to its counterpart on the
other side, which were fully erupted and in occlusal
contact (except for the second and third permanent
molars due to the age limit of the patients).
– No abnormality in number (agenesis, extraction, or
supernumerary teeth) or shape of teeth.
– Normal overbite and overjet (2–4 mm).
– Coincidence of mandibular and maxillary midlines.
– Lack of interproximal caries.
– Minimal crowding (crowding of 1–2 mm) and small
rotations were ignored.
The exclusion criteria were:
– Previous history orthodontic treatment.
– Presence of restorations extending to the contact areas,
cuspal tips (incisal edges) or cervical areas.
Study cast preparation
Standard orthodontic study casts were prepared and trim-
med symmetrically [13]. In case of any artifact or bubbles
on the final cast, impression taking and casting were
repeated. To facilitate the point registration, the casts were
soaked in black color water-soluble dye (Pars Co., Tehran,
Iran) and the points were defined by means of a superfine-
tip marker (nail design polish, Victoria, Taiwan, Taiwan)
(Fig. 1). The facial-axis point (FA Point) is defined as the
midpoint of the facial axis of the clinical crown, which is
the most prominent part of the central lobe on each crown’s
facial surface except for the molars [1]. The first molar FA
Odontology (2017) 105:229–236
Fig. 1 Photographic image of a model set with determined facial-
axis (FA) points of all permanent teeth, considering the FA point of
first molars as the reference value
Point coordinates were identified with a digital orthodontic
gauge (Unitek, US) and used as a reference point on all
permanent teeth registered on dental casts (except second
molars). For the first molars, the facial axis of the clinical
crown is represented by the mesiobuccal groove that sep-
arates the 2 large facial cusps. The maxillary midline was
constructed using second rugae and midpoint of fovea
palatine and was transferred to mandibular arch when casts
were in maximum interdigitation. The curves passing
through the FA points of the canines, premolars, and first
molars, along with the incisal edges of the anterior teeth
were studied using the 3D laser scanner.
3D surface laser scan acquisition
The dental casts were laser scanned with a 3D surface laser
scanner (Patent No. 69383, Iran). This system consisted of
a laser scanning unit, a computer-aided design software
program, and dental cast analyzing software.
The measuring device of the laser scanning unit con-
sisted of two Class 2 laser diodes operating at 685 nm with
the power output of 1 mW. Each slit-ray laser projected a
line 100 lm thick at a distance of 180 mm. Two sets of
charged coupled device video cameras (768 9 493 pixels,
Hitachi KPM 1, Japan) with the average distance of 19 cm
to the casts captured the reflected images and transferred it
to a computer. The maximum area of test scanning was
6 9 6 cm2. The cast was secured to the horizontal surface
of a rotating table, which rotated with an accuracy of
0.009°. The precision of the applied laser scanner has been
231
validated previously [14]. Considering the pixel coordi-
nates and the characteristics of the camera lens, a linear
equation was written and the coordinates were measured.
The image processing unit received the coordinates of each
selected point as visual signals in analog format from the
CCD camera, converted them to digital information and
stored them on a computer. This information consisted of
overlapped and repeated coordinates of each point. Thus,
prior to the construction of 3D images, the repeated data
were deleted. Accordingly, a 3D wire frame model was
created. Triangulation was used to determine the location
of each point with a measurement error of less than
0.1 mm. The generation of 3D graphics of each dental cast
took approximately 20 min. A standard graph format was
created to enable comparisons of the patients. The data
were first translated, shifting the midpoint between FA
points of the maxillary central incisors to the origin of the
graph (x–y intersection). A two dimensional coordination
of each dental landmark was subsequently obtained by
reducing the Z-coordinate of each recorded landmark to
zero. Then, it was rotated, relocating the midpoint of the
first molars to the y axis. The positions of the rotated ref-
erence points and the curve were confirmed on the graphic
display of the software program. This method was applied
to the data of each set of upper and lower FA points, and
average curves were created.
Variables measured and the proposed well-fitting
arch form
The measurements included the widths and depths of the
dental arches at the maxillary and mandibular first molar
region and were defined as follows:
1. Intermolar width (IMW): distance between the FA
points of the first molars.
2. Intermolar depth (IMD): distance between the contact
of the central incisors and a line that connects the FA
point of the first molars.
To propose a well-fitting maxillary arch form, we used a
general form of the beta function. The beta function in the
present study is a curve based on two parameters of
mandibular intermolar depth and intermolar width. The
beta function is an empirical formulation, a low order
polynomial curve expressed as:
   
x 1 0:8 1 x 0:8
y ¼ 3:0314  D  þ  
w 2 2 w
For the purpose of this study, W is the intermolar width
(IMW) and D is the intermolar depth (IMD).
123
232 Odontology (2017) 105:229–236

