In vitro comparison of the accuracy (trueness and ...of 3 main steps: surface scan-ning, restoration design, and manufacture.5 ... a rounded shoulder with a 0.6 mm radius; and an occlusal

RESEARCH AND EDUCATION

Supported byaResearcher,bAssociate PrcPrivate pracdProfessor, DeProfessor an

THE JOURNA

In vitro comparison of the accuracy (trueness and precision)of six extraoral dental scanners with different

scanning technologies
Pablo González de Villaumbrosia, DDS, PhD,a Francisco Martínez-Rus, DDS, PhD,b
Ana García-Orejas, DDS, PhD,c María Paz Salido, DDS, PhD,d and Guillermo Pradíes, DDS, PhDe

ABSTRACTStatement of problem. The fabrication of prosthetic restorations using computer-aided design andcomputer-aided manufacturing (CAD-CAM) procedures depends on scanning surfaces. However,limited information is available regarding the effect of extraoral scanning systems on theaccuracy of the fabrication process.

Purpose. The purpose of this in vitro study was to evaluate and compare the accuracy (trueness andprecision) and resolution of 6 CAD-CAMextraoral scanners by comparing features and scan technology.

Material and methods. A master die was fabricated to simulate a dental preparation. The die wasmeasured with a coordinate measuring machine (CMM) to obtain an accurate digital CAD referencemodel (CRM). The master die was then scanned 10 times with 3 structured light scanners, 2 laserscanners, and 1 contact scanner. The resulting laboratory scan data (LSD) were converted to astereolithography (STL) format. The discrepancies between measurements were compared 3-dimensionally and at 3 selected areas of a virtual sagittal cut using CAD software. TheKruskaleWallis 1-way analysis of variance was first performed to compare scanners and then togroup data according to scanner type. The Spearman rank correlation coefficient was used totest the association between resolution and all other variables (a=.05).

Results. For all 6 scanners, the mean resolution value was 133.9 (SD 93.9) points/mm2. The value fortrueness was 38.8 (SD 6.2) mm and for precision 45.5 (SD 4.8) mm. Trueness values were 20.3 mm (SD32.7) at the axial surfaces, 46.6 mm (SD 25.9) at the margin of the preparation, and 55.8 mm (SD 29.3)at the center of the occlusal groove. The ZENO Scan was the most accurate and precise of the 6scanners for most of the variables measured.

Conclusions. The reliability of CAD-CAM scanners is not affected by a specific technology (light,laser, or contact) but by definite parameters. In addition, the entire scanning procedure is moreaccurate if the scanned surfaces are smooth and regular. (J Prosthet Dent 2016;116:543-550)

Computer-aided design andcomputer-aided manufacturing(CAD-CAM) technology hasbrought a whole new range oftechniques and materials todentistry. This technology hasevolved since its introduction inthe 1980s1-3 and is now part ofdaily practice. The CAD-CAMsystems produce higher-qualityrestorations by using industri-ally prepared materials and astandardized manufacturingprocess.4 The workflow consistsof 3 main steps: surface scan-ning, restoration design, andmanufacture.5

Two main types of dentalCAD-CAM scanners exist6:intraoral scanners are usedchairside to scan patients’dental arches; extraoral scan-ners are used in the dentallaboratory to scan casts. Both
types of scanners develop a digital model of the patient’smouth using CAD software. At the end of theprocess, 3D printers or milling machines that use CAM
grant No. 263-2008 from the University Complutense of Madrid.Department of Buccofacial Prostheses, Faculty of Odontology, Universityofessor, Department of Buccofacial Prostheses, Faculty of Odontology, Untice, Madrid, Spain.epartment of Buccofacial Prostheses, Faculty of Odontology, University Cod Chairman, Department of Buccofacial Prostheses, Faculty of Odontolog

L OF PROSTHETIC DENTISTRY

technology fabricate the definitive physical model orrestoration using ceramics, polymers, metals, or othermaterials.7-12

Complutense of Madrid, Madrid, Spain.iversity Complutense of Madrid, Madrid, Spain.

mplutense of Madrid, Madrid, Spain.y, University Complutense of Madrid, Madrid, Spain.

543

http://crossmark.crossref.org/dialog/?doi=10.1016/j.prosdent.2016.01.025&domain=pdf

Figure 1. Polyetheretherketone master die.

Clinical ImplicationsThe accuracy of the extraoral CAD-CAM scannerstested was clinically acceptable. CAD-CAM dentalpreparations must be smooth (no sharp edges orundermined areas) to ensure more accurate scannerreadings.

