The Saudi Journal for Dental Research (2017) 8, 11–18

King Saud University

The Saudi Journal for Dental Research

www.ksu.edu.sa
www.sciencedirect.com

ORIGINAL ARTICLE

Evaluation of virtual models (3Shape Ortho

System) in assessing accuracy and duration of
model analyses based on the severity of crowding
Syed Abid Altaf Bukhari a,*, Kaladhar A. Reddy c, Madhukar R. Reddy c,
Sheeba H. Shah b

a
Department of Orthodontics, College of Dentistry, Jazan University, Saudi Arabia
b
Department of Paediatric and Preventive Dentistry, Kothiwal Dental College, Moradabad, Uttar Pradesh, India
c
Department of Orthodontics, SVS Institute of Dental Sciences, Mahabubnagar, Telangana, India

Received 13 February 2016; revised 2 May 2016; accepted 8 May 2016

Available online 5 July 2016

KEYWORDS Abstract Objective: The purpose of this study was to determine the accuracy and time taken to
Accuracy; perform the space analysis (arch length discrepancy and Bolton’s) based on the severity of crowding
Crowding; using new virtual model software (3Shape Ortho System, 3Shape A/S, Copenhagen) with that of
Model analysis; conventional plaster models. Materials and methods: Models from 45 patients, divided into three
Virtual models groups based on the severity of crowding (group 1: <2.5 mm, group 2: 2.5–5 mm, and group 3:
>5 mm) were included in this study. The mesiodistal width of each tooth from first molar to first
molar was measured to the nearest 0.1 mm with digital calipers, and the arch length tooth mass dis-
crepancy and Bolton’s ratio were calculated for each model of every group. The total time required
to perform the space analysis was recorded in seconds using a stopwatch. This process was repeated
to record the digital measurements with the 3Shape Ortho System software. The difference between
the two methods in all the three groups was calculated, and a paired t test was used to analyze the
data. Results: The results showed no statistically significant difference between the mesiodistal
width measurements, arch length discrepancy and Bolton’s values in all the three groups. However,
the total time required to perform the space analysis between the two methods revealed statistically
significant differences, with digital models averaging more time in all the groups. Conclusion:
Virtual models can be used as an alternative to routine plaster models in model analysis procedures,
independent of severity of crowding.
Ó 2016 The Authors. Production and hosting by Elsevier B.V. on behalf of King Saud University. This is
an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

* Corresponding author.
1. Introduction
Peer review under responsibility of King Saud University.

Plaster study models have been the ‘‘gold standard” in

orthodontic diagnosis and treatment planning, and also play
Production and hosting by Elsevier a pivotal role in orthognathic surgery, restorative dentistry

http://dx.doi.org/10.1016/j.sjdr.2016.05.004
2352-0035 Ó 2016 The Authors. Production and hosting by Elsevier B.V. on behalf of King Saud University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
12 S.A.A. Bukhari et al.