Table 1 Mean and Standard deviation (SD) of intermolar width (IMW) and intermolar depth (IMD) for the sample, according to the maxillary
and mandibular arches as well as the correlation coefficient (r) between them
n = 54 Maxillary arch (mm), mean (SD) Mandibular arch (mm), mean (SD) Correlation coefficient (r) Sig.

IMW 57.92 (4.75) 54.19 (5.31) 0.828 P \ 0.0001

IMD 31.59 (2.90) 28.10 (2.59) 0.649 P \ 0.0001

Statistical analysis
IBM SPSS version 22.0 for Windows (IBM Corp.,
Armonk, NY, USA) was used to perform statistical anal-
yses. The mean and standard deviation (with 95 % confi-
dence interval) of IMW and IMD measurements were
calculated for maxillary and mandibular arches of the 54
subjects (study casts) for both females and males. The
Kolmogorov–Smirnov test was used to assess the normal
distribution of variables. Differences between the mea-
surements of the maxillary and mandibular arches were
analyzed using paired t-test for all study samples; unpaired
t-test was used to analyze differences between genders.
In order to assess the intra-observer reliability, IMW and
IMD were measured twice, at two weeks interval, on 10
randomly selected dental casts. The differences between
the first and second measurements analyzed by the paired
t-test, which was insignificant (P [ 0.05). The intraob-
server agreement for both measurements (IMW, IMD) was
excellent [intraclass correlation coefficient (ICC) [0.90].
At the later stage, a beta function formula for upper arch
was presented by means of the MATLAB Curve Fitting
toolbox. By entering 2 parameters of a mandibular dental
arch into this formula, one could predict the maxillary
dental arch form. The accuracy of the proposed prediction
model was tested by two quantitative measurements on 10
randomly selected dental casts from initial samples,
including:
1. Arch length measurement using integral calculation
(below formula), and calculating the Pearson’s corre-
lation coefficient between the actual and predicted
curves (adjusted r square),
whole arch and anterior segment (six anterior teeth,
from canine to canine):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
RMS½Dy2  ¼ r2 þ l2
The above formula is a general formula for calculation
of RMS; where, r and l are the standard deviation and
mean of Dys, respectively.
Results
To calculate the cumulative frequency for a true distribu-
tion of variables (IMW and IMD), we used Q–Q plot
supplemented by Kolmogorov–Smirnov test, which indi-
cated that the normal probability distribution function was
well fitted to the population for both variables.
Table 1 presents data for maxillary and mandibular
IMW and IMD measurements. The mean (SD) of the
maxillary and mandibular IMW were 57.92 (4.75) and
54.19 (5.31) mm, respectively. The corresponding val-
ues for the maxillary and mandibular IMD were 31.59
(2.90) mm and 28.10 (2.59) mm, respectively. There was
no gender dimorphism (unpaired student t-test,
P [ 0.05) for both variables (IMW, IMD) (Table 2).
Therefore, the null hypothesis was not rejected. The
correlation coefficients (r) between upper and lower
IMW and IMD were 0.82 and 0.65, respectively
(P \ 0.0001) (Table 1).
The mean differences between IMW and IMD were
3.73 and 3.49 mm, respectively, which was significant
(Table 3, P \ 0.05).