544 Volume 116 Issue 4

Extraoral scanners use 3 main technologies: laser,structured light, and contact. As they do not requirephysical contact, optical scanners (both laser and struc-tured light) are not affected by the density of the objectbeing scanned.13 They also tend to be faster than contactscanners.14 However, contact scanners are not affectedby the surface shine, brightness, or other optical prop-erties of the object being scanned.15

Although one of the objectives of computer-aidedtechnology is to increase the accuracy of themanufacturing process,14-39 few studies have analyzedthe effect of scanning systems on the accuracy of fit,which is one of most important criteria for the long-termsuccess of restorations.40-44

The purpose of this in vitro study was to assess thereliability e resolution and accuracy e of 6 commonlyused extraoral dental scanners. The null hypothesis wasthat no differences would exist between the results inthese parameters of the scanners.

MATERIAL AND METHODS

Six commonly used extraoral CAD-CAM scanners eachtook 10 scans of a master die. Each scan was compared 3-dimensionally with a digital version of the master die andthe CAD reference model (CRM) obtained using a co-ordinate measuring machine (CMM). The resolutionused for the reference measurement was 0.01 mm.

Resolution was defined as the number of points perunit area measured by each scanner; accuracy was inter-preted as the sum of trueness and precision, with truenessbeing the degree of closeness to the true value and precisionbeing the deviation of measurements from themean value.

First, a single master die in the shape of a preparedtooth was designed using CAD software (Rhinoceros 4.0SR; Robert McNeel & Associates). This design was milledfrom a polyetheretherketone (PEEK) disk (Ketron PEEK1000; Quadrant), a material with high dimensional sta-bility and optical properties that allow an unhinderedscan of the surface.45

The master die had the following characteristics:sharp edges and undermined areas as well as smoothsurfaces; 5 mm high from the margin of the preparationto the occlusal surface; 5 degrees of occlusal convergence;a rounded shoulder with a 0.6 mm radius; and anocclusal groove with a radius of 0.6 mm (Fig. 1).

THE JOURNAL OF PROSTHETIC DENTISTRY

The master die was then measured with a CMMused in the quality control of industrial components(Global Performance 7.10.7 SF; Hexagon Metrology).This device uses a 1-mm-diameter ruby touch probe torecord the surface of an object. According to ISO 10360-2, the probe had a maximum possible error of 1.7mm at a room temperature of 18�C to 22�C. Thesemeasurements were used to create a digital version ofthe physical die, the CRM. As reported elsewhere,16-18

CMMs are considered the most accurate tools to mea-sure the dimensions of objects, despite being slowerthan other methods.19-21 CMMs are able to create anaccurate CRM and, therefore, a high validity level ofcomparison between the CRM and laboratory scandata (LSD).

Six dental CAD-CAM extraoral scanners were tested.IScan D101 (Imetric), Lava Scan ST (3M ESPE), andSmartOptics Activity 101 (Smart Optics) are structuredlight scanners; D640 (3Shape) and ZENO Scan S100(Wieland) are laser scanners; and Renishaw Incise(Renishaw) is a contact scanner. Each of the 6 scannersperformed 10 scans of the master die. These LSD wereconverted into a stereolithography (STL) format. The 60resulting files contained the 3D coordinates of everypoint scanned over the surface of the master die (Fig. 2).The scans were made using parameters regularly used indaily practice, with no modification of algorithms fortriangulation, smoothing, or software parameters.

Each LSD file was aligned with the CRM using a “bestfit alignment” mathematical algorithm. A sample size of10 000 points with a tolerance of 0.001 mm was used inthis process. Each LSD file was compared with the CRMusing 3D analysis software (Geomagic Qualify 12.1.12;Geomagic). In order to remove outliers from the sample,all LSD points more than 0.5 mm from the margin of thepreparation were rejected. Afterward, the distance fromthe surface of the CRM to every point of the LSD fileswas calculated. These distances were given a plus sign ifthe point was outside of the CRM (yellow and orange on

González de Villaumbrosia et al

Figure 2. Representative laboratory scan data image obtained from LavaScan.

October 2016 545

the color map in Fig. 3) and a negative sign if inside theCRM (blue on the color map in Fig. 3). In addition, avirtual cut was made to superimpose every LSD on theCRM. This cut was made along the major axis (as seenfrom the plan view) of the master die (Fig. 4). Threedifferent points were chosen inside the cut to measurethe distance between LSD and CRM. The locations ofthese points are shown in Figure 4.

All these data were analyzed according to thefollowing variables: resolution, trueness, precision, anddiscrepancy at the margin of the preparation, at the axialsurfaces, and at the occlusal groove (Table 1).

