- Email: [email protected]
Accuracy of transferring analog dental casts to a virtual articulator
CLINICAL RESEARCH
Accuracy of transferring analog dental casts to a virtual articulator }d Úry, DMD, MSc,a Cinzia Fornai, PhD,b and Gerhard W. Weber, PhDc Elo The conventional analog ABSTRACT method of transferring patient Statement of problem. Complex digital workflows have been developed to create virtual dental information to semiadjustable patients. Direct and indirect digital methods are available for transferring analog patient or fully adjustable articulators information to virtual articulators. The direct method consists solely of digital workflows. The involves several clinical and indirect method combines analog steps and digital procedures, representing an intermediate laboratory steps (Fig. 1). These solution between the analog and direct digital approach. Studies that have investigated the overall accuracy of the virtual working space are sparse. steps are followed by a detailed static and dynamic Purpose. The purpose of this clinical study was to investigate the accuracy of the virtual dental occlusal analysis of the gypspace using the indirect digital workflow. sum casts in the programmed Material and methods. Mounted gypsum casts of 18 patients were used for indirect scanning. The articulators, forming part of maxillary casts were mounted in their skull-related position with a kinematic facebow. The the definitive diagnosis. mandibular casts were mounted in centric relation to the maxillary casts. The obtained digitized The widespread use of casts were transferred to a virtual articulator. An occlusal analysis was performed both in the analog and virtual environments, and the coordinates of matching analog and virtual contact digital technology and points were measured. The trueness and precision of the indirect transferring procedure were computer-assisted techniques assessed. in dentistry has led to considerable changes, including the Results. A total of 194 analog points was considered in the reference. Ninety-three percent of all analog points matched a virtual correspondent, and 96% of the analog first contacts between creation of the virtual dental the casts were also present as first contacts in the virtual space. The trueness of the data patient.1 Instead of the contransfer, corresponding to the spatial distance between the matching analog and virtual points, ventional procedure, direct or was 0.55 ±0.31 mm. The maximum recorded deviation was 1.02 mm. indirect digital workflows can Conclusions. The correspondence between the number and position of analog and virtual contacts be used to transfer a patient’s was high. The mean absolute deviation of the matching point-pairs was better than that reported data into a virtual dental space for the direct digital method. Under the conditions described, the virtual dental space created (VDS), where surface models with the indirect digital method can be reliably used for virtual occlusal analysis in clinical from direct digital scans of the practice. (J Prosthet Dent 2019;-:---) dental arches or indirect digital only digital data. The indirect digital workflow represents scans of the gypsum casts are placed in a virtual articuan intermediate option between the analog and direct lator. The protocol of the direct digital workflow is digital methods (Fig. 1). It includes all the conventional comparable with that of the analog procedures but uses
C.F. was financially supported by the Swiss National Science Foundation (grants no. 31003A_156299 and 31003A_176319), the Siegfried Ludwig-Rudolf Slavicek Foundation, Vienna, Austria (FA547016), A.E.R.S. Dental Medicine Organizations GmbH, Vienna, Austria (FA547013). a Postgraduate doctoral student, Department of Evolutionary Anthropology, University of Vienna, Vienna, Austria; and Private practice, ConfiDent Dental Office, Sopron, Hungary. b Postdoctoral Fellow, Institute of Evolutionary Medicine, University of Zurich, Zurich, Switzerland; and Lecturer, Department of Evolutionary Anthropology, University of Vienna, Vienna, Austria. c Associate Professor, Department of Evolutionary Anthropology, University of Vienna, Vienna, Austria; and Head of Core Facility for Micro-Computed Tomography, University of Vienna, Vienna, Austria.