and prosthodontics.1 Philipp Pfaff first described an The purpose of this study was to evaluate the accuracy of
impression-taking technique by using heated sealing wax to mesiodistal widths of each tooth and space analysis (arch
obtain a negative representation of the dental arches followed length-tooth material discrepancy and Bolton’s analysis)
by Chapin A. Harris in 1839, who advocated using a calcined between the plaster and virtual study models in all the three
plaster to fabricate casts from wax impressions.2 Later groups based on severity of crowding present (mild, moderate
advances brought about even more accurate and dimensionally and severe) by using 3Shape technologies. In addition the
stable impression materials such as elastic polyether and amount of time that is required to perform the complete space
polyvinylsiloxane; yet irreversible hydrocolloid alginate has analysis among different groups between the plaster and digi-
remained the most common impression material used in the tal models will be measured to ascertain which one is time
orthodontic office today with continued use of plaster, namely efficient.
Type II Dental Stone, for fabrication of the casts.3
Study models provide a three dimensional view of a
patient’s occlusion and are more amenable to routine measure- 2. Materials and methods
ments like tooth size, arch length, arch width, overjet, overbite,
midline discrepancy, curve of Spee etc.4 The disadvantages of The sample comprised 45 sets of maxillary and mandibular
plaster study models are a tendency to breakage, wear from study models that were undergoing orthodontic treatment in
continued measurements contributing to inaccuracy, and stor- Department of Orthodontics and Dentofacial Orthopedics,
age space required in a busy orthodontic practice.5,6 The con- SVS Institute of Dental Sciences, Mahabubnagar, Andhra
cept of three dimensional (3D) virtual orthodontic models Pradesh, India. The selection criteria were: permanent denti-
seems very promising in eliminating the problems of conven- tion erupted from first molar to first molar, no missing teeth
tional plaster models, and also to simplify the office manage- from first molar to first molar, no voids or blebs in the plaster
ment and communication between different specialties.7,8 or digital models, no fractures of the teeth on the plaster mod-
Ortho CadTM was the first company to introduce a digital els. The sample size was divided into 3 groups, each with 15
model service to the orthodontic market in early 19999 fol- sets of study models (maxillary and mandibular) based upon
lowed by e-modelsTM of Geo Digm Corp in 2001.10 Many stud- the severity of crowding present in either of the arches22,23:
ies have been done to evaluate the reliability of virtual group 1: <2.5 mm; group 2: 2.5–5 mm and group 3: >5 mm.
models,11–20 however no study has been carried out to assess All the measurements were performed by single author
the accuracy and time required to perform the space analysis (SAB), who was trained sufficiently in using both methods:
based on the severity of the crowding. We thought it would measuring with a digital caliper and manipulation and record-
be appropriate to conduct a study to evaluate whether the ing of measurements in virtual models with the software. The
severity of crowding will affect the accuracy and time duration mesiodistal width of each tooth for maxillary and mandibular
to conduct the model analysis between plaster and digital mod- models from first molar of one side to the first molar of the
els. 3Shape Ortho System* is a recently introduced 3-D repre- other side; overall arch length for the each model; arch length
sentation of a patient’s dentition on the computer screen with discrepancy (ADL) for each model; Bolton ratio for each set of
an accuracy of 20 lm tested with MitytoyoÒ gauges (Fig. 1). model and time to perform the complete space analysis for
The system utilizes a propertied laser scanner R700 which each set of models were measured in all the three groups.
projects a laser line onto the surface of the model or impres- The plaster and digital measurements were repeated for 15 sets
sion, 3-axis motion system and two high resolution charge- of randomly selected models, 5 from each group after 10 days
coupled-device cameras, one on either side of the laser that by the same examiner to check the intra-examiner error. The
observes the profile of the line as it falls on the object. The entire samples of plaster models were measured first, followed
two-camera system reduces scanning time, because less reori- by the virtual models.
entation of the model is required to capture surface detail that The data for compiling the measurement results of the
would be missed by a single camera due to shadowing.21 study were collected under the following categories:

1. The mesiodistal width of each tooth from right first molar

to left first molar was measured between the anatomic
mesial and distal contact points in each model using a
Digital calipers (model) and measurements were rounded
to the nearest 0.1 mm.
2. The arch length in plaster models were recorded by placing
a brass wire starting from mesial of 1st molar running along
the cusp tips of other teeth up to the mesial of 1st molar of
the opposite side and a metal ruler was used to measure the
dimensions for both maxillary and mandibular arches of
each group.
3. Using the M-D measurements and arch perimeter values,
arch length discrepancy of each model and Bolton analysis
of each set of models was performed manually.
4. The time required to take all measurements in plaster mod-
els and to perform the arch length discrepancy and Bolton
analysis manually was recorded in seconds using a
Figure 1 3Shape R700 Scanner with models in occlusion. stopwatch.
Evaluation of virtual models 13

5. Measurements for the digital models were usually done analysis by using 3Shape model was recorded using a
from the occlusal aspect unless both contacts could not stopwatch.
be visualized. The rotating tool and zooming tool were used
to find a precise location and more amenable view for dig-
ital measurements. Using Orthoanalyzer of the 3Shape Statistical tests
model software, the mesial and distal contact points were
identified and marked. The values were calculated automat- Descriptive statistical analysis has been carried out in the pre-
ically by the software, once the mesial and distal points sent study. Results on continuous measurements are presented
were marked (Figs. 2–4). as mean ± SD (Min–Max) and results on categorical measure-
6. The arch perimeter was recorded by marking points on ments are presented in number (%). Significance is assessed at
the maxillary and mandibular arches of the 3Shape 5% level of significance. Analysis of variance (ANOVA) has
digital models using Orthoanalyzer software’s dedicated been used to find the significance of study parameters between
tool for each model in all the three groups. The three or more groups of patients, Student t test (two tailed, inde-
values for arch perimeter were analyzed automatically pendent) has been used to find the significance of study param-
once the completion of marking the points was eters on a continuous scale between two groups (Inter group
finished. analysis) on metric parameters. Levene’s test for homogeneity
7. The time required to identify the contact points for measur- of variance has been performed to assess the homogeneity.
ing the mesiodistal width, arch perimeter and to calculate Mann Whitney U test has been used to find the significance
the arch perimeter-tooth material discrepancy and Bolton’s between two groups for parameters on non-interval scale.