For each variable, all the data were grouped (withoutdistinguishing between scanners) and in 6 groups (1 foreach scanner). First, a descriptive statistical analysis wasperformed for all the variables, followed by an inferentialstatistical analysis. The null hypothesis stated that nodifference between scanners or types of scanners existsfor each of the variables. After checking for normalitywith the Kolmogorov-Smirnov tests, a Kruskal-Wallis 1-way analysis of variance was performed to comparescanners. Data were then grouped according to the typeof scannerdstructured light, laser, and contactdtocompare the 3 groups and assess the correlation betweenresolution and all other variables, using all data as awhole without any distinction between types of scanners.The statistical analysis was achieved with a significancelevel of 95% using software (SPSS v19 for Macintosh;IBM Corp).

RESULTS

The global resolution (for all 6 scanners groupedtogether) was 133.93 points/mm2 and the global truenesswas 38.8 mm. The descriptive results of the data from allthe scanners are summarized in Table 2.

The value of every variable for every scanner and thestatistical differences for the variables between the


scanners are summarized in Tables 3-5, SupplementalTables 1-3. The data were arranged in these tables inincreasing order of scanner resolution and discrepancy.Statistically significant differences were found betweensome of the scanners for some of the variables; trueness,for example, ranged from 29 mm (ZENO Scan) to 46 mm(Lava Scan) and precision from 37.5 mm (ZENO Scan) to50.6 mm (Lava Scan).

The value of every variable according to the type ofscanner and the statistical differences between the typesare summarized in Table 6. Types of scanners (laser,structured light, and contact) were arranged in increasingorder of resolution and discrepancy, revealing statisticallysignificant differences. Trueness ranged from 35.5 mm(laser) to 41.6 mm (structured light) and precision from42.1 mm (laser) to 48.3 mm (structured light).

Table 7 shows the correlation between scanner reso-lution and the other variables tested.

DISCUSSION

Every CAD-CAM process has multiple steps, with eachstep a potential source of error. As a result, each procedurein any CAD-CAM workflow is very important andcan affect overall performance.4 The objective of the pre-sent study was to assess 6 commonly used CAD-CAMlaboratory scanners that are crucial links in the chain ofdigital workflow in restoration design and manufacturing.

The null hypothesis of the present study was thatdifferences between scanners or types of scanners foreach of the variables tested (Table 1) were not statisticallysignificant. In fact, this hypothesis was rejected becausestatistically significant differences were observed.

The trueness of the 6 scanners (38.8 mm) was higherthan the accuracy declared by themanufacturers (between10 and 20 mm).39 This may be because the master die wasdesigned with sharp edges and undermined areas as wellas smooth surfaces to assess the scanners’ performanceunder extreme conditions. This then raises the question ofwhether the scanners would perform better if they werescanning an “ideal” shape, that is, one with only straightsurfaces and rounded edges. The measurements obtainedat the axial surfaces of the preparation on the virtual cut,which have a smoother surface (Supplemental Table 2),confirm that scanners do, in fact, perform better if thesurface is smoother. The mean discrepancy on the virtualcuts was smaller (20.3 mm) than that found on themarginsof the preparation (46.6 mm) or on the occlusal grooves(55.8 mm) (Table 2). Furthermore, each scanner had adiscrepancy at the axial surfaces smaller than at themargin of the preparation and occlusal grooves(Supplemental Tables 1-3). In other words, extraoralscanners are more accurate when scanning smooth sur-faces than when scanning sharp edges or underminedareas.


Figure 3. Four different views of one 3D comparison (using Geomagic Qualify software) of CAD reference model (CRM) and laboratory scan data (LSD)obtained from Renishaw Incise scan. Blue areas show LSD points inside of CRM. Yellow/orange areas show LSD points outside of CRM.


These results agree with those of other publishedstudies. Vlaar et al,22 using a calibration ball, found adiscrepancy of between 7.7 and 13.9 mm; Persson et al,16

using a CMM and a calibrated square gauge, found thetrueness of a contact scanner to be 10 mm; DeLonget al,17 using a CMM and a structured light scanner,found the range of discrepancy to be 18 mm to 30 mm;and Del Corso et al23 had similar results using a struc-tured light scanner (14 to 21 mm). The results of severalother studies of scanner trueness cannot be comparedwith those of the present study, either because differentvalidation methods were used14,15,24-33 or because pre-cision rather than trueness was assessed.13,26,34,35

Nevertheless, the trueness found in these studies wasless than that found in the present study (29 to 46 mm),possibly because of the design of the master die.

The resolution of the scanners ranging from 299.83points/mm2 to 34.29 points/mm2. This fact reasserts theneed to assess the effect of this variable on scannerperformance.