THE JOURNAL OF PROSTHETIC DENTISTRY
1
2
Volume
Clinical Implications The indirect digital workflow for transferring analog data to the virtual dental space is accurate and can be used confidently for occlusal analysis.
steps before the gypsum casts are mounted in centric relation (CR) and maximal intercuspal position (ICP). Maxillary and mandibular casts are then digitized with a laboratory scanner, and their skull-related position and spatial correlation are recorded. The resulting 3D surfaces are registered in the virtual coordinate system according to their CR and ICP. The accuracy of the various phases and procedures of the digital workflow have been evaluated. The accuracy of the complete-arch digital scans has been compared with that of conventional impressions.2-12 Ender et al3 reported that highly accurate conventional impression materials provided significantly higher precision than current intraoral scanners. Güth et al8 found that the accuracy of the indirect digitalization was in the mid-range of the various direct digitalization systems tested. Ahlholm et al2 recommended the conventional impression techniques for large fixed dental prostheses, and in a literature review, Goracci et al6 concluded that up-to-date clinical evidence collected on complete-arch intraoral scanning is lacking. Virtual occlusal records in computer-aided design and computer-aided manufacturing (CAD-CAM) systems align the virtual mandibular cast to the maxillary cast. Iwaki et al13 described the dimensional stability of optical and conventional occlusal records. They reported that, for multiple restorations in a quadrant, interarch distances increased significantly from reference values with an optical interocclusal record. Solaberrieta et al14 reported that the combination of left and right lateral virtual interocclusal records (VIRs) is the most convenient for virtual articulation. The virtual articulation accuracy of indirect scanner systems was also evaluated by Yee et al.15 They compared interarch, interocclusal linear, and global virtual distances with reference values obtained from conventionally articulated gypsum casts and concluded that the different levels of distortion from reference values were due to differences in scanner accuracy and virtual articulation algorithms. Intraoral scanners provide an accurate quantitative reproduction of analog contacts in a virtual environment for casts mounted in ICP.1,16 In contrast, VIRs in CR or different therapeutic positions (TPs) are difficult to capture. From the few in vitro protocols published,16,17 it can be concluded that virtual articulation is more accurate if a VIR is registered with the conventional CR record placed intraorally rather than scanning the entire CR record extraorally.17,18
THE JOURNAL OF PROSTHETIC DENTISTRY
-
Issue
-
The virtual facebow transfer locates the maxillary surface model in the virtual articulator in its skull-related position, creating a standard coordinate system between patient data and the VDS. In this coordinate system, electronically recorded mandibular movements can be displayed for a dynamic occlusal analysis. Virtual facebow transfer protocols have been described.19-21 Based on a single patient and 6 matching points, Solaberietta et al22 reported a mean deviation of 0.752 ±0.456 mm between the virtual occlusal points and their digitized analog references using direct digital workflows. The purpose of the present study was to evaluate the global accuracy of the VDS by using the indirect digital workflow. For this purpose, spatial coordinates of analog and virtual contact points obtained from a static occlusal protocol performed in both environments were compared. The null hypothesis was that the accuracy of the indirect digital workflow would be sufficiently precise to be applied for virtual occlusal analysis in clinical practice. MATERIAL AND METHODS Mounted maxillary and mandibular gypsum casts of 18 nonedentulous patients with temporomandibular disorders attending the ConfiDent Dental Office (Sopron, Hungary) were used for indirect digitalization. All included patients received a detailed diagnosis before extensive treatment. Written consent was obtained from the patients. Factors such as sex, age, and craniomandibular relation were not considered relevant to the outcome of this study. Patient data were collected over 2 appointments. First, conventional polyvinyl siloxane impressions were made of both dental arches with a 2-step putty-wash technique (Aquasil Ultra DECA Heavy, Aquasil Ultra LV; Dentsply Sirona).23-26 Two pairs of casts were prepared per patient. The definitive casts poured in Type IV stone (GC Fujirock EP Classic; GC Corp) were equipped with a Zeiser base plate system (Zeiser Dentalgeräte GmbH). Each tooth of the definitive mandibular cast, except the central and lateral incisors, was double pinned, but not sectioned. The second pair of casts was used to prepare the acrylic resin base plate for the centric interocclusal record. In a second appointment, after a 10-minute deprogramming of the masticatory muscles,27 a CR record was made (Primobyte base plate; Primobyte detail paste; Primotec, Joachim Mosch e.K.). Then, electronic axiography was performed (CADIAX 4; GAMMA Medizinisch-wissenschaftliche Fortbildungs-GmbH), and the kinematic axis of the mandible was identified.28,29 The kinematic axis and the left infraorbital point defined an individual axis-orbital plane.30 This plane served as a reference for constructing a unitary coordinate system of the patient’s axiography data, lateral
Úry et al
-
2019
3
ANALOG METHOD
INDIRECT DIGITAL METHOD
DIRECT DIGITAL METHOD
Direct digital scan
Conventional impression
Gypsum casts
Virtual interocclusal record
ICP
Interocclusal record
Virtual surface models
CR, TP
ICP CR? TP?
