Clin Oral Invest (2014) 18:16871694

DOI 10.1007/s00784-013-1132-y

ORIGINAL ARTICLE

Accuracy of full-arch scans using intraoral scanners

Sebastian B. M. Patzelt & Archontia Emmanouilidi &
Susanne Stampf & Joerg R. Strub & Wael Att

Received: 23 October 2012 / Accepted: 31 October 2013 / Published online: 17 November 2013
# Springer-Verlag Berlin Heidelberg 2013

Abstract
Objectives This study aimed to evaluate the accuracy of
intraoral scanners in full-arch scans.
Materials and methods A representative model with 14 prepared abutments was digitized using an industrial scanner
(reference scanner) as well as four intraoral scanners (iTero,
CEREC AC Bluecam, Lava C.O.S., and Zfx IntraScan).
Datasets obtained from different scans were loaded into 3D
evaluation software, superimposed, and compared for accuracy. One-way analysis of variance (ANOVA) was implemented
to compute differences within groups (precision) as well as
comparisons with the reference scan (trueness). A level of
statistical significance of p <0.05 was set.
Results Mean trueness values ranged from 38 to 332.9 m.
Data analysis yielded statistically significant differences between CEREC AC Bluecam and other scanners as well as
between Zfx IntraScan and Lava C.O.S. Mean precision
values ranged from 37.9 to 99.1 m. Statistically significant
differences were found between CEREC AC Bluecam and
Lava C.O.S., CEREC AC Bluecam and iTero, Zfx Intra Scan
and Lava C.O.S., and Zfx Intra Scan and iTero (p <0.05).
Conclusions Except for one intraoral scanner system, all
tested systems showed a comparable level of accuracy
S. B. M. Patzelt (*) : A. Emmanouilidi : J. R. Strub : W. Att
Department of Prosthodontics, School of Dentistry,
University Medical Center Freiburg, Hugstetter Strasse 55,
79106 Freiburg, Germany
e-mail: sebastian.patzelt@uniklinik-freiburg.de
S. B. M. Patzelt
Department of Periodontics, School of Dentistry,
University of Maryland Baltimore, 650 West Baltimore Street,
Baltimore, MD 21201, USA
S. Stampf
Institute of Medical Biometry and Medical Informatics, Department
of Medical Biometry and Statistics, University Medical Center
Freiburg, Freiburg, Germany

for full-arch scans of prepared teeth. Further studies are

needed to validate the accuracy of these scanners under
clinical conditions.
Clinical relevance Despite excellent accuracy in single-unit
scans having been demonstrated, little is known about the
accuracy of intraoral scanners in simultaneous scans of multiple abutments. Although most of the tested scanners showed
comparable values, the results suggest that the inaccuracies of
the obtained datasets may contribute to inaccuracies in the
final restorations.
Keywords Intraoral scanner . Digital impression . Full-arch
scan . Accuracy of intraoral scanners . Precision of intraoral
scanners . Trueness of intraoral scanners

Introduction
Since the introduction of the computer-aided design/
computer-aided manufacturing technology (CAD/CAM) in
the early 1980s, continuous advancements have evolved,
and the indication spectrum has been widely expanded. In
general, CAD/CAM systems are comprised mainly of three
components: (1) the data capture component (digitizers), (2)
the design component (CAD software), and (3) the
manufacturing component (CAM). The first component, the
data acquisition process, is the basic prerequisite before further processes can take place. Here, three-dimensional data of
the anatomical structures, namely abutments and edentulous
spans, are collected using several technologies such as contact
digitization, laser scanning, or the recently introduced optical
cameras. Until recently, laboratory scanners have been dominantly used for this digitization process. With the introduction
of new intraoral scanner systems, the digitization process has
become easier. This is because intraoral digitization, termed as
computer-aided impressioning (CAI), allows the clinician to
directly acquire the data from the prepared abutments without