The correlation between resolution and the othervariables was irregular (Table 7). For some of the vari-ables, the correlation was positive, meaning that whenthe resolution increased, the value of the variable also


increased, while for others, the correlation was negative.The strongest correlation was with the discrepancy at themargin of the preparation (Spearman rank correlationcoefficient of −0.825). The higher the resolution, thesmaller the discrepancy at sharp edges. This findingagrees with those of other published studies, whichconclude that resolution may affect the accuracy ofscanners, especially when reading complex sur-faces.15,17,19,26,36,37 This may be because every scannerrecords the position of certain points over the surface ofan object. Noncontinuous reading of sharp edges meansthat scanners with a higher resolution tend to measureedges more accurately than those with a lower onebecause the scanner software fills in the gaps betweenpoints on the very edge of the scanned object andchanges a sharp edge into a short radius curve. However,a higher resolution does not imply a higher accuracy(trueness or precision) but only the capability ofrecording in greater detail, as can be inferred from theirregular correlation between resolution and the accuracyvariables in this study.

However, for each scanner, precision mean valueswere greater than those for trueness, meaning thattheir relative precision is low. This may be another


0.4000.3830.3660.3490.3320.3150.2980.2810.2640.2470.2300.2130.1970.1800.1630.1460.1290.1120.0950.0780.0610.0440.0270.010–0.010–0.027–0.044–0.061–0.078–0.095–0.112–0.129–0.146–0.163–0.180 D:

Dx:Dy:Dz:

C014–0.0420.002–0.001–0.042

–0.197–0.213–0.230–0.247–0.264–0.281–0.298–0.315–0.332–0.349–0.366–0.383–0.400

D:Dx:Dy:Dz:

C007

b

a

c

b

a

–0.0230.001–0.023–0.002

D:Dx:Dy:Dz:

C016–0.048–0.003–0.048–0.004

D:Dx:Dy:Dz:

C013–0.0000.000–0.000–0.000

D:Dx:Dy:Dz:

C012–0.0090.001–0.000–0.009

D:Dx:Dy:Dz:

C010–0.0320.002–0.000–0.032

D:Dx:Dy:Dz:

C008–0.0070.0010.007–0.001

D:Dx:Dy:Dz:

C013–0.0330.0020.001–0.033

D:Dx:Dy:Dz:

C0060.002–0.000–0.0010.001

D:Dx:Dy:Dz:

C005–0.023–0.000–0.015–0.017

D:Dx:Dy:Dz:

C0170.024–0.0010.0010.024

Figure 4. Example of virtual 3D cut for geometric comparison between CAD reference model (CRM) and laboratory scan data (LSD) obtained fromZENO Scan S100. The picture shows which points were selected on all cuts to measure discrepancy between CRM and LSD files: “a” points show marginof preparation discrepancy, “b” points show axial surface discrepancy, and “c” point shows occlusal groove discrepancy.

Table 1. Variables of study

Variable Description

Resolution Number of points per mm2 of LSD

Trueness (mean discrepancy) Average discrepancy (distance LSDeCRM) of all points of LSD (consideringonly their absolute value)

Precision Standard deviation of the distancesbetween points of LSD and CRM

Discrepancy at the margin ofpreparation

Average discrepancy at margin ofpreparation showed on virtual cut

Discrepancy at axial surfaces Average discrepancy at center of axialsurfaces showed on virtual cut

Discrepancy at occlusal groove Average discrepancy at lowest part ofocclusal groove showed on virtual cut

CAD, computer-aided design; CRM, CAD reference model; LSD, laboratory scan data.

October 2016 547

consequence of the design of the master die, with itssharp edges and undermined areas. Rudolph et al15

noted the presence of points with a very high discrep-ancy at every scan of an object with irregular surfaces.This may lead to less precision.

Of note, every scanner had a discrepancy value with aminus sign at the margin of the preparation, meaningthat all the readings of the scanners were inside theCRM. This would be because the scanner software joinsthe points recorded by a discontinuous reading into acurve that falls inside the object. This could lead to amisfitting definitive restoration or incomplete seating.

The contact scanner used in the present study had thehighest resolution (216.4 points/mm2) and the lowestdiscrepancy at themargin of the preparation (28.9 mm) andocclusal groove (31 mm). However, laser scanners had thebest results for trueness (35.5 mm), precision (42.1 mm), anddiscrepancy at the axial surfaces (7.3 mm).

Structured light scanners did not have the best per-formance for any variable. Nevertheless, they aregenerally considered the best scanners because they cangather much more information per scan than can othertypes. It is possible that being the biggest group ofscanners (3 of the 6 studied) tested lowered the averagefor every variable, with the worst scanners balancing outthe results of the best ones, thus yielding overall poor


results. However, when comparing the results for eachindividual model, a structured light scanner obtained thebest results for several variables, including resolution(Table 3) and discrepancy at the margin of the prepara-tion (Supplemental Table 1).