Virtual articulation
Electronic condylography Kinematic facebow
Maxillary cast mounting
Virtual facebow
Face scan
ICP
Mandibular cast mounting
CR, TP Indirect digital scan digitized casts Virtual articulation
Transferring files to a virtual articulator
General medical and dental anamnesis
Clinical functional analysis
Chronic pain anamnesis
Standardized radiological analysis
Instrumental functional analysis Static and dynamic occlusal protocol
Virtual static and dynamic occlusal protocol
Diagnosis
Figure 1. Flowchart illustrating analog and digital approaches for transferring patient information to articulator. CR, centric relation; ICP, intercuspal position; TP, therapeutic position.
radiographs, cone beam computed tomography (CBCT) data, and mounted casts. Definitive casts of each patient were mounted in a fully adjustable articulator (Reference SL; GAMMA Medizinisch-wissenschaftliche Fortbildungs-GmbH) by means of the kinematic facebow and the CR record. The mounting gypsum (Artifix; Amann Girrbach AG) was allowed to harden for 24 hours. Immediately after, the casts were digitized with a high-resolution laboratory Úry et al
scanner (Activity 885; smart optics Sensortechnik GmbH) with an accuracy of 10 mm according to the manufacturer. The scanner was calibrated as indicated by the manufacturer. To register the VIR, an object holder was used to transfer the articulator-related position of the casts to the scanner (FINOSCAN RELATION Scan Fixator Gamma; Fino GmbH). The Scan Fixator had been previously synchronized with the articulator (Fig. 2). In the CAD software (dental Scan; smart optics Sensortechnik THE JOURNAL OF PROSTHETIC DENTISTRY
4
Volume
-
Issue
-
Figure 3. Virtual maxillary and mandibular surface models aligned to virtual interocclusal record.
Figure 2. Synchronization process between articulator and scan fixator. A, Calibration key in articulator. B, Calibration key in scan fixator, resetting systems to zero. C, Scan fixator holding articulator-related position of casts in scanner.
GmbH), the maxillary and mandibular casts were virtually aligned with the VIR using a best-fit algorithm (Fig. 3).31 On the same day, the mandibular cast was sectioned, and a step-by-step static occlusal analysis was performed. The first contact point between the casts was identified by rotating them against each other and was marked with 8-mm articulating paper (Arti-Fol; Dr Jean Bausch GmbH & Co KG). The contacting mandibular tooth was numbered and removed. Rotation continued THE JOURNAL OF PROSTHETIC DENTISTRY
to the next contact following the same procedure. The procedure ended when the whole sequence of contacts was registered, namely until 1 or more maxillary anterior teeth contacted the mandibular anterior teeth. The contact sequences were ordered numerically, and the contacting maxillary teeth were listed. Multiple contact points on the same tooth were differentiated according to their position on the occlusal surface. The spatial coordinates of the maxillary contact points were recorded with a calibrated mechanical 3D Digitizer (GAMMA Medizinisch-wissenschaftliche Fortbildungs-GmbH) and reported in a spreadsheet (Excel; Microsoft Corp). The sampling was repeated in 3 different sessions by 2 dentists (E.U., ZS.D.B.). The measurements were carried out with magnifying loupes and a headlight. Samplings by the same examiner on the same cast were at least 1 week apart. In each session, the maxillary contacts of a cast were measured 3 times consecutively. Thus, every contact was sampled 9 times by each operator over the 3 sampling sessions. Contacts were of different shapes and sizes and could appear in isolation or as close clusters. For standardization, each operator measured the estimated center of each contact area, regardless of its shape. When the centers of contact areas were less than 1 mm apart, the central point of the cluster rather than the central point of each area was collected. The reoriented surface models were loaded in the virtual articulator (CADIAS 3D; GAMMA Medizinischwissenschaftliche Fortbildungs-GmbH), where a virtual segmentation was performed to separate each tooth from its neighbors. As for the analog casts, a virtual static occlusal protocol was carried out with the segmented surface models in CR (Fig. 1). The segmented mandibular surface model was automatically rotated against the maxillary surface model with 0.05-degree increments. If contacts between the surfaces were detected, they were marked on the maxillary teeth as intersecting areas. The contacting antagonists were automatically removed and Úry et al
-
2019
5
Figure 4. Entire sequence of contacts. A, Different colors indicate collisions at different rotational steps. B, Contact points in order of collision. Rotational angles, incisal pin changes between consecutive collisions, as well as colliding maxillary and mandibular teeth were enumerated.
