ORIGINAL ARTICLE

Dental models made with an intraoral scanner:

A validation study
Anne Margreet R. Cuperus,a Marit C. Harms,b Frits A. Rangel,c Ewald M. Bronkhorst,d Jan G. J. H. Schols,e
and K. Hero Breuningf
Nijmegen, The Netherlands

Introduction: Our objectives were to determine the validity and reproducibility of measurements on stereolitho-
graphic models and 3-dimensional digital dental models made with an intraoral scanner. Methods: Ten dry
human skulls were scanned; from the scans, stereolithographic models and digital models were made. Two
observers measured transversal distances, mesiodistal tooth widths, and arch segments on the skulls and
the stereolithographic and digital models. All measurements were repeated 4 times. Arch length discrepancy
and tooth size discrepancy were calculated. Statistical analysis was performed by using paired t tests.
Results: For the measurements on the stereolithographic and digital models, statistically significant differences
were found. However, these differences were considered to be clinically insignificant. Digital models had fewer
statistically significant differences and generally the smallest duplicate measurement errors compared with the
stereolithographic models. Conclusions: Stereolithographic and digital models made with an intraoral scanner
are a valid and reproducible method for measuring distances in a dentition. (Am J Orthod Dentofacial Orthop
2012;142:308-13)

S
uccessful treatment planning in dentistry re-
quires precise diagnostic information and an
extensive diagnosis. In orthodontics, dental
model analysis is an essential part of this process.1 Den-
tal models can be used to evaluate occlusion and
perform measurements more easily and accurately
than in the patient's mouth.2 Measurements typically
made for orthodontic diagnosis are overjet, overbite,
tooth sizes, arch lengths, and transversal distances.3
Space can be analyzed by calculating the arch length
discrepancy.4 Disproportions among sizes of maxillary
and mandibular teeth can be defined by using the tooth
size discrepancy calculations according to Bolton.5 For
dental model analysis, plaster models have been the
standard for years. However, plaster models are subject
From Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.
a
Orthodontist, Department of Orthodontics and Craniofacial Biology.
b
Orthodontist, Department of Orthodontics and Craniofacial Biology.
c
Orthodontist and PhD student, Department of Orthodontics and Craniofacial Bi-
ology.
d
Associate professor, Department of Preventive and Curative Dentistry.
e
Chef de clinique, Department of Orthodontics and Craniofacial Biology.
f
Assistant professor, Department of Orthodontics and Craniofacial Biology.
The authors report no commercial, proprietary, or financial interest in the prod-
ucts or companies described in this article.
Reprint requests to: K. Hero Breuning, Radboud University Nijmegen
Medical Centre, Postbox 9101, 6500 HB Nijmegen, The Netherlands;
e-mail, h.breuning@dent.umcn.nl.
Submitted, October 2011; revised and accepted, March 2012.
0889-5406/$36.00
Copyright Ó 2012 by the American Association of Orthodontists.
doi:10.1016/j.ajodo.2012.03.031
308
to loss, fracture, and degradation and require physical
storage space.6,7 To overcome these disadvantages,
3-dimensional digital dental models are an alternative.
Additional advantages of digital models are easy
storage and exchange with electronic data transfer.
Digital models can be virtually manipulated, precise
cross-sectional views can be created, and they can be
magnified.8 Commercially available digital models can
be produced by a direct or an indirect method. Indirect
methods begin with dental impressions. Digital models
can then be obtained by laser scanning of plaster
models or computed tomography imaging of the
impressions or plaster models. The direct method uses
an intraoral scanner to scan directly in the patient's
mouth, making dental impressions redundant.8,9 This
can be advantageous for patients with a gag reflex or
with cleft lip and palate, who are at risk of aspiration
and respiratory distress during taking of the dental
impressions.10,11 Recently, the validity of digital
models produced with an indirect method was
evaluated in a systematic review by assessing the
agreement of measurements on digital and plaster
models.12 It was concluded that digital models offer
a high degree of validity, and measurement differences
are likely to be clinically acceptable. To our knowledge,
no study has assessed the validity of measurements on
digital models obtained with a direct method with an
intraoral scanner. Therefore, the aim of this study
was to determine the validity and reproducibility of
Cuperus et al
Fig 1. Examples of a skull model, a stereolithographic model, and a digital model. SM,
Stereolithographic model; DM, digital model.
orthodontic measurements made on stereolithographic
and digital models created with an intraoral scanner.