No type of scanner systematically presented the bestvalues for every variable. Even if the type of scanneraffected specific aspects of the quality of the scans,highest values were not reached consistently for everyvariable in this study. In other words, the type of scannerdid not have a major effect on the overall results, only onspecific aspects of the scanning procedure.


Table 3.Differences in resolution between scanners (Pairwise tests)

ScannerResolution

(points/mm2), Mean

P Values of Differences

3Shape Imetric Lava Scan Incise SmartOptics ZENO Scan

SmartOptics 34.3 <.001 .010 <.001 <.001 1.000 .200

ZENO Scan 70.5 .010 .200 <.001 <.001 .200 1.000

Imetric 85.5 .200 1.000 <.001 .010 .010 .200

3Shape 97.0 1.000 .200 .010 .200 <.001 .010

Incise 216.4 .200 .010 .200 1.000 <.001 <.001

Lava Scan 299.8 .010 <.001 1.000 .200 <.001 <.001

Table 2.Descriptive statistical analysis of all scanners together

VariableDiscrepancy,

Mean SD Minimum Median MaximumInterquartile

Range

Resolution (points/mm2) 133.9 93.9 33.4 90.2 300.9 145.5

Trueness (mm) 38.8 6.2 23 39.7 48.1 7.1

Precision (mean discrepancy)(mm)

45.5 4.8 33 46 54 7

Discrepancy at margin ofpreparation (mm)

46.6 25.9 12.3 39.4 112.3 27

Discrepancy at axial surfaces(mm)

20.3 29.2 0 7.3 87.5 9

Discrepancy at occlusal groove(mm)

55.8 29.3 9 52 162 41.5

SD, Standard deviation.

Table 4.Differences in trueness between scanners (Pairwise tests)

ScannerTrueness (mm),

Mean


3Shape Imetric Lava Scan Incise Smart Optics ZENO Scan

ZENO Scan 29.0 <.001 <.001 <.001 .052 .405 1.000

SmartOptics 35.7 <.001 <.001 <.001 .265 1.000 .405

Incise 37.1 .056 .010 <.001 1.000 .265 .052

3Shape 42.1 1.000 .514 .039 .056 <.001 <.001

Imetric 43.2 .514 1.000 .159 .010 <.001 <.001

Lava Scan 46.0 .039 .159 1.000 <.001 <.001 <.001

Table 5.Differences in precision between scanners (Pairwise tests)

ScannerPrecision (mm),

Mean



ZENO Scan 37.5 <.001 <.001 <.001 .073 .052 1.000

Incise 43.8 .047 <.001 <.001 1.000 .878 .073

SmartOptics 44.0 .067 <.001 <.001 .878 1.000 .052

3Shape 46.6 1.000 .091 .033 .047 .067 <.001

Imetric 50.2 .091 1.000 .654 <.001 <.001 <.001

Lava Scan 50.6 .033 .654 1.000 <.001 <.001 <.001


As mentioned before, the misfit of restorations is aconcern,40-43 and can compromise abutment teeth orperiodontal tissue and even lead to failure. Thediscrepancies measured in the present study are withina clinically acceptable level (120 mm) that wasestablished by McLean and Von Fraunhofer44 in 1979and is still commonly used as a reference. Neverthe-less, the discrepancy of 120 mm may be large enoughfor bacteria to grow, leading to biological and me-chanical problems.46,47


Manufacturers usually state that optical scanners arethe most accurate, but most milling centers have a con-tact scanner for work that requires the greatest accuracyor to use as a gold standard to calibrate their opticalscanners. Further studies to evaluate dental CAD-CAMscanners should be conducted to clarify the relativeaccuracy of the types of scanner.

Although all the tested scanners were stored inair-conditioned buildings, room temperature was notmonitored. Changes in size of up to 1 mm may have


Table 6.Differences in resolution, trueness, precision, and discrepancies on virtual cut (Pairwise tests) between types of scanner

Scanner Type Mean


Contact Laser Structured Light

Resolution (points/mm2)

Laser 83.7 .003 1.000 .509

Structured light 139.9 .009 .509 1.000

Contact 216.4 1.000 .003 .009

Trueness (mm)

Laser 35.5 .982 1.000 .005

Contact 37.1 1.000 .982 .027


Precision (mm)

Laser 42.1 .912 1.000 <.001

Contact 43.8 1.000 .912 <.001

Structured light 48.3 <.001 <.001 1.000

Discrepancy at margin of preparation (mm)

Contact 28.9 1.000 .041 .002

Laser 39.2 .041 1.000 .253


Discrepancy at axial surfaces (mm)

Laser 7.3 .014 1.000 .021

Contact 9.5 1.000 .014 .425


Discrepancy at occlusal groove (mm)