Figure 2 Set point stage of analysis.

Figure 3 Define cut stage of analysis.

14 S.A.A. Bukhari et al.

Figure 4 Finish stage of analysis.

Table 1 Comparison of arch length tooth material discrepancy between plaster and digital models of maxillary and mandibular
arches in group 1.
ADL in mm Maxillary arch Mandibular arch
Plaster model Digital model P value Plaster model Digital model P value
Arch perimeter 75.60 ± 3.74 75.86 ± 3.81 0.851 64.50 ± 3.84 64.19 ± 4.12 0.832
Total tooth material 73.38 ± 4.00 73.17 ± 3.86 0.884 64.83 ± 3.59 63.92 ± 3.65 0.497
Discrepancy 2.22 ± 2.21 2.69 ± 1.97 0.541 0.33 ± 1.65 0.27 ± 1.43 0.297

3. Results statistically significant difference with a P value of 0.674 for

maxillary and 0.927 for mandibular arches (Table 3 and Figs. 9,
The results showed excellent correlation of individual mesio- 10). Comparison of Bolton’s analysis between plaster and
distal tooth measurements for both maxillary and mandibular digital models among the three groups (group 1 with 0.917
arches between plaster and digital models in each group (mild, ± 0.03 mm and 0.913 ± 0.026 mm, group 2 with 0.909 ±
moderate and severe crowding) of the study. There was no sta- 0.020 mm and 0.908 ± 0.022 mm and group 3 with
tistically significant difference between the three groups. 0.901 ± 0.02 mm and 0.898 ± 0.020 mm respectively), indi-
Arch length-tooth mass discrepancy comparison between cated no statistically significant differences among the three
the plaster and digital models for group 1 (models with crowd- compared groups (Table 4 and Fig. 11).
ing <2.5 mm) showed an average discrepancy of 2.22 However, when the total time taken to do the complete
± 2.21 mm and 2.69 ± 1.97 mm for the maxillary arch and analysis of each model (measuring the M-D with of each tooth,
0.33 ± 1.65 mm and 0.27 ± 1.43 mm for the mandibular arch perimeter length and calculation of arch perimeter-tooth
arch respectively (Table 1 and Figs. 5, 6). There was no statis- material discrepancy and Bolton’s analysis) was compared for
tically significant difference between the plaster and digital the plaster and digital models, the results showed moderate to
models in both maxillary and mandibular models (P value of strong statistically significant differences in all the three
0.541 for maxillary and 0.297 for mandibular arches). Group groups. For group 1 with an average time of 6.58
2 models (crowding 2.5–5 mm) had a discrepancy of 0.06 ± 0.41 min for plaster and 7.49 ± 1.04 min for digital models,
± 2.91 mm and 0.19 ± 3.09 mm for the maxillary and (P = 0.028) suggesting moderate statistical significant differ-
4.27 ± 1.29 mm and 4.20 ± 1.93 mm for the mandibular ences. The average time of 6.05 ± 0.19 min for plaster and
arches between plaster and digital models respectively, 7.50 ± 0.53 min for digital models for group 2 showed strong
revealed no statistically significant difference (Table 2 and statistical significant differences (P < 0.001). In group 3 the
Figs. 7 and 8). For group 3 (models with crowding >5 mm) average time for plaster and digital analysis was 6.14
the discrepancy values of 3.90 ± 3.42 mm and 4.42 ± 0.37 min and 7.57 ± 1.02 min respectively, also suggestive
± 3.21 mm for the maxillary and 6.28 ± 2.32 mm and of strong statistical significant results (P < 0.001) (Table 5
6.20 ± 2.12 mm for the mandibular also showed no and Fig. 12).
Evaluation of virtual models 15

3 0.3
2.69 0.2
2 Plaster
2.22 0.1 0.19
Plaster model model
1 0
Digital model -0.06 Digital
0 -0.1
model
PLASTER DIGITAL PLASTER DIGITAL
MODEL MODEL MODEL MODEL

Figure 5 Comparison of arch perimeter-tooth material discrep- Figure 7 Comparison of arch perimeter-tooth material discrep-
ancy in Plaster and Digital in maxillary models of group 1. ancy in Plaster and Digital in maxillary models of group 2.