MATERIAL AND METHODS
Ten dry human skulls were selected from the files of
the Department of Orthodontics and Craniofacial Biol-
ogy at Radboud University Medical Centre, Nijmegen,
The Netherlands. The inclusion criteria were (1) full
permanent dentition from the left first molar to the right
first molar in both jaws, with 1 missing or deciduous
tooth per skull accepted, and (2) dentition without
abnormal morphology, with defects of teeth accepted
when they did not affect the mesiodistal or buccolingual
diameter of the crown.
The dentitions were scanned with a chair-side oral
scanner (Lava, 3M ESPE, Seefeld, Germany) according
to the manufacturer's manual to create stereolitho-
graphic files. After scanning, the files of the scanner
were sent to the company by e-mail. This company
uses a computer program to correct for missing data,
and the files were returned to our university by the Inter-
net. These files were then transformed into digital
models by software (OrthoProof, Nieuwegein, The Neth-
erlands). A dental laboratory (Oosterwijk, Utrecht, The
Netherlands) used the same files to produce stereolitho-
graphic models with a 3-dimensional printer, made from
American Journal of Orthodontics and Dentofacial Orthopedics
309
epoxy resin. An example of the 3 models used in this
study is shown in Figure 1.
Two experienced observers (A.M.R.C. and M.C.H.)
measured the dentitions on the skulls on the stereolitho-
graphic and digital models. A digital caliper (Mauser,
Winterthur, Switzerland) was used for the measurements
on the dentition and the stereolithographic models. The
beaks of the digital caliper were sharpened to a fine edge
to improve measurement accuracy. All measurements on
the digital models were made by using the software pro-
gram DigiModel (version 2.3.6; OrthoProof). The digital
models could be rotated, magnified, and sliced during
measuring. The computer screens used were 15.4 in,
with a screen resolution of 1280 3 800 pixels and 32-
bit color. The observers repeated all measurements 4
times with a minimum interval between repetitions of
a week.
The following measurements were made on the skulls
and the stereolithographic and digital models: intermo-
lar distances, intercanine distances, mesiodistal tooth
widths, and dental arch segments. There were recorded
to the nearest 0.01 mm. The definitions of these mea-
surements are given in the Table.
All measurements were made in both arches, and the
mesiodistal widths were measured for all incisors,
canines, premolars, and first molars. The arch length
September 2012  Vol 142  Issue 3
310 Cuperus et al

Table. Measurement definitions

Measurement
Intermolar distance
Intercanine distance
Mesiodistal tooth width
Dental arch segments
Definition
Transversal distance between the mesiobuccal cusp tips of the first permanent molars
Transversal distance between the crown tips of the permanent canines
Greatest mesiodistal diameter measured perpendicular to the axis of the tooth
Segments were measured by using estimated contact points in a nicely aligned dental arch:
Incisor segment:
The estimated distal contact point of the lateral incisor to the estimated mesial contact point of the right
central incisor
Canine segment:
The estimated mesial contact point of the first premolar to the estimated distal contact point of the lateral
incisor
Premolar segment:
The estimated mesial contact point of the first molar to the estimated mesial contact point of the first
premolar

discrepancy, and the anterior Bolton and overall Bolton
ratios were calculated.
Statistical analysis
For all models, both means and duplicate measure-
ment errors of all distances were compared by using
paired t tests. This was done for both observers sepa-
rately and combined. The level of significance was set
at P #0.05.