1688

the need to make the conventional impression, pour the cast,

and finally digitize using the laboratory scanner [1, 2]. Thus,
the time needed for the digitization as well as fabrication
process is reduced, and potential errors, such as expansion,
shrinkage, and distortion of impression materials and/or the
gypsum master model, which can evolve at different steps are
eliminated or minimized. In this context, the scientific evidence and clinical observation have shown that the currently
available impression materials provide excellent reproduction
of abutment tooth dimensions [39]. On the other hand,
inherent problems and/or disadvantages of conventional impressions still exist, such as improper tray selection, disinfection of impression, separation of impression material from the
impression tray, distortion of conventional impressions before
pouring, and storage of impressions for potential remaking of
casts and dies [10, 11]. The elimination of these steps, i.e.,
with the use of CAI, is claimed to yield scans with an accuracy
level that surpasses that of the accuracy of conventional impressions and accelerates the fabrication process. Studies
comparing the accuracy of conventional impressions and
intraoral scanners have been limited to up to three abutments [12, 13]. In full-arch and complete oral rehabilitation cases, CAI of all prepared abutments is necessary.
Moreover, the different scanning techniques (laser beam
based or light beam based), implemented in different
intraoral scanners, may yield different scanning accuracies.
So far, studies evaluating the accuracy of full-arch scans
made by intraoral scanners are not available. Such an
evaluation would provide valuable information about the
accuracy of different scanning techniques implemented in
CAI systems for full-arch scans. As different scanning
technologies are implemented, it can be hypothesized that
the contemporary scanners yield different accuracies. Therefore,
the aim of this study was to evaluate the accuracy of full-arch
scans including 14 prepared abutments using different intraoral
scanners.

Materials and methods

Reference model
Fourteen teeth of an upper-jaw study model (KaVo,
Biberach, Germany) were prepared with a circular
supragingival chamfer (posterior teeth 0.60.8-mm chamfer
and anterior teeth 0.40.6-mm chamfer). The model was
replicated using a polyurethane (PU) material (Alpa-Pur,
Shore A 70, CHT BEZEMA R. Beitlich GmbH, Tbingen,
Germany) and was used as a reference model. The PU material has an impact resistance of 25 kJ/m2, an ultimate elongation of 7 %, a tear strength of 32 N/mm2, a flexural strength of
>20.0 N/mm2, a shrinkage of 0.2 %, a thermal shape retention
of 80 C, and an E modulus of 1,300 N/mm2.

Clin Oral Invest (2014) 18:16871694

Reference scan and CAI

Four intraoral scanner systems (CEREC AC Bluecam,
CEREC 3D Service Pack V3.85 Sirona, Bensheim, Germany;
Lava C.O.S., Lava software 3.0, 3M ESPE, St. Paul, USA;
iTero, software version 4.0, Cadent Inc., Carlstadt, USA;
Zfx IntraScan, software version 0.9 RC33 2.8, MHT S.p.A.,
Arbizzano di Negrar, Italy/MHT Optic Research AG,
Niederhasli, Switzerland) as well as an industrial reference
scanner (IScan D101, Imetric 3D GmbH, Courgenay,
Switzerland) were used for the study. All scans were
made within the same day in the same room under
ambient conditions (temperature 211 C, relative humidity 553 %, and air pressure 7615 mmHg) with
the following scan order.
First, the reference model was digitized using the industrial
scanner (n =5). According to the manufacturers data, the
scanner has a point spacing of 70 m, a noise level of 5 m,
a repeatability level of 10 m, and an accuracy of 20 m
depending on the surface properties of the scanned object and
the operating temperature of 1530 C. This scanner utilizes a
white-light 3D scanning technology and provides the
possibility to scan a complete arch. After the scan
process, the generated datasets were used to validate
the manufacturers data. One dataset was randomly selected
as the reference dataset R1 for the trueness measurements of
all test scanners.
Second, the reference model was scanned with the test
intraoral scanners. An experienced dentist performed the
scans after a learning phase with the scanning devices for a
period of 1 week. Five independent scans (n =5) were performed using each scanner. The scans were carried out with an
intermission of 510 min between each scan to allow the
scanner to cool down. For the Lava C.O.S. system, a special
scanning protocol was applied. The protocol consisted of a
precalibration (Lava calibration tool, 3M ESPE, St.
Paul, USA), followed by a zick-zack scanning process
[14] and a final recalibration. All other scans were
performed in accordance with the manufacturers information. To avoid potential impurities of the reference
model due to the essential powdering process for the
CEREC AC Bluecam and the Lava C.O.S., the following scan
order was adapted:
(a)
(b)
(c)
(d)

Reference scanner (n =5)

iTero (n =5)
Zfx IntraScan (n =5)
Lava C.O.S. (Lava Powder, 3M ESPE, St. Paul, USA)+
removal of the powder with soft brush and air (n =5)
(e) CEREC AC Bluecam (Optispray, Sirona Dental Systems
GmbH, Bensheim, Germany)+removal of the powder
with soft brush and air (n =5)
(f) Reference scanner R2 (n =1)