0.4 -4.15
0.2 -4.2
0.27 Plaster -4.2 Plaster
0 model -4.27 model
-4.25
-0.2 -0.33 Digital
Digital -4.3
-0.4 model model
PLASTER DIGITAL PLASTER DIGITAL
MODEL MODEL MODEL MODEL

Figure 6 Comparison of arch perimeter-tooth material discrep- Figure 8 Comparison of arch perimeter-tooth material discrep-
ancy in Plaster and Digital in mandibular models of group 1. ancy in Plaster and Digital in mandibular models of group 2.

4. Discussion The results of our study for mesio-distal tooth measure-

ments showed no statistical significant differences between
Traditionally various measurements like tooth size, arch plaster and virtual models in all the three groups thus concur-
length, arch width and different model analysis can be per- ring with the findings of Quimby et al.16 and Bootvong et al.20
formed on conventional plaster models using vernier calipers However, there are some other studies which showed a statis-
or needle pointed dividers. In the new electronic era, virtual tically significant difference in the mesiodistal width measure-
models were developed as an alternative to the conventional ments between the plaster and digital models such as,
plaster models. All software’s that provide 3D image virtual Schirmer and Wiltshire12 concluded that the digitized measure-
model systems are similar with respect to what assessments ments were smaller than the manual measurements with an
can be undertaken on them but differ in the software they pro- average discrepancy in arch length of 4.7 mm in the maxilla
vide and the method of scanning the models or impressions. and 3.1 mm in the mandible and; in Mullen et al.18 study this
None of the previous studies evaluated the accuracy and time difference was 1.48 mm and 1.5 mm in maxilla and mandible
required to perform the model analysis in digital models based respectively.
on the severity of crowding present. So in this present study, Arch length tooth material discrepancy values in this study
we did an investigation to determine whether the virtual mod- did not find any statistical significant differences between the
els can be used as an alternative to plaster models in assessing plaster and virtual models in all the three groups which is in
the accuracy and duration of model analyses depending on the accordance with the findings of Quimby et al.16 and Bootvong
severity of crowding (group 1: <2.5 mm, group 2: 2.5–5 mm, et al.20, but Mullen et al.18 showed significant differences in
and group 3: >5 mm) by using the recently introduced 3Shape ALD values when comparing digital and plaster models rang-
Ortho System software from 3Shape A/S Company. ing from 1.5 ± 1.36 mm and 1.47 ± 1.55 mm for mandibular

Table 2 Comparison of arch length tooth material discrepancy between plaster and digital models of maxillary and mandibular
arches in group 2.
ADL in mm Maxillary arch Mandibular arch
Plaster model Digital model P value Plaster model Digital model P value
Arch perimeter 74.53 ± 4.69 74.76 ± 4.53 0.894 61.13 ± 4.24 61.3 ± 4.33 0.917
Total tooth material 74.75 ± 5.13 74.57 ± 5.21 0.923 65.41 ± 4.35 65.48 ± 4.28 0.962
Discrepancy 0.06 ± 2.91 0.19 ± 3.09 0.820 4.27 ± 1.29 4.20 ± 1.93 0.900
16 S.A.A. Bukhari et al.

Table 3 Comparison of Arch length tooth material discrepancy between plaster and digital models of maxillary and mandibular
arches in group 3.
ADL in mm Maxillary arch Mandibular arch
Plaster model Digital model P value Plaster model Digital model P value
Arch perimeter 71.8 ± 3.91 71.98 ± 3.93 0.902 59.73 ± 3.75 59.79 ± 3.62 0.968
Total tooth material 75.65 ± 4.68 76.35 ± 4.82 0.687 65.96 ± 3.94 65.99 ± 4.01 0.983
Discrepancy 3.90 ± 3.42 4.42 ± 3.21 0.674 6.28 ± 2.32 6.20 ± 2.12 0.927

0.917
0.92

0.913
0.915

0.909

0.908
-3.5 0.91
-3.9

0.901
0.905
Plaster

0.898
-4 model 0.9
-4.42
Digital 0.895
-4.5 model
PLASTER DIGITAL 0.89
MODEL MODEL 0.885
GROUP I GROUP II GROUP III

Figure 9 Comparison of arch perimeter-tooth material discrep- Plaster model Digital model
ancy in Plaster and Digital in maxillary models of group 3.
Figure 11 Comparison of Bolton’s analysis: ratio/value.