The mean measurements on the skulls were used as the
standard values. Mean differences in measurements
between skulls and either stereographic or digital models
are informative but do not reveal how often a relatively
large deviation from the standard values occurred. There-
fore, a large deviation was counted by using arbitrarily
chosen cutoffs: 0.1 mm for mesiodistal widths, 0.2 mm
for segments, 1.0 mm for transversal distances and arch
length discrepancies, and 1.5 mm for tooth size discrep-
ancies. Deviations larger than the cutoffs were compared
by using the Fisher exact test version of the chi-square test.
RESULTS
The reproducibility of the measurements was tested.
At only 3 times, a significant difference was found, but
this difference was not statistically significant. The cor-
relation coefficients of the absolute measurements were
high (mean, 0.986; range, 0.876-0.999). We concluded
that the outcome of the measurement is independent
of the method and the observer. Therefore, only pooled
comparisons are presented.
The mean values for dental measurements, arch
length discrepancies, and tooth size discrepancies on
the skulls and the stereolithographic and digital models
were compared with the paired t test. Of the 44 variables,
37 showed statistically significant differences between
the skulls and the stereolithographic models. All of these
differences showed greater distances on the skulls than
on the stereolithographic models, except for segments
33 and 43 and the arch length discrepancy. Fifteen
statistically significant differences were found between
the measurements on the skulls and the digital models.
In these 15 cases, the distances measured on skulls
were greater, except for segment 33 and the arch length
discrepancy. When the measurements on the stereoli-
thographic and digital models were compared, in 26
cases, statistically significant differences were seen. In
all cases (except for arch length discrepancy), greater
values on the digital models than on the stereolitho-
graphic models were found. An example of the deviation
of the distances and measurements of the dentition in
the first quadrant is shown in Figure 2.
The measurement errors for the arch length and
tooth size discrepancies on the skulls and the stereoli-
thographic and digital models were determined. When
comparing these measurements on the skulls and the
stereolithographic models, 7 of the 44 measurements
were significantly different, with larger measurement
errors for the stereolithographic models. Between the
skulls and the digital models, 16 statistically significant
differences were found, with larger measurement errors
for the skulls. When the measurements on the stereoli-
thographic and digital models were compared, 31 statis-
tically significant differences were seen. Again, the
stereolithographic models had greater measurement
errors. An example of the measurement errors for the
distances and measurements in the first quadrant is
given in Figure 3.
Relatively large deviations of the measurements on
the stereolithographic and digital models by using the
cutoffs were recorded. For the transversal measure-
ments, this difference was always less than 0.1 mm for
the digital models. On the stereolithographic models,
however, 1.32% of the transversal measurements

September 2012  Vol 142  Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics
Cuperus et al
Fig 2. An example of the deviation of the distances and measurements in percentages in the first quad-
rant. SM, Stereolithographic model; DM, digital model; m-d, mesiodistal.
deviated by more than the 0.1-mm cutoff. For the
mesiodistal widths, the 0.1-mm difference was exceeded
by 11.86% on the digital models and 30.08% on the
stereolithographic models. For the segments, 7.92%
of the digital model and 10.00% of the stereolitho-
graphic model measurements deviated by more than
the 0.2-mm cutoff. The 1.0-mm cutoff for the arch
length discrepancy was exceeded in 2.50% of the stereo-
lithographic models but not on the digital models. For
tooth size discrepancy, neither the stereolithographic
nor the digital models exceeded the 1.5-mm cutoff.
DISCUSSION
The results of this study can only be compared with
studies that used indirect methods to produce digital
models, since no other validation studies for stereolitho-
graphic or digital models with an intraoral scanner have
been published. Therefore, we can only compare our
results with studies that used indirect methods to
produce digital models. When distances measured on
the digital models and the skulls were compared, high
percentages of statistically significant differences were
found, but these differences were relatively small. Signif-
icantly small differences were also found in other stud-
ies.9-11 In our study, significant but relatively small
differences on the skulls and the stereolithographic
models were found in 84.1% of the measurements.