Contact 31.0 1.000 .756 <.001

Laser 34.3 .756 1.000 <.001

Structured light 78.4 <.001 <.001 1.000

Table 7. Correlation between scanner resolution and remainder of thescanner variables

VariableSpearman Rank

Correlation Coefficient P

Trueness (mean discrepancy) 0.656 <.001

Precision 0.512 <.001

Discrepancy at margin of preparation -0.825 <.001

Discrepancy at axial surfaces -0.421 <.001

Discrepancy at occlusal groove -0.166 .202

October 2016 549

occurred because of differences in temperature betweenthe controlled CMM room environment and the roomswhere the scanners were stored. In addition, the presentstudy evaluated only 1 stage of the complete CAD-CAMprocedure, which may include other possible sources oferror.

More studies are needed with respect to the completeCAD-CAM process, and in vivo studies should be per-formed to evaluate what can be considered a clinicallyacceptable misfit.

CONCLUSIONS

The present study found the following:

1. Dental preparations must have smooth surfaceswithout sharp edges or undermined areas foroptimal CAD-CAM scanner reading.


2. The resolution of a scanner affected its ability toread the sharp contours of a scanned surface but didnot affect its general trueness or precision.

3. The accuracy of every extraoral CAD-CAM scannerstudied was clinically acceptable.

4. The different technologies (light, laser, or contact)do not affect scanners’ overall reliability, but specificaspects of the scanning procedure do.

REFERENCES

1. Duret F, Preston JD. CAD/CAM imaging in dentistry. Curr Opin Dent 1991;1:150-4.

2. Strub JR, Rekow ED, Witkowski S. Computer-aided design and fabrication ofdental restorations: current systems and future possibilities. J Am Dent Assoc2006;137:1289-96.

3. Goodacre CJ, Garbacea A, Naylor WP, Daher T, Marchack CB, Lowry J. CAD/CAM fabricated complete dentures: concepts and clinical methods ofobtaining required morphological data. J Prosthet Dent 2012;107:34-46.

4. Prasad R, Al-Kheraif AA. Three-dimensional accuracy of CAD/CAM titaniumand ceramic superstructures for implant abutments using spiral scan micro-tomography. Int J Prosthodont 2013;26:451-7.

5. Freedman M, Quinn F, O’Sullivan M. Single unit CAD/CAM restorations: aliterature review. J Ir Dent Assoc 2007;53:38-45.

6. Tamim H, Skjerven H, Ekfeldt A, Ronold HJ. Clinical evaluation of CAD/CAM metal-ceramic posterior crowns fabricated from intraoral digital im-pressions. Int J Prosthodont 2014;27:331-7.

7. Kachata PR, Geissberger MJ. Dentistry a la carte: in office CAD/CAM tech-nology. J Calif Dent Assoc 2010;38:323-30.

8. Fasbinder DJ. Digital dentistry: innovation for restorative treatment.Compend Contin Educ Dent 2010. 31 Spec No 4:2-11; quiz 12.

9. Fielding GA, Bandyopadhyay A, Bose S. Effects of silica and zinc oxidedoping on mechanical and biological properties of 3D printed tricalciumphosphate tissue engineering scaffolds. Dent Mater 2012;28:113-22.

10. Ebert J, Ozkol E, Zeichner A, Uibel K, Weiss O, Koops U, et al. Direct inkjetprinting of dental prostheses made of zirconia. J Dent Res 2009;88:673-6.


http://refhub.elsevier.com/S0022-3913(16)00109-8/sref1


























11. Metzger MC, Hohlweg-Majert B, Schwarz U, Teschner M, Hammer B,Schmelzeisen R. Manufacturing splints for orthognathic surgery using athree-dimensional printer. Oral Surg Oral Med Oral Pathol Oral RadiolEndod 2008;105:e1-7.

12. Bosch G, Ender A, Mehl A. A 3-dimensional accuracy analysis of chairsideCAD/CAM milling processes. J Prosthet Dent 2014;112:1425-31.

13. Persson A, Andersson M, Oden A, Sandborgh-Englund G. A three-dimen-sional evaluation of a laser scanner and a touch-probe scanner. J ProsthetDent 2006;95:194-200.

14. Chan DC, Chung AK, Haines J, Yau EH, Kuo CC. The accuracy of opticalscanning: influence of convergence and die preparation. Oper Dent 2011;36:486-91.

15. Rudolph H, Luthardt RG, Walter MH. Computer-aided analysis of the in-fluence of digitizing and surfacing on the accuracy in dental CAD/CAMtechnology. Comput Biol Med 2007;37:579-87.

16. Persson M, Andersson M, Bergman B. The accuracy of a high-precision digitizer for CAD/CAM of crowns. J Prosthet Dent 1995;74:223-9.