Clin Oral Invest (2014) 18:16871694

Datasets
To standardize the data file format, datasets obtained from
different scans were converted to the Standard Tessellation
Language (STL) file format directly via the manufacturer of
the system or intraoral scanner by using a manufacturercertified software. For the CEREC system, the datasets were
converted with a CAD/CAM software (Rhinoceros, McNeel,
Seattle, WA, USA, with the plug-in Dental Shaper,
CIMSystem S.r.l., Cinisello Balsamo, Italy). For the Lava
C.O.S. system and iTero system, the datasets were converted
by the manufacturers. The iTero datasets were provided in
3shape- and Dentalwings-compatible STL files. For the Zfx
IntraScan, the files were automatically saved in the STL file
format after the scanning process.
Data evaluation and 3D comparisons
The STL files were loaded into 3D evaluation software
(Geomagic Qualify 2012, Geomagic, Morrisville, USA).
After visualization, the datasets were reviewed for data that
should be excluded during the evaluation process (e.g., obvious optical artifacts not interfering with actual surfaces and
unimportant areas). Then, the 14 abutments were separated
from the surrounding structures. By using the best-fitalgorithm method of the comparison software, all datasets
were superimposed. Overall 3D deviations (x, y, and z) between datasets obtained from the reference scanner and from
different intraoral scanners (trueness) as well as deviations
between datasets obtained from the same scanner (precision)
were calculated (Fig. 1). To validate the manufacturers data of
the reference scanner, the obtained datasets were superimposed
as well. Furthermore, the reference dataset of the final
reference scan R2 and the randomly selected reference dataset
R1 were also evaluated for deviations.
The 3D comparison software provided positive and negative discrepancies and visualized the deviations in a colorcoded image of the superimposition, representing expansions
or contraction, respectively. Only absolute values were used

1689

for the further analysis resulting in two values per scan and in
total ten values per device.
Statistical analyses
Statistical analyses were performed for absolute discrepancies
(n =10 per device). To detect differences between scans from
different devices as well as between scans within the same
device, a one-way ANOVA was fitted. A level of statistical
significance of p <0.05 was set. Least-square means of device
and interaction effects were calculated with their 95 % confidence intervals. Model assumptions, i.e., normal distribution
of residuals, were checked by evaluation of histograms and
normal probability plots. Box plots were used to illustrate
experiment results. All calculations were performed with the
statistical software SAS 9.1.2. (SAS Institute Inc., Cary, USA)
using PROC MIXED by an independent statistician.

Results
The validation of the repeatability of the reference scanner
revealed no statistically significant differences between the five
initial reference scans (n =10) and between R1 and R2 (p >
0.05). The absolute mean deviation of the reference datasets
was 3.31.7 and 1.50.7 m (R1 and R2, respectively).
The visual analysis of superimposed datasets showed deviations in the horizontal plane of the scanner datasets. Compared with the reference dataset, most datasets revealed a
horizontal expansion in the region of the molars and posterior
parts (Fig. 2). Additionally, elusive optical artifacts were
found in some datasets, especially of the CEREC AC
Bluecam, interfering with the actual scanned tooth surface.
The highest accuracy was observed in the Lava C.O.S.
(trueness 38.014.3 m; precision 37.919.1 m), whereas
the least accuracy was obtained with the CEREC AC (trueness
332.964.8 m; precision 99.137.4 m) (Table 1). Data
analysis yielded statistically significant differences for the
mean trueness values between CEREC AC Bluecam and the

Fig. 1 Illustration of the data acquisition and the 3D comparison. RM reference model, DD 15 datasets 15 of each scanning device, RD reference
dataset, 3DC 3D-comparison