Significant differences on the skulls and the digital
models were found in 34.1% of the measurements.
American Journal of Orthodontics and Dentofacial Orthopedics
311
Similar significant differences between measurements
were also found in other studies.13-15 For correction
of relative large deviations caused by measuring
smaller and larger distances, arbitrary cutoffs where
used. With these cutoffs, again the digital models
had smaller percentages of high deviations compared
with the stereolithographic models. Significant
measurement differences were found in 84.1% of the
stereolithographic models and in 34.1% of the
digital models. In general, the measurements on the
stereolithographic and digital models tended to be
smaller than those on the skulls. Measurements on the
stereolithographic models tended to be even smaller
than those on the digital models. Both the scanning
process and the procedure used by 3M ESPE to convert
the data from the scanner into stereolithographic files
could explain some differences between the accuracy
of the models. The restricted scanning time and the
difficulties in scanning of contact points in the dental
arch caused small amounts of missing data on the
scans. To obtain a complete stereolithographic model,
an algorithm was used by 3M ESPE to correct for
missing data. This algorithm might also have
influenced the measurements. The layered structure of
the resin as a result of the 3-dimensional printing
process and the minimal shrinkage of the epoxy resin
during the production of the stereolithographic models
(about 0.12%) will influence the accuracy of these
models.3
September 2012  Vol 142  Issue 3
312
Fig 3. An example of the measurement errors in millimeters in the first quadrant. SM, Stereolitho-
graphic model; DM, digital model; m-d, mesiodistal.
The measuring method (a digital caliper and a specific
software measuring program) could also have influenced
the measurements. The smaller values found for signif-
icant percentages of the measurements on the stereoli-
thographic models could be explained by difficult
access for the digital caliper between the teeth. Access
was easier on the skulls because of the slight mobility
of the teeth. There were no access problems for measure-
ments on the digital models because of the digital
measurement method.
When the differences of measurements on the denti-
tions for both the stereolithographic and digital models
were compared with the skulls, we concluded that these
differences were small and comparable with validation
studies on digital models produced with an indirect
method. For transversal distances, other validation
studies found differences between plaster and digital
models of 0.04 to 0.62 mm.9,12-14 Differences found in
this study were well within this range. The reported
ranges for differences in measurement comparing
mesiodistal widths in plaster and digital models are
variable.10,12,14-16 Differences in mesiodistal widths in
this study were well within the range reported in the
literature. Large variations in space analysis on plaster
and digital models have been reported. Differences for
space analysis in this study were comparable14,17 or
smaller15,16 than those published in other studies.
For tooth size discrepancy measurements, only 1
small (0.114 mm) but statistically significant difference
September 2012  Vol 142  Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics
Cuperus et al
was found. Other studies reported comparable small
mean differences for tooth size discrepancies.7,10,11
However, in 2 of these studies, the values of the tooth
size discrepancy had fairly large ranges: 0.0 to 5.6
mm10 and –5.09 to 3.75 mm.11 The measurement error
can be seen as a measurement for reproducibility. In
general, the measurement errors on the digital models
were smaller than were those on the skulls and the ster-
eolithographic models. Therefore, it can be concluded
that digital model measurements showed better repro-
ducibility than traditional caliper measurements. The
measuring method on digital models, with the possibility
to zoom in and without physical barriers during measur-
ing, could explain the smaller measurement errors in this
study. The measurement errors on the skulls were smaller
compared with the stereolithographic models, probably
because the better accessibility for the digital caliper
on the skulls positively influenced the accuracy of
the measurements. The stereolithographic and digital
models used in this study seem to be clinically acceptable
for measuring distances and segments in a dentition.
The differences fell within ranges that were considered
clinically acceptable in other studies.7,9-17 However, no
studies with which to compare our data were found.
CONCLUSIONS
Based on this study, we concluded that measure-
ments on stereolithographic models and digital models
Cuperus et al
made with an intraoral scanner are valid and reproduc-
ible. The statistically significant differences in absolute
measurements and measurement errors for the measure-
ments on the digital models were generally smaller than
were those for the stereolithographic models.