Table 5 Comparison of total time required for analysis.

-6.15 Total time Plaster model Digital model P value

-6.2 required (min)
-6.2 Plaster
Group 1 6.58 ± 0.41 7.49 ± 1.04 0.028*
model
-6.25 -6.28 Group 2 6.05 ± 0.19 7.50 ± 0.53 <0.001**
Digital Group 3 6.14 ± 0.37 7.57 ± 1.02 <0.001**
-6.3 model Group F = 30.915; F = 0.988;
PLASTER DIGITAL comparison P < 0.001** P = 0.381
MODEL MODEL
*
Moderately significant.
**
Strongly significant.
Figure 10 Comparison of arch perimeter-tooth material dis-
crepancy in Plaster and Digital in mandibular models of group 3.
7.59

7.5

7.5
8
6.92

7
5.81

Table 4 Comparison of Bolton’s analysis: ratio/value.

5.8

6
Bolton’s Group 1 Group 2 Group 3 P value 5
analysis
4
Plaster 0.917 0.909 0.901 0.151 3
model ± 0.03 ± 0.020 ± 0.02
2
Digital 0.913 0.908 0.898 0.253
1
model ± 0.026 ± 0.022 ± 0.020
P value 0.636 1.000 0.707 – 0
GROUP I GROUP II GROUP III

Plaster model Digital model

Figure 12 Comparison of total time required for analysis.

and maxillary respectively greater than the values of e-models.
The results of Bolton analysis in our study showed a good cor-
relation between the plaster and digital models in all the three Time, no doubt remains a critical factor in an orthodon-
groups similar to that of Paredes et al.24 study. Mullen et al.18 tist’s busy practice and performing any model analysis has
found no significant differences between the Bolton ratios cal- always been a laborious and time consuming procedure.
culated using plaster models and emodels which contradicts Hence, for any technological breakthrough in this aspect it
our results. becomes imperative to be the frontrunner rather than a
Evaluation of virtual models 17