Figure 5. A, Maxillary contact points after analog static occlusal protocol. B, Automated virtual static occlusal protocol shows almost perfect point-match. Maxillary right first molar had 2 mesial contacts on analog cast, but only 1 detected on virtual surface model (blue arrows).
excluded from further calculations. The rotation continued until the last mandibular contact was detected. Contacting teeth were automatically listed in their order of contact (Fig. 4). The center of each maxillary virtual contact area was sampled twice by 2 operators (E.U. and A.P., the latter an information technology professional). To assess the precision of the 3D Digitizer, the repeatability of measuring a reference point located at the base of the digitizer was first evaluated. Between consecutive measurements (n=50), the measuring stylus was moved from its position along all the 3 spatial axes. In addition, predefined points were measured on a gypsum cast with and without removing the cast from the base of the digitizer between consecutive samplings. For both analog and virtual points, the means and standard deviations were calculated by operator and by operator and sampling session for the total of the samplings at a confidence interval of approximately 95%. Two standard deviations were reported because of their greater practical significance. Precision was divided into Úry et al
Table 1. Trueness of indirect transfer to virtual space (approximately 95% CI) Metric Analogdvirtual difference
Mean ±2 Standard Deviations (mm)
Maximum (mm)
0.55 ±0.31
1.02
CI, confidence interval.
repeatability (same operator, within a short period) and reproducibility (different operators or same operator over a long period). Repeatability is the average spatial distance of all analog points from the global centroid calculated per day and per operator. Reproducibility is the average distance of all analog points from the centroid of all days and operators together. In addition, the trueness of the transferring method was assessed, namely the spatial deviation between the analog and virtual global centroids. Bland-Altman plots were generated to visualize the interoperator error for the analog measurements as well as the analog and virtual mean deviations from the analog centroid. Excel THE JOURNAL OF PROSTHETIC DENTISTRY
6
Volume
-
Issue
-
Mean Analog Deviations Minus Mean Virtual Deviations
0.2
0.0
+1.96 SD –0.16
–0.2 Mean –0.48
–0.4
–0.6 –1.96 SD –0.80 –0.8
–1.0
0.2
0.3
0.4
0.5
Mean Analog and Virtual Deviations From Analog Centroid Figure 6. Bland-Altman plot of deviations from global analog centroid for 181 matching analog and virtual measurements. Scatter along horizontal axis reflects that some points more difficult to identify than others. All points on vertical axis had negative values, indicating that deviations of virtual points from analog centroids always larger than those of analog points to their centroids. Variation of deviation depended on magnitude of deviation itself. SD, standard deviation.
(Microsoft Corp) and R packages32 were used for statistical analysis.