1690

Clin Oral Invest (2014) 18:16871694

Fig. 2 Illustration of the horizontal expansion of the superimposed

datasets. Left: Superimposed datasets. Arrows indicate areas of the expansion. Dark red areas on the superimposed dataset represent areas of
high positive deviations (centripetal-aimed expansion), whereas dark

blue areas represent negative deviations (tested dataset is compromised

compared with control dataset). Right: Schematic representation of the
expansion direction

other scanners (p <0.05) as well as between Zfx IntraScan and

Lava C.O.S. (p <0.05) (Fig. 3) and for mean maximum trueness values between CEREC AC Bluecam and other scanners
(p <0.05) (Fig. 4).
Data analysis of the mean precision values revealed statistically significant differences between CEREC AC Bluecam
and Lava C.O.S., CEREC AC Bluecam and iTero, Zfx Intra
Scan and Lava C.O.S., and Zfx Intra Scan and iTero (p <0.05)
(Fig. 5). For maximum mean precision values, statistically
significant differences were found between CEREC AC
Bluecam and the other scanners (p <0.05) (Fig. 6).
Furthermore, statistically significant differences between
several scans of each scanner were identified (p <0.05).

accuracy values of conventional impressions and digital impressions (CEREC AC) were reported to be 12.52.5 m
(precision) and 20.42.2 m (trueness), respectively, 32.4
9.6 m (precision) and 58.615.8 m (trueness) for the
digital approach. Maximum differences of up to 170 m were
found in posterior areas [15]. The authors concluded that a
digital impression cannot fully replace the conventional impression for restorative procedures yet. Additional studies
focused on the accuracy of scans of single prepared tooth or
quadrant scans. The trueness values of single abutments
ranged between 19.2 m [16] and 27.9 m [13], while the
precision values were 10.81.8 m [16]. Moreover, trueness
and precision values of quadrant scans (canine to second
molar; CEREC AC Bluecam) were 35 and 13 m, respectively [16]. Another study reported mean positive and negative
deviations and averaged maximal positive and negative deviations of 17/1319 and 134/123 m, respectively, for
digitizing of a premolar and a molar with a chamfer preparation of a four-unit fixed dental prosthesis (FDP) [17]. However, it is unclear whether the provided values are overall
results or deviations of the single abutments. A recent study
evaluating the accuracy of FDPs showed values of 3068 m
for the marginal inaccuracy and 2988 m for the internal fit
[18]. Van der Meer et al. evaluated linear discrepancies of
intraoral scanners between cylinders screwed on implant analogs in a stone model [14]. The authors reported

Discussion
Full-arch scans using different intraoral scanners with only
three prepared teeth were previously compared in two studies.
The reported trueness values for the CEREC AC Bluecam and
the Lava C.O.S. were 4914.2 and 40.314.1 m, respectively, while precision values were 30.97.1 and 60.1
31.3 m, respectively. In contrast, trueness and precision
values of conventional full-arch impressions using polyether
showed comparable values of 5521.8 and 61.321.8 m,
respectively [12]. However, in a recent publication, the

Table 1 Mean trueness/precision and mean maximum trueness/precision values the obtained standard deviation (SD)
Scanner

iTero (3shape)
iTero (Dentalwings)
CEREC AC Bluecam
Lava C.O.S.
Zfx IntraScan
IScan D101 (reference scanner)

Trueness

Precision

Mean valuesSD (m)

Mean maximum
valuesSD (m)

Mean valuesSD (m)

Mean maximum
valuesSD (m)

49.013.6
49.614.0
332.964.8
38.014.3
73.726.6
Not determined (20 manufacturer)

1919.431138.94
1860.711121.42
4858.78227.13
1080.36899.07
2091.021218.39
Not determined

40.411.3
40.511.2
99.137.4
37.919.1
90.226.7
3.31.7 (10 manufacturer)

683.80723.08
691.90741.05
4337.57552.20
821.351002.20
1779.301317.45
12.83.6

Clin Oral Invest (2014) 18:16871694

1691

Fig. 3 Illustration of absolute

mean trueness (comparison of
scanner datasets to the reference
dataset) values (m) of
statistically significant differences
between devices (marked with
asterisks). iTero-3s 3shape STL
file, iTero-dw Dentalwings STL
file

discrepancies of 2.2 m (Lava C.O.S.) to 287.5 m (CEREC)

between the measuring points on the cylinders [14]. With the
exception of the CEREC AC Bluecam system, the majority of
obtained values in this study showed a level of accuracy that is
comparable with those reported in previous studies.
The results of this study yielded differences in trueness and
precision values between the tested scanners. The highest
Fig. 4 Illustration of absolute
mean maximum trueness
(comparison of scanner datasets
to the reference dataset) values
(m) which represent
predominantly artifacts
interfering with and distorting the
actual surfaces as well as
statistically significant differences
between devices (asterisks).
iTero-3s 3shape STL file, iTerodw Dentalwings STL file

accuracy was achieved by the Lava C.O.S system, while the

least accuracy was found with the CEREC AC Bluecam
system. The differences in accuracies between the tested systems might be related to different scanning technologies as
well as data processing algorithms. The iTero scanner uses a
parallel confocal imaging technique [19] and a red laser light
beam without the need for a reflective coating. Therefore, a