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2 The proposed beta function formula
ZLM
dy
S¼ 1þ dx
dx Based on the above mentioned descriptive measurements
RM
and mandibular arch dimension data, we were able to
where, ddyx is the first derivative of the arc curve with estimate a modified beta function formula for predicting
respect to horizontal coordinate, S is the arc length, maxillary arch form:
LM is the left first molar and RM is the right first y ¼ 3:0314  ðD þ 3:49Þ
molar.    0:8
x 1 0:8 1 x
2. Root mean square calculation of differences between  þ  
w þ 3:73 2 2 w þ 3:73
the y coordinates of the actual and proposed curves of

123
Odontology (2017) 105:229–236 233

Table 2 The effect of gender

Gender Number Mean (SD) Std. error P value
on measured intermolar width
(IMW) and intermolar depth IMW maxillary arch 29 female 57.06 (4.60) 0.85 0.812
(IMD) according to maxillary
and mandibular arches 25 male 58.93 (4.81) 0.96
IMW mandibular arch 29 female 53.81 (5.31) 0.98 0.934
25 male 54.63 (5.37) 1.07
IMD maxillary arch 29 female 30.82 (2.51) 0.46 0.271
25 male 32.42 (3.15) 0.63
IMD mandibular arch 29 female 27.77 (2.39) 0.44 0.320
25 male 28.47 (2.80) 0.56

Table 3 Statistical comparison of intermolar width (IMW) and Table 4 The measured mean maxillary arch length and proposed
intermolar depth (IMD) differences [mean (SD)] between the max- maxillary arch in 10 patients (mm), as well as the Pearson correlation
illary and mandibular arches coefficient value between measured and proposed maxillary arch
lengths
Maxillary and mandibular arch Mean diff. Sig. (2-
comparison (SD) tailed) Arch length calculation Mean (mm) Pearson correlation
coefficient (r)
IMW difference 3.73 (2.99) P \ 0.0001
IMD difference 3.49 (2.32) P \ 0.0001 Actual upper arch length 90.23 0.98
Proposed upper arch length 87.06

and the root of this mean indicated the amount of RMS

Formula verification
To determine the accuracy of the proposed formula, we
examined 10 study models from the original sample. We
calculated the real arch length (found by the curve fitting
toolbox of MATLAB software) and the predicted arch
length (from the beta function formula) using integration.
Furthermore, the Pearson’s correlation coefficient showed
that the value for r was 0.98, indicating a strong positive
association between measured maxillary arch length and
predicted arch length derived from the generated formula
(Table 4).
The average calculated goodness of fit for the whole
arch due to proposed beta function formula using adjusted
r square was 0.97 (Table 5; Fig. 2a).
To report the observed differences between the actual
and proposed arch forms, the RMS was measured for the
whole arch by considering all 12 dental landmarks
(Table 5) and particularly at anterior areas (six anterior
teeth, from canine to canine) of the named 10 patients
(Table 6; Fig. 2b). The average reported goodness of fit
due to proposed beta function formula using adjusted r
square at anterior area was 0.90 mm and the mean values
of RMS at anterior arch area and 0.92 mm, respectively.
The distance of each clinical bracket attachment point from
the drawn curve was measured and their RMS was calcu-
lated by the given formula. This means that the distance of
12 CBPs was measured from the corresponding points
along the y axis on the polynomial curve fitted to them and
squared. The mean of the squared distances was calculated
P
p x2
(RMS = n ). The smaller the RMS value, the greater
the adaptation of the arch to the clinical bracket points.
This calculation was also performed for the six anterior
teeth.
Discussion
Previous studies evaluated arch dimensions in different
malocclusions [15–17]. Slaj et al. [18] examined the dental
arch dimensions in malocclusions and found larger inter-
canine and intermolar widths in Class III malocclusions
when compared to Class I or II malocclusions. Braun et al.
[9] reported that dental arches in patients with Class III
malocclusions were wider than that of subjects with Class I
malocclusions. The mandible of Class III cases may
demonstrate a wider variety of relationships between the
dental and basal arch forms compared to patients with
normal occlusion and are often wider in Class III maloc-
clusions [19].
In the present study, the differences between upper and
lower arch in a group of normal occlusion samples helped
generate a beta function formula. The mean maxillary and
mandibular inter-molar width and depth values in males
were greater than in females, but the differences were not
statistically significant (Table 2).
The possible explanation for the observed natural dif-
ferences in arch dimensions can be the development of
alveolar sockets during the age of 8-13 years and the

123
234 Odontology (2017) 105:229–236

Table 5 The calculated goodness of fit for the whole maxillary arch prediction (12 dental landmarks) with the proposed beta function formula
by means of adjusted r-square measure and Root Mean Square of the differences in 10 patients (mm)
Number 1 2 3 4 5 6 7 8 9 10 Average