17. DeLong R, Heinzen M, Hodges JS, Ko CC, Douglas WH. Accuracy of asystem for creating 3D computer models of dental arches. J Dent Res 2003;82:438-42.

18. Jemt T, Hjalmarsson L. In vitro measurements of precision of fit of implant-supported frameworks. A comparison between “virtual” and “physical” as-sessments of fit using two different techniques of measurements. ClinImplant Dent Relat Res 2012;14 Suppl 1:e175-82.

19. Tomita S, Shin-ya A, Gomi H, Shin-ya A, Yokoyama D. Machining accuracyof crowns by CAD/CAM system using TCP/IP: influence of restorative ma-terial and scanning condition. Dent Mater J 2007;26:549-60.

20. Arnetzl G, Pongratz D. Milling precision and fitting accuracy of Cerec Scanmilled restorations. Int J Comput Dent 2005;8:273-81.

21. Al-Fadda SA, Zarb GA, Finer Y. A comparison of the accuracy of fit of 2methods for fabricating implant-prosthodontic frameworks. Int J Prosthodont2007;20:125-31.

22. Vlaar ST, van der Zel JM. Accuracy of dental digitizers. Int Dent J 2006;56:301-9.

23. Del Corso M, Aba G, Vazquez L, Dargaud J, Dohan Ehrenfest DM. Opticalthree-dimensional scanning acquisition of the position of osseointegratedimplants: an in vitro study to determine method accuracy and operationalfeasibility. Clin Implant Dent Relat Res 2009;11:214-21.

24. Quaas S, Rudolph H, Luthardt RG. Direct mechanical data acquisition ofdental impressions for the manufacturing of CAD/CAM restorations. J Dent2007;35:903-8.

25. Luthardt RG, Kuhmstedt P, Walter MH. A new method for the computer-aided evaluation of three-dimensional changes in gypsum materials. DentMater 2003;19:19-24.

26. Persson AS, Andersson M, Oden A, Sandborgh-Englund G. Computer aidedanalysis of digitized dental stone replicas by dental CAD/CAM technology.Dent Mater 2008;24:1123-30.

27. Kusnoto B, Evans CA. Reliability of a 3D surface laser scanner for orthodonticapplications. Am J Orthod Dentofacial Orthop 2002;122:342-8.

28. Keating AP, Knox J, Bibb R, Zhurov AI. A comparison of plaster, digitaland reconstructed study model accuracy. J Orthod 2008;35:191-201. discus-sion 175.

29. Hayasaki H, Martins RP, Gandini LG, Saitoh I, Nonaka K. A new way ofanalyzing occlusion 3 dimensionally. Am J Orthod Dentofacial Orthop2005;128:128-32.

30. Sousa MV, Vasconcelos EC, Janson G, Garib D, Pinzan A. Accuracy andreproducibility of 3-dimensional digital model measurements. Am J OrthodDentofacial Orthop 2012;142:269-73.

31. Santoro M, Galkin S, Teredesai M, Nicolay OF, Cangialosi TJ. Comparison ofmeasurements made on digital and plaster models. Am J Orthod DentofacialOrthop 2003;124:101-5.

32. Redlich M, Weinstock T, Abed Y, Schneor R, Holdstein Y, Fischer A. A newsystem for scanning, measuring and analyzing dental casts based on a 3Dholographic sensor. Orthod Craniofac Res 2008;11:90-5.


33. Zilberman O, Huggare JA, Parikakis KA. Evaluation of the validityof tooth size and arch width measurements using conventional andthree-dimensional virtual orthodontic models. Angle Orthod 2003;73:301-6.

34. Ender A, Mehl A. Accuracy of complete-arch dental impressions: a newmethod of measuring trueness and precision. J Prosthet Dent 2013;109:121-8.

35. Kuroda T, Motohashi N, Tominaga R, Iwata K. Three-dimensional dentalcast analyzing system using laser scanning. Am J Orthod Dentofacial Orthop1996;110:365-9.

36. Persson AS, Oden A, Andersson M, Sandborgh-Englund G. Digitization ofsimulated clinical dental impressions: virtual three-dimensional analysis ofexactness. Dent Mater 2009;25:929-36.

37. Luthardt RG, Sandkuhl O, Herold V, Walter MH. Accuracy of mechanicaldigitizing with a CAD/CAM system for fixed restorations. Int J Prosthodont2001;14:146-51.

38. Nedelcu RG, Persson AS. Scanning accuracy and precision in 4 intraoralscanners: an in vitro comparison based on 3-dimensional analysis. J ProsthetDent 2014;112:1461-71.