1692

Clin Oral Invest (2014) 18:16871694

Fig. 5 Illustration of absolute

mean precision (comparison of
scanner datasets of a single
device) values (m) of
statistically significant differences
between devices (asterisks).
iTero-3s 3shape STL file, iTerodw Dentalwings STL file

color wheel is needed to capture the objects [20]. Color wheels

have the disadvantages of limited wavelength selection, vibration, relatively slow switching speed, and potential image
shift [21]. These disadvantages in turn yield deviations in both
trueness and precision values. On the other hand, the CEREC
AC Bluecam is based on the basic principles of confocal
microscopy and active triangulation technique with blue
light-emitting diodes (LEDs) [2225]. Therefore, a uniform
Fig. 6 Illustration of absolute
mean maximum precision values
(comparison of scanner datasets
of a single device) values (m)
which represent predominantly
artifacts interfering with and
distorting the actual surfaces as
well as statistically significant
differences between devices
(asterisks). iTero-3s 3shape STL
file; iTero-dw Dentalwings STL
file

reflective surface, usually achieved by using a powder to coat

the structures, is needed. Active triangulation systems obtain
distance information by measuring the angles of a plane
triangle and deal with a nonuniform radiance with nonaxial
symmetry. This leads to deviations from the geometric image
such as defocusing because the radial symmetry in the pupil is
disrupted [26]. Distortion caused by the noise in the frame
buffer can lead to measurement uncertainties as well [27].

Clin Oral Invest (2014) 18:16871694

However, blue LEDs have a shorter wavelength, which can

result in a higher precision of virtual datasets [28]. The high
deviations in accuracy values of the CEREC AC Bluecam
system seem to be mainly caused by the aforementioned
technological difficulties. As for the Zfx IntraScan, it utilizes
a smart pixel sensor; a technology based on confocal microscopy combined with Moire effect detection [29]. An aspherical moveable lens is used to obtain the necessary imaging
quality in all focal planes, typically, curved surfaces. As a
result, the scanned surfaces appear distorted, and a mathematical algorithm is applied to compensate these effects [28, 29].
Thus, the quality of the compensating algorithm and the
scanning technique can influence the processing procedure
of the images as well as the accuracy of the final dataset,
which may explain the higher deviations of accuracy observed
in this system.
Finally, the Lava C.O.S. utilizes an active wavefront sampling technique with structured light projection (22 lens system, 192 blue LED cells, and three CCD sensors). Similar to
the CEREC AC Bluecam system, a coating is also required,
but only in a significantly smaller amount. Here, it is noteworthy to mention that the powdering is not used as an
antireflective coating but rather the particles are used in terms
of very small connectors through the scanning process. The
single rotating aperture built into this device avoids overlapping of images and allows projection of images at several
positions, which in turn increases the spatial resolution and
enhances the measurement sensitivity [28]. The accuracy of
active wavefront sampling devices mainly depends on the
number of sampling positions during the scanning process.
In other words, an increase of sampling positions (captured
images) leads to a higher accuracy [30].
The visual observation revealed horizontal deviations in
the distal parts of the datasets. These deviations might be
related to incorrect software stitching processes and a summation of matching errors of the captured data during processing
[15, 31, 32]. Ender and Mehl described similar findings of
highest deviations in posterior areas due to errors in software
processing [15]. Causes for elusive optical artifacts, found in
the datasets, could not be clearly identified. It can be assumed
that optical reflections may have caused these artifacts. Most
of these artifacts found in the CEREC datasets have probably
yielded the higher deviation values. These optical artifacts are represented by the maximum mean values. It is
likely that the artifacts are responsible for the high
deviations. The more optical artifacts identified during
the visual examination of the datasets, the higher the
resulting deviations were.
In addition to technology-related errors, shaking of hands
during the scanning process is considered a universal problem
contributing to measurement uncertainties and distortion of
captured images, leading ultimately to errors in datasets of all
handheld scanners [33].