Adjusted r square 0.92 0.99 0.98 0.98 0.98 0.98 0.98 0.95 0.98 0.99 0.97
RMS (root mean square) 2.68 1.32 1.41 1.13 1.35 1.62 1.08 2.09 1.35 0.94 1.49

Fig. 2 a A sample of the

graphic representation of the
superimposition of point
registration on dental cast on the
beta function of the maxillary
arches (12 points). b Graphic
representation of the
superimposition of point
registration on dental cast on the
beta function at anterior area (6
points) for the same patient

Table 6 Comparison of the estimated maxillary anterior arch (from canine to canine) with actual maxillary anterior arch, reported by means of
adjusted r square measure and Root Mean Square (RMS) in 10 patients (mm)
Number 1 2 3 4 5 6 7 8 9 10 Average

Adjusted r square 0.93 0.92 0.40 0.98 0.97 0.98 0.96 0.97 0.99 0.98 0.90
RMS (root mean square) 0.85 1.14 3.08 0.61 0.76 0.62 0.69 0.57 0.28 0.60 0.92
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi P
n P
n
RMS ¼ ðE½DyÞ2 þ r2 , E½Dy ¼ 1n Dyi , Dyi ¼ yproposed
i yreal 1
i , r¼ n ðDyi  E½DyÞ2
i¼1 i¼1
r standard deviation, n measured points

outward divergence of their processes in the following mandible. Bishara et al. [20] reported a slight decrease of
years. In the mandible, the slight increase of inter-molar arch length during the period of 8–18 years in both the
width is due to conformity of molars to occlusion and slight maxilla and mandible. This is probably due to the mesial
bone deposition on the outer surfaces of the body of the migration of posterior teeth during this time, lingual