39. Tapie L, Lebon N, Mawussi B, Fron Chabouis H, Duret F, Attal JP. Under-standing dental CAD/CAM for restorationsethe digital workflow from amechanical engineering viewpoint. Int J Comput Dent 2015;18:21-44.

40. Sachs C, Groesser J, Stadelmann M, Schweiger J, Erdelt K, Beuer F. Full-archprostheses from translucent zirconia: accuracy of fit. Dent Mater 2014;30:817-23.

41. Keul C, Stawarczyk B, Erdelt KJ, Beuer F, Edelhoff D, Güth JF. Fit of 4-unitFDPs made of zirconia and CoCr-alloy after chairside and labsidedigitalizationea laboratory study. Dent Mater 2014;30:400-7.

42. Katsoulis J, Mericske-Stern R, Rotkina L, Zbären C, Enkling N, Blatz MB.Precision of fit of implant-supported screw-retained 10-unit computer-aided-designed and computer-aided-manufactured frameworks made from zirco-nium dioxide and titanium: an in vitro study. Clin Oral Implants Res 2014;25:165-74.

43. Mously HA, Finkelman M, Zandparsa R, Hirayama H. Marginal andinternal adaptation of ceramic crown restorations fabricated withCAD/CAM technology and the heat-press technique. J Prosthet Dent2014;112:249-56.

44. McLean JW, von Fraunhofer JA. The estimation of cement film thickness byan in vivo technique. Br Dent J 1971;131:107-11.

45. Rae PJ, Brown EN, Orler EB. The mechanical properties of poly(ether-ether-ketone) (PEEK) with emphasis on the large compressive strain response.Polymer 2007;48:598-615.

46. Lang NP, Kiel RA, Anderhalden K. Clinical and microbiological effects ofsubgingival restorations with overhanging or clinically perfect margins. J ClinPeriodont 1983;10:563-78.

47. Sailer I, Fehér A, Filser F, Gauckler LJ, Lüthy H, Hämmerle CH. Five-yearclinical results of zirconia frameworks for posterior fixed partial dentures. Int JProsthodont 2007;20:383-8.

Corresponding author:Dr Guillermo PradíesDepartment of Buccofacial ProsthesesFaculty of OdontologyUniversity Complutense of MadridPlaza de Ramón y Cajal s/n.28040 MadridSPAINEmail: [email protected]

AcknowledgmentsThe authors thank Mr Ricardo García for statistical advice; and Mr Adrián Her-nández for technical support.

Copyright © 2016 by the Editorial Council for The Journal of Prosthetic Dentistry.



















































































































mailto:[email protected]

Supplemental Table 2.Differences in axial surface discrepancy between scanners (Pairwise tests)

ScannerAxial Surface

Discrepancy (mm), Mean



3Shape 2.7 1.000 .163 .165 <.001 <.001 .012

Lava Scan 5.4 .165 .995 1.000 .045 <.001 .260

Incise 9.5 <.001 .046 .045 1.000 .041 .380

ZENO Scan 12.0 .012 .263 .260 .380 .003 1.000

Imetric 12.6 .163 1.000 .995 .046 <.001 .263

SmartOptics 79.6 <.001 <.001 <.001 .041 1.000 .003

Supplemental Table 1.Differences in margin of preparation discrepancy between scanners (Pairwise tests)

ScannerMargin of PreparationDiscrepancy (mm), Mean



Lava Scan 21.1 .004 <.001 1.000 .279 <.001 .004

Incise 28.9 .077 <.001 .279 1.000 <.001 .076

3Shape 39.1 1.000 .073 .004 .077 <.001 .995

ZENO Scan 39.4 .995 .074 .004 .076 <.001 1.000

Imetric 54.0 .073 1.000 <.001 <.001 .137 .074

SmartOptics 97.2 <.001 .137 <.001 <.001 1.000 <.001

Supplemental Table 3.Differences in occlusal groove discrepancy between scanners (Pairwise tests)

ScannerOcclusal Groove

Discrepancy (mm), Mean



ZENO Scan 20.6 .009 <.001 <.001 .300 <.001 1.000

Incise 31.0 .115 <.001 .002 1.000 <.001 .300

3Shape 48.0 1.000 .024 .136 .115 <.001 .009

Lava Scan 63.7 .136 .442 1.000 .002 .081 <.001

Imetric 82.0 .024 1.000 .442 <.001 .327 <.001

SmartOptics 89.6 <.001 .327 .081 <.001 1.00 <.001

October 2016 550.e1

González de Villaumbrosia et al THE JOURNAL OF PROSTHETIC DENTISTRY

Documents

In vitro comparison of the accuracy (trueness and ...of 3 main steps: surface scan-ning, restoration design, and manufacture.5 ... a rounded shoulder with a 0.6 mm radius; and an occlusal