We thank 3M ESPE, Dental laboratorium Oosterwijk,
and OrthoProof for their cooperation in this research.
REFERENCES
1. Han UK, Vig KW, Weintraub JA, Vig PS, Kowalski CJ. Consistency
of orthodontic treatment decisions relative to diagnostic records.
Am J Orthod Dentofacial Orthop 1991;100:212-9.
2. Bolton WA. Disharmony in tooth size and its relation to the analysis
and treatment of malocclusion. Angle Orthod 1958;28:113-30.
3. Rheude B, Sadowsky PL, Ferriera A, Jacobson A. An evaluation of
the use of digital study models in orthodontic diagnosis and treat-
ment planning. Angle Orthod 2005;75:300-4.
4. Fleming P, Marinho V, Johal A. Orthodontic measurements on
digital study models compared with plaster models: a systematic
review. Orthod Craniofac Res 2011;141:1-16.
5. Horton HM, Miller JR, Gaillard PR, Larson BE. Technique compar-
ison for efficient orthodontic tooth measurements using digital
models. Angle Orthod 2010;80:254-61.
6. Mah J, Hatcher D. Current status and future needs in craniofacial
imaging. Orthod Craniofac Res 2003;6:10-6.
7. White AJ, Fallis DW, Vandewalle KS. Analysis of intra-arch and
interarch measurements from digital models with 2 impression
materials and a modeling process based on cone-beam computed
tomography. Am J Orthod Dentofacial Orthop 2010;137:
456.e1-9.
American Journal of Orthodontics and Dentofacial Orthopedics
313
8. Pani SC, Hedge AM. Impressions in cleft lip and palate—a novel two
stage technique. J Clin Pediatr Dent 2008;33:93-6.
9. Santoro M, Galkin S, Teredesai M, Nicolay OF, Cangialosi TJ.
Comparison of measurements made on digital and plaster models.
Am J Orthod Dentofacial Orthop 2003;124:101-5.
10. Watanabe-Kanno GA, Abr~ao J, Miasiro Junior H, Sanchez-Ayala A,
Lagravere MO. Reproducibility, reliability and validity of measure-
ments obtained from Cecile3 digital models. Braz Oral Res 2009;
23:288-95.
11. Yourtee D, Emery J, Smith RE, Hodgson B. Stereolithographic
models of biopolymers. J Mol Graph Model 2000;18:26-8,
59-60.
12. Keating AP, Knox J, Bibb R, Zhurov AI. A comparison of plaster,
digital and reconstructed study model accuracy. J Orthod 2008;
35:191-201.
13. Goonewardene RW, Goonewardene MS, Razza JM, Murray K.
Accuracy and validity of space analysis and irregularity index
measurements using digital models. Aust Orthod J 2008;24:83-90.
14. Bootvong K, Liu Z, McGrath C, H€agg U, Wong RW, Bendeus M,
et al. Virtual model analysis as an alternative approach to plaster
model analysis: reliability and validity. Eur J Orthod 2010;32:
589-95.
15. Tomassetti JJ, Taloumis LJ, Denny JM, Fischer JR Jr. A comparison
of 3 computerized Bolton tooth-size analyses with a commonly
used method. Angle Orthod 2001;71:351-7.
16. Stevens DR, Flores-Mir C, Nebbe B, Raboud DW, Heo G, Major PW.
Validity, reliability, and reproducibility of plaster vs digital study
models: comparison of peer assessment rating and Bolton analysis
and their constituent measurements. Am J Orthod Dentofacial
Orthop 2006;129:794-803.
17. Leifert MF, Leifert MM, Efstratiadis SS, Cangialosi TJ. Comparison
of space analysis evaluations with digital models and plaster dental
casts. Am J Orthod Dentofacial Orthop 2009;136:16.e1-4.
September 2012  Vol 142  Issue 3