123
Odontology (2017) 105:229–236
positioning of the incisors as a result of differential growths
of the maxilla and mandible, and natural torque of molars
and incisors [21]. Racial diversity could possibly influence
the dental arch form [22]. Chang et al. [23] compared
Chinese males and females with normal occlusion with the
average age of 15 years. The researchers found that the
normal occlusion group had dimorphism (male [ female)
in all widths. Similar findings reported by Kuntz et al. [15].
They both concluded that males had significantly larger
inter-molar widths than females, for both upper and lower
arches [15, 23]. Dissimilarities between the present and
previous findings [15, 23] probably represent racial dif-
ferences, as well as sample size and age range differences.
The minimum age of the subjects chosen for this study
was based on the available evidence reporting no signifi-
cant change in the first molar and canine arch widths after
the age of 13 in females and 16 in males [15, 20, 24–28] as
no significant future transverse growth changes can be
expected.
The beta mathematical function was selected for this
project over other formulae due to its accuracy and the
restricted number of prerequisite dental landmarks to rep-
resent the dental arch. Unlike conic sections model, it has
no inherent limitation of fitting the arch to specific shapes,
and unlike other mathematical models, the beta function
uses only two parameters and does not utilize the remain-
ing dental landmarks. In fact, decreasing the number of
pre-requisite measurements increases the error of curve
fitting ability; but on the other hand, this decreases the
possible irregularities of achieved curvatures [4]. As it is
stated in the literature, the reduction in the amount of error
should not be the only factor to be considered for selection
of the preferred type of mathematical model. The selection
of preferred mathematical model should be based on the
objectives and applications; this characteristic can serve as
an advantage in specific clinical situations such as in cleft
patients, where the intercanine area and the anterior seg-
ment of the upper arch are strongly affected by the pres-
ence of alveolar defect.
Another specific characteristic of beta-function mathe-
matical model is that it does not reflect asymmetry [4]. In
fact, one can take advantage of this inherent symmetrical
function in specific clinical situations like unilateral cleft
lip and palate, in which we might need to construct a
symmetric ideal model [4].
Mandibular inter-canine and inter-molar widths are
accurate indices of the muscular balance inherent in each
patient and dictate the limits of arch expansion during
treatment. These are important to establish and maintain
for proper functional balance and as an aid in treating and
retaining orthodontic patients [29]. An analytical equation
of dental arch shape is necessary to describe the relation-
ship between arch width and depth, which influence the
235
arch form [30]. Studies of retention and relapse recommend
maintenance of the original intercanine and intermolar
width when possible [29]. Arch length is determined by the
sum of the mesiodistal widths of all teeth in the arch; an
increased width would result in a decreased arch depth,
which can impact lip support and facial esthetics [30].
The occlusal views of dental arches in Angle Class I
occlusions have been described previously by the math-
ematical beta function with average correlation coeffi-
cients of 0.98 for the mandible and 0.97 for the maxilla
[9]. The beta function is shown to be an accurate repre-
sentation of planar projections of natural arch forms
defined by spatial coordinates of labial landmarks in
maxillary and mandibular arches [31]. In the present
study, the coefficient of determination measured by
means of adjusted r-square measure also reported high
level of total arch fitness among Iranian subjects. The
reported RMS measures, particularly at the anterior area
(0.92 mm), compared to the total arch measure
(1.49 mm) were within the acceptable range. This
implies the minimization of the postulated drawback of
beta function mathematical formulation.
Although Braun recommended the use of intercanine
values for checking the adaptation of beta-function to the
dental arch, the anterior curvature is a better determi-
nation of the differences between proposed model and
actual FA points of anterior teeth based on three reasons
[9]. (1) Highest variation of the scanning process at the
canine area, based on a previous study [14], (2) The
canine’s location in the curvature; which is at the region
that the slope of dental arch is changed. (3) Its sequence
of eruption; that is the latest tooth erupted in maxillary
arch causing it’s displacement to be the most in com-
parison to other teeth. Therefore, the accuracy of pre-
sented formula at anterior area (from canine to canine)
was reported by means of same accuracy measures as the
overall arch, using integration (Adjusted r square and
RMS) (Table 6; Fig. 2b). This way, we demonstrated a
better estimation for the fitness of the presented model
on six anterior FA points, instead of only considering
canine dimension.
Using integral calculation also revealed the accuracy of
prediction of an ideal maxillary arch form using
mandibular measurements; which can simplify the clinical
diagnostic setup, treatment plan simulations, and ultimately
custom-made arch wire production in orthodontic patients
with dentofacial deformity. Furthermore, by considering
the upper and lower limits of observed IMW differences
between the maxillary and mandibular arches, one spatial
envelope of acceptable difference variations can be drawn