Table 2. Precision of analog and virtual point measurements (approximately 95% CI) Mean ±2 Standard Deviations (mm)
RESULTS A total of 194 analog and 212 virtual points were identified. Analog points were considered as the reference. Ninety-three percent of the analog points (n=181) matched a virtual correspondent (Fig. 5). The first contact point in CR is of particular diagnostic importance as it may be a deflective occlusal interference associated with temporomandibular disorders.33 Ninety-six percent of the analog first contacts (n=47) were also found virtually. Only 2 data sets showed differences between analog and virtual first contacts. In both situations, multiple analog first contacts could only be partially reproduced by the virtual method in the first rotational step (1 of 2, and 4 of 5, respectively). The mean spatial distance between analogs and their corresponding virtual points was 0.55 ±0.31 mm, which corresponds to the trueness of the transferring method. The maximum deviation recorded was 1.02 mm (Table 1). The Bland-Altman plot shows the range of deviations of the analog and virtual mean points from the analog centroid. Values along the vertical axis were always negative, meaning that the deviation of the virtual points from the analog centroid was always higher than the deviation of the corresponding analog points to their centroid. The regular scatter of the points, aligned along a descending diagonal, showed a strong correlation
THE JOURNAL OF PROSTHETIC DENTISTRY
Metric
Operator 1
Operator 2
Average Operator 1+2
Analog reproducibility (all days)
0.073 ±0.108
0.085 ±0.123
0.079 ±0.116
Analog repeatability (all days)
0.040 ±0.085
0.048 ±0.095
Virtual repeatability (all days)
0.044 ±0.090 0.023 ±0.049
CI, confidence interval.
between the deviations of the virtual points from the analog centroid and the deviations of the analog points to their centroid (Fig. 6). The spatial difference between analog and correspondent virtual points could result from both analog and virtual sampling approaches. First, the global interoperator reproducibility for analog measurements was examined. The mean was calculated from all measurements of the 2 operators over the 3 sessions (Table 2). Overall reproducibility for both operators was 0.079 ±0.116 mm. Thus, approximately 95% of the measurements were within a sphere of 0.232 mm in diameter. The Bland-Altman plot shown in Figure 7 visualizes reproducibility, namely the deviation of the analog point measurements of the operators from the global analog centroid. Interoperator error is negligible with a mean difference of 0.013 mm. The cone-shaped scatter of the points from left to right indicates that interoperator error along the vertical axis
Úry et al
2019
7
Mean Operator 1 Minus Mean Operator 2
-
0.10
0.05
+1.96 SD 0.040
0.00
Mean –0.013
–0.05
–1.96 SD –0.066
–0.10
0.04
0.06
0.08
0.10
0.12
0.14
Grand Mean Operator 1 and Operator 2 Figure 7. Bland-Altman plot of deviations for 194 analog measurements of both operators from global analog centroid. Minimal shift of operators’ mean from zero (−0.013 mm) implied that interoperator error negligible. Only few measurements outside 2 standard deviationeconfidence interval (blue dashed lines). Rather symmetric cone-shaped scatter of points from left to right indicated lack of systematic errors. Differences between operators (vertical axis) increased with increasing deviations from centroid, meaning that interoperator error increased for points for which intraoperator error higher. SD, standard deviation.
increased for points which were more difficult to identify and thus also showed a higher intraoperator error. Second, the global intraoperator repeatability for analog measurements was calculated, which was, as expected, more precise than reproducibility with a mean of 0.044 ±0.090 mm (Table 2). Third, the error introduced by measuring the assumed center of a virtual contact area was assessed by examining the mean repeatability of the 2 operators (Table 2). Because virtual surfaces do not change over time, reproducibility was not evaluated. For the virtual measurements, the mean repeatability was 0.023 ±0.049, and thus, approximately 95% of the virtual measurements were within a sphere of 0.098 mm in diameter. Finally, the precision of the mechanical 3D Digitizer was 0.025 ±0.030 mm for the reference point and 0.028 ±0.048 mm for the predefined points on a gypsum cast (Table 3). Removing and repositioning the cast between successive measurements led to an error of 0.027 ±0.036 mm with negligible impact on the measures. DISCUSSION This clinical study evaluated the accuracy of transferring gypsum casts from a conventional articulator to a virtual environment using indirect digitalization. The primary requirement for an individualized reconstruction of occlusion with CAD-CAM technology is the accurate transfer of the patient’s analog data to the VDS. Both
Úry et al
Table 3. Precision of 3D digitizer (approximately 95% CI) Mean ±2 Standard Deviations (mm)
Maximum (mm)
Reference point measurement
0.025 ±0.030
0.063
Cast measurement w/o removal
0.028 ±0.048
0.136
Cast measurement w removal
0.027 ±0.036
0.090
Procedure
CI, confidence interval; w, with; w/o, without.