1693

In regard to the reference scanner, the repeatability and the

validity of the provided manufacturers data could be verified
by superimposition of the different reference datasets. Although the trueness of the reference scanner was only
20 m, pre-investigations of our research group as well as
previous studies [12] showed results for full-arch scans not
better than 35 m. In light of this knowledge, it seems to be
sufficient to apply this reference scanner for this particular
study design and to waive the utilization of a reference scanner
with accuracy at the nanometer level, as described in other
studies [12]. Additionally, the dimensions of the full-arch
model hampered a utilization of a micro-CT. A micro-CT scan
of a volume of approximately 883 cm would provide a
maximal accuracy of 5060 m.
Previous studies used metal-based [12, 15, 18] or gypsum
[14] reference models. Indeed, a metal-based model has a
reliable durability and is resistant against mechanical irritations, but there is still the problem of optical surface reflections that may lead to scanning artifacts with all applied
scanners. Seelbach et al. overcame this problem by
sandblasting the surfaces [18]. In contrast, a gypsum model
is not reflecting; however, it is sensitive to outer influences
such as water and mechanical irritations. In the present study,
a PU reference model was used. This PU material represents a
specially developed material for high-precision dental applications. To minimize the risk of water-uptake- or temperaturerelated distortions of the model and consequently of the
datasets, all measurements were performed within the same
day under similar conditions. Furthermore, the superimposition of the dataset of the reference model after all scans (R2)
and the initial as reference-defined dataset (R1) revealed no
statistically relevant deviations. Thus, the model dimensions
were stable during the scanning processes.
In general, results of this study provide evidence that IOS
systems have a variable level of overall accuracy in digital
datasets obtained from full-arch scans. It should be highlighted that the obtained values do not provide information about
the accuracy of single abutments or the quality of restorations
produced from such data. Although demonstrated to be technology related, several other factors, such as patient and hand
movements during scanning as well as the presence of saliva
and reflections from tooth and adjacent structures, might
influence accuracy values. Hence, it should be expected that
the datasets obtained from intraoral scans have a lower level of
accuracy compared with those obtained in the current study.

Conclusions
Except for one intraoral scanner system, all tested systems
showed comparable levels of trueness and precision values in
full-arch scans of prepared teeth. Further studies are needed to
validate the accuracy of these scanners under clinical conditions.

1694
Acknowledgment The authors would like to express their gratitude to
Sirona (Bensheim, Germany), 3M ESPE (St. Paul, USA), Cadent Inc.
(Carlstadt, USA), and Zfx GmbH (Dachau, Germany) for providing
intraoral scanners. Furthermore, the authors want to thank MDT Siegbert
Witkowski and MDT Wolf Woerner (Freiburg, Germany) for their help in
data processing.
Conflict of interest The authors have no conflict of interest.

References
1. Quaas S, Rudolph H, Luthardt RG (2007) Direct mechanical data
acquisition of dental impressions for the manufacturing of CAD/
CAM restorations. J Dent 35(12):903908
2. Schoenbaum TR (2010) Decoding CAD/CAM and digital impression units. Dent Today 29(2):140, 142, 144145
3. Clancy JM, Scandrett FR, Ettinger RL (1983) Long-term dimensional stability of three current elastomers. J Oral Rehabil 10(4):325333
4. Endo T, Finger WJ (2006) Dimensional accuracy of a new polyether
impression material. Quintessence Int 37(1):4751
5. Jamani KD, Harrington E, Wilson HJ (1989) Rigidity of elastomeric
impression materials. J Oral Rehabil 16(3):241248
6. Shafa S, Zaree Z, Mosharraf R (2008) The effects of custom tray
material on the accuracy of master casts. J Contemp Dent Pract
9(6):4956
7. Shetty P, Rodrigues S (2006) Accuracy of elastomeric impression
materials on repeated pours. J Indian Prosthodont Soc 6(2):6171
8. Wostmann B, Rehmann P, Balkenhol M (2009) Accuracy of impressions obtained with dual-arch trays. Int J Prosthodont 22(2):158160
9. Zweig A (2009) Improving impressions: go digital! Dent Today
28(11):100, 102, 104
10. Christensen GJ (2008) Will digital impressions eliminate the current
problems with conventional impressions? J Am Dent Assoc 139(6):
761763
11. Christensen GJ (2009) Impressions are changing: deciding on conventional, digital or digital plus in-office milling. J Am Dent Assoc
140(10):13011304
12. Ender A, Mehl A (2011) Full arch scans: conventional versus digital
impressionsan in-vitro study. Int J Comput Dent 14(1):1121
13. Luthardt RG, Loos R, Quaas S (2005) Accuracy of intraoral data
acquisition in comparison to the conventional impression. Int J
Comput Dent 8(4):283294
14. van der Meer WJ, Andriessen FS, Wismeijer D, Ren Y (2012)
Application of intra-oral dental scanners in the digital workflow of
implantology. PLoS One 7(8):e43312. doi:10.1371/journal.pone.
0043312
15. Ender A, Mehl A (2013) Accuracy of complete-arch dental impressions: a new method of measuring trueness and precision. J Prosthet
Dent 109(2):121128. doi:10.1016/S0022-3913(13)60028-1
16. Mehl A, Ender A, Mormann W, Attin T (2009) Accuracy testing of a
new intraoral 3D camera. Int J Comput Dent 12(1):1128