based on individual mandibular measures for each patient,
that can guide the clinicians to judge and select the
appropriate treatment modality for each patient.
123
236
Conclusion
Based on the present findings, the mean values for maxil-
lary IMW and IMD were greater than the corresponding
mandibular values, but no gender dimorphism was present.
The proposed beta function formula used for predicting
maxillary arch form based on two mandibular descriptive
measures was found to have a high level of accuracy for
maxillary arch prediction and may be used as a guide to
fabricate customized arch wires or used as an aid in max-
illary reconstructive surgery.
Acknowledgments This study was extracted from a postgraduate
thesis at the School of Dentistry, Shahid Beheshti University of
Medical Sciences. The authors wish to thank the Dentofacial Defor-
mities Research Center, Research Institute of Dental Sciences, Shahid
Beheshti University of Medical Sciences for the support of this project.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
References
1. Ronay V, Miner RM, Will LA, Arai K. Mandibular arch form: the
relationship between dental and basal anatomy. Am J Orthod
Dentofacial Orthop. 2008;134:430–8.
2. Felton MJ, Sinclair PM, Jones DL, Alexander RG. A computer-
ized analysis of the shape and stability of mandibular arch form.
Am J Orthod. 1987;92:478–83.
3. De La Cruz AR, Sampson P, Little RM, Artun J, Shapiro PA.
Long term changes in arch form after orthodontic treatment and
retention. Am J Orthod. 1995;107:518–30.
4. AlHarbi S, Alkofide EA, AlMadi A. Mathematical analyses of
dental arch curvature in normal occlusion. Angle Orthod.
2008;78:281–7.
5. BeGole EA. Application of the cubic spline function in the
description of dental arch form. J Dent Res. 1980;59:1549–56.
6. Sampson PD. Dental arch shape: a statistical analysis using conic
sections. Am J Orthod. 1981;79:535–48.
7. Pepe SH. Polynomial and catenary curve fits to human dental
arches. J Dent Res. 1975;54:1124–32.
8. Ferrario VF, Sforza C, Miani A Jr, Tartaglia G. Mathematical
definition of the shape of dental arches in human permanent
healthy dentitions. Euro J Orthod. 1994;16:287–94.
9. Braun S, Hnat WP, Fender DE, Legan HL. The form of the
human dental arch. Angle Orthod. 1998;68:29–36.
10. Maeda T, Oyama A, Okamoto T, Funayama E, Furukawa H,
Hayashi T, Murao N, Sato Y, Yamamoto Y. Combination of
Tessier clefts 3 and 4: case report of a rare anomaly with 12
years’ follow-up. J Craniomaxillofac Surg. 2014;42:1985–9.
11. Kim J, Lagravére MO. Accuracy of Bolton analysis measured in
laser scanned digital models compared with plaster models (gold
123
Odontology (2017) 105:229–236
standard) and cone-beam computer tomography images. Korean J
Orthod. 2016;46:13–9.
12. Stewart J. Essential Calculus. In: Applications of Integration. Arc
length, chap 7, sec 7.4. 2011. p. 392–98.
13. Quimby ML, Vig KW, Rashid RG, Firestone AR. The accuracy
and reliability of measurements made on computer-based digital
models. Angle Orthod. 2004;74:298–303.
14. Nouri M, Massudi R, Bagheban AA, Azimi S, Fereidooni F. The
accuracy of a 3-D laser scanner for crown width measurements.
Aust Orthod J. 2009;25:41–7.
15. Kuntz TR, Staley RN, Bigelow HF, Kremenak CR, Kohout FJ,
Jakobsen JR. Arch widths in adults with Class I crowded and
Class III malocclusions compared with normal occlusions. Angle
Orthod. 2008;78:597–603.
16. Al-Khateeb SN. Abu Alhaija ES. Tooth size discrepancies and
arch parameters among different malocclusions in a Jordanian
sample. Angle Orthod. 2006;76:459–65.
17. Uysal T, Usumez S, Memili B, Sari Z. Dental and alveolar arch
widths in normal occlusion and Class III malocclusion. Angle
Orthod. 2005;75:809–13.
18. Slaj M, Spalj S, Pavlin D, Illes D, Slaj M. Dental archforms in
dentoalveolar Class I, II and III. Angle Orthod. 2010;80:919–24.
19. Suk KE, Park JH, Bayome M, Nam Y-O, Sameshima GT, Kook
Y-A. Comparison between dental and basal arch forms in normal
occlusion and Class III malocclusions utilizing cone-beam com-
puted tomography. Korean J Orthod. 2013;43(1):15–22.
20. Bishara SE, Jakobsen RJ, Treder J, Nowak A. Arch width
changes from 6 weeks to 45 years of age. Am J Orthod Dento-
facial Orthop. 1997;111:401–9.
21. Henrikson J, Persson M, Thilander B. Long-term stability of
dental arch form in normal occlusion from 13 to 31 years of age.
Eur J Orthod. 2001;23:51–61.
22. Burris BG, Harris EF. Maxillary arch size and shape in American
blacks and whites. Angle Orthod. 2000;70:297–302.
23. Chang HF, Shiau YY, Chen KC. The relationship of dental
crowding to tooth size, dental arch width, and arch depth. Proc
Natl Sci Counc Repub China B. 1986;10:229–35.
24. Moorrees CFA, Gron AM, Lebret LML, Yen PKJ, Frohlich FJ.
Growth studies of the dentition: a review. Am J Orthod.
1969;55:600–16.
25. Borzabadi-Farahani A. Orthodontic considerations in restorative
management of hypodontia patients with endosseous implants.
J Oral Implantol. 2012;38:779–91.
26. Sillman J. Dimensional changes of the dental arches: longitudinal
study from birth to 25 years. Am J Orthod. 1964;50:824–42.
27. Knott VB. Longitudinal study of dental arch widths at four stages
of dentition. Angle Orthod. 1972;42:387–94.
28. DeKock WH. Dental arch depth and width studied longitudinally
from 12 years of age to adulthood. Am J Orthod. 1972;62:56–66.
29. Lyotard N, Hans M, Nelson S, Valiathan M. Short-term post
orthodontic changes in the absence of retention. Angle Orthod.
2010;80:1045–50.
30. Proffit WR, Fields H. Contemporary orthodontics Mosby. St
Louis. 2000;78(124):315.
31. Braun S, Hnat WP, Leschinsky R, Legan HL. An evaluation of
the shape of some popular nickel titanium alloy preformed arch
wires. Am J Orthod Dentofacial Orthop. 1999;116:1–12.