follower in such ever growing technological advancements. with the 3Shape software being more time consuming than
Our study also took this most important factor into consider- the routine plaster study models in all the groups.
ation and compared the total time taken to perform space 4. The 3Shape Ortho System software measurements are as
analysis (i.e. measuring M-D tooth sizes and arch length accurate and reliable as the routine digital caliper measure-
including calculation of arch length tooth material discrepancy ments irrespective of severity of crowding but the amount
and Bolton analysis) between the routine plaster models and of time taken to calculate the analysis in 3Shape system
3Shape digital models in all the groups. The results of our needs to be further investigated.
study showed 6.58 ± 0.41 min and 7.49 ± 1.04 min for group
1 (P value-0.028), 6.05 ± 0.19 min and 7.50 ± 0.53 for group
2 (P < 0.001) and 6.14 ± 0.37 min and 7.57 ± 1.02 in group 3 Conflict of interest
(P < 0.001) for plaster and digital models respectively. To our
surprise this showed a markedly strong statistical and clinical None declare.
significance between the plaster and digital model methods.
The time taken for digital model analysis was higher than that
for plaster models, not in agreement with the previous studies. References
Tomassetti et al.13 found that time taken to calculate a Bolton
ratio using Quick Ceph was 1.85 min, as compared to an aver- 1. Han KU, Vig KWL, Weintraub JA, Vig PS, Kowalski CJ.
age time of 8.06 min with Vernier calipers. Mullen et al.18 com- Consistency of orthodontic treatment decisions relative to diag-
pared the time it took to calculate the Bolton ratio using nostic records. Am J Orthod Dentofacial Orthop 1991;100:212–9.
2. Glenner RA. Dental impressions. J Hist Dent 1997;45:127–30.
plaster versus e-model in seconds and found that the plaster
3. Peluso MJ, Josell SD, Levine SW, Lorei BJ. Digital models – an
models were on an average of 65.6 s slower than doing the cal- introduction. Semin Orthod 2004;10:226–38.
culation using e-models. The range was +157 s to 47 s. 4. Callahan C, Sadowsky PL, Ferreira A. Diagnostic value of plaster
The probable reason why it took more time for digital mod- models in contemporary orthodontics. Semin Orthod 2005;11
els in our study is due to the fact that we evaluated the total (2):94–7.
time required to perform space analysis as compared to other 5. Proffit WR. Contemporary orthodontics. 4th ed. Mosby; 2000. p.
studies previously done in which either M-D measurement 138.
time or analysis time was evaluated individually. The prerequi- 6. Scholz RP. Indefinite storage of orthodontic records. J Clin
site for performing the analysis in 3Shape software was to Orthod 1988;22:734–5.
mark the mesial and distal contact point for each tooth after 7. Bell A, Ayoub AF, Siebert P. Assessment of the accuracy of a
three dimensional imaging system for archiving dental study
which the software defines the cut for each tooth and then sep-
models. J Orthod 2003;30:219–23.
arates the tooth, adds root and then finalizes the procedure. 8. Mah J, Freshwater M. 3D digital dental models using laser
The main limitation for the software is this inherent tool, technology. J Clin Orthod 2003;37:101–3.
which makes performing the analysis time consuming. 9. OrthoCADTM, USA. Available at: <http://www.orthocad.com>.
However, this one complete step of marking, defining, separat- 10. Geodigm EmodelTM, USA. Available at: <http://
ing, adding root and finalizing is the only procedure that needs www.dentalemodels.com>.
to be undertaken for any type of tool to be used, whether, it is 11. Yen CH. Computer aided space analysis. J Clin Orthod
for analysis or virtual treatment setup or fabrication of any 1991;24:236–8.
appliances. The time factor for completing all this procedure 12. Schirmer UR, Wiltshire WA. Manual and computer-aided space
may seem to take a bit longer but for sure when the same soft- analysis: a comparative study. Am J Orthod Dentofacial Orthop
1997;112:676–80.
ware is employed for other orthodontic setup purposes, makes
13. Tomasserti JJ, Taloumis LJ, Denny JM, Fischer Jr JR. A
things easier and hassle-free by not redoing all the steps once comparison of 3 computerized Bolton tooth-size analyses with a
again. In the light of above results, 3Shape Ortho System commonly used method. Angle Orthod 2001;71:351–7.
can be used as an alternative to plaster study models and at 14. Garino F, Garino GB. Comparison of dental arch measurement
the same time it becomes imperative and a driving force for between stone and digital casts. World J Orthod 2002;3:250–4.
one to evaluate and compare the services of 3Shape. 15. Santro M, Galkin S, Teredesai M, Nicolay OF, Cangialosi TJ.
Comparison of measurements made on digital and plaster models.
5. Conclusions Am J Dentofacial Orthop 2003;124:101–5.
16. Quimby NL, Vig KWL, Rashid RG, Firestone AR. The accuracy
and reliability of measurements made on computer based digital
In light of the results obtained in this study, the following con- models. Angle Orthod 2004;74, 298-03.
clusions can be drawn. 17. Costalos PA, Sarraf K, Cangialosi TJ, Efstratiadis S. Evaluation
of the accuracy of digital model analysis for the American Board
1. No statistically significant difference was found in mesio- of Orthodontics objective grading system for dental casts. Am J
distal tooth measurements recorded using digital calipers Orthod Dentofacial Orthop 2005;128:624–9.
and 3Shape Ortho System in all the three groups (mild, 18. Mullen SR, Martin CA, Ngan P, Gladwin M. Accuracy of space
moderate and severe crowding). analysis with emodels and plaster models. Am J Orthod Dentofa-
cial Orthop 2007;132:346–52.
2. Comparison of arch length discrepancy and Bolton ratio
19. Leifert MF, Leifert MM, Efstratiadis SS, Cangialosi TJ. Com-
calculations between the plaster and digital models in all parison of space analysis evaluations with digital models and
the groups revealed that severity of crowding will not affect plaster dental casts. Am J Orthod Dentofacial Orthop 2009;136:16,
the accuracy and reliability of virtual models. el-4.
3. The time taken to do the measurements and calculation of 20. Bootvong K, Liu Z, McGrath C, Hagg U, Wong RWK, Bendeus
model analysis was statistically and clinically significant, M, et al. Virtual model analysis as an alternative approach to
18 S.A.A. Bukhari et al.

plaster model analysis: reliability and validity. Eur J Orthod 23. Peck H, Peck S. An index for assessing tooth shape as applied to
2010;32:589–95. mandibular incisors. Am J Orthod 1972;61, 384-01.
21. Barry Mark. The cutting edge In-office digital study models. J 24. Paredes V, Gandia JL, Cibrian R. Determination of Bolton tooth-
Orthod 2011;45:385–9. size ratios by digitization, and comparison with the traditional
22. Little RM. The irregularity index: a quantitative score of method. Eur J Orthod 2006;28:120–5.
mandibular anterior alignment. Am J Orthod 1975;68:554–63.