Clin Oral Invest (2014) 18:16871694

17. Guth JF, Keul C, Stimmelmayr M, Beuer F, Edelhoff D
(2012) Accuracy of digital models obtained by direct and
indirect data capturing. Clin Oral Investig. doi:10.1007/
s00784-012-0795-0
18. Seelbach P, Brueckel C, Wostmann B (2012) Accuracy of digital and
conventional impression techniques and workflow. Clin Oral
Investig. doi:10.1007/s00784-012-0864-4
19. Babayoff N, Glaser-Inbari I (2000) Imaging a three-dimensional
structure by confocal focussing an array of light beams.
International Publication WO 00/08415
20. Birnbaum NS, Aaronson HB, Stevens C, Cohen B (2009) 3D digital
scanners: a high-tech approach to more accurate dental impressions.
Insid Dent 5:7074
21. Wachman ES, Niu W, Farkas DL (1997) AOTF microscope for
imaging with increased speed and spectral versatility. Biophys J
73(3):12151222
22. Brandestini M, Moermann WH (1989) Method of and apparatus for
making a prosthesis, especially a dental prosthesis. US Patent
4663720
23. Schwotzer A (2007) Measuring device and method that operates
according to the basic principles of confocal microscopy. US Patent
2007/0296959
24. Thiel F, Pfeiffer J, Fornoff P (2008) Apparatus and method
for optical 3D measurement. International Publication WO
2008/092791
25. Schmidt V (2010) 3D dental camera for recording surface structures
of a measuring object by means of triangulation. International
Publication WO 2010/012838 A1
26. Seitz G, Tiziani HJ (1996) Resolution limits of active triangulation
systems by defocusing. Opt Eng 32(6):13741383
27. Kovcs T (2004) Active triangulation scanner development focusing
on the accuracy of the detection. In: 5th International Symposium of
Hungarian Researchers: Sponsored by IEEE Computational
Intelligence Chapter, Budapest, Magyarorszg, 2004.11.11
2004.11.12. pp 183194
28. Logozzo S, Franceschini G, Kilpel A, Caponi M, Governi L, Blois L
(2011) A comparative analysis of intraoral 3D digital scanners for
restorative dentistry. Internet J Med Technol 5(1). http://ispub.com/
IJMT/5/1/10082
29. Berner M (2010) Optical system for a confocal microscope. US
Patent 2010/0085636
30. Figerio F (2006) 3-dimensional surface imaging using active
wavefront sampling (PhD Thesis). Massachusetts Institute of
Technology, Massachusetts, USA. http://dspace.mit.edu/handle/
1721.1/38258
31. Besl PJ, McKay ND (1992) A method for registration of 3-D shapes.
In: IEEE Transactions on Pattern Analysis and Machine Intelligence.
14(2):239256
32. Yang C, Medioni G (1992) Object modeling by registration of
multiple range images. Image Vis Comput 10(3):145155
33. Mada SK, Smith ML, Smith LN, Midha PS (2003) Overview of
passive and active vision techniques for hand-held 3D data acquisition. In: Shearer A, Murtagh FD, Mahon J, Whelan PF (eds) OptoIreland 2002: Optical metrology, imaging, and machine vision,
Galway, Ireland, 2002. SPIE Digital Library. http://proceedings.
spiedigitallibrary.org/proceeding.aspx?articleid=879224