direct and indirect digital solutions use iterative best-fit alignment for superimposing the various surface models. Best-fit alignment is a nonspecific surface alignment method that globally minimizes the distance of each measured point to its reference point, often with an iterative least-square fitting algorithm. Best-fit algorithms allocate the same weight to each point during calculations. Redundant data points (representing noise) or surfaces differing markedly in resolution or margins can disturb the superimposition and considerably reduce the transferring accuracy. On the basis of the results of the present study, the null hypothesis, that the accuracy of the indirect digital workflow would be sufficiently precise to be applied for virtual occlusal analysis in clinical practice, was supported. The main advantage of the indirect method is that the axis-orbital planeerelated maxillary position and differently assigned mandibular positions (ICP, CR, TP) are easily transferable to the VDS. Moreover, the number of surface alignments is reduced to only 2,
THE JOURNAL OF PROSTHETIC DENTISTRY
8
Volume
namely the superimposition of both maxillary and mandibular surface models to the VIR. The fact that some analog steps are still required can be an advantage because it offers desirable control options. Moreover, the indirect technology is readily available in most dental offices and does not require costly chairside technology. The high precision of 0.023 ±0.049 mm achieved in measuring a virtual point in the CAD software was similar to the precision of the 3D Digitizer. In contrast, measuring an analog point on a gypsum cast was less precise (0.044 ±0.090 mm repeatability and 0.079 ±0.116 mm reproducibility). The lower precision of the analog samplings might originate from different sources, such as the digitizer, different skills of the operators, and gradual deterioration of the cast surface after repeated positioning of the stylus tip of the digitizer. The trueness of the indirect transfer (mean spatial distance between an analog point and its virtual counterpart) was 0.55 ±0.31 mm. In a comparable study, Solaberrieta et al22 tested the complete direct digital workflow and reported lower trueness (0.75 ±0.46 mm) for transferring analog points to a VDS, which may reflect greater error owing to multiple surface alignments in their protocol. A further disadvantage of the direct digital method is that transferring CR or TP positions to the virtual space still requires an analog occlusal record. The virtual facebow transfer is also more complicated and demands a costly face scanner. The authors are unaware of available data for estimating a threshold for a clinically acceptable mean deviation between analog and virtual points. From a clinical point of view, the high coincidence of the first contact points in the present study and the results of the virtual static and dynamic occlusal protocol both supported the diagnoses consistently. Accordingly, the protocol applied here yielded results applicable in clinical practice. CONCLUSIONS Based on the protocol and indirect scanning system used, the following conclusions were drawn: 1. A high correspondence was found between analog and virtual environments regarding the number and spatial position of occlusal contacts. The virtual instrumental functional analysis is reliable, contributing to a valid diagnosis. 2. The precision of both analog and virtual point measurements was high. The trueness (0.55 ±0.31 mm) of the indirect transferring method can be used as a current reference, but individual steps of the protocol should be further investigated and
THE JOURNAL OF PROSTHETIC DENTISTRY
-
Issue
-
improved. Based on the limited literature data, the indirect digital transferring method is better than the direct digital method regarding accuracy. REFERENCES 1. DeLong R, Ko CC, Anderson GC, Hodges JS, Douglas WH. Comparing maximum intercuspal contacts of virtual dental patients and mounted dental casts. J Prosthet Dent 2002;88:622-30. 2. Ahlholm P, Sipilä K, Vallittu P, Jakonen M, Kotiranta U. Digital versus conventional impressions in fixed prosthodontics: A review. J Prosthodont 2018;27:35-41. 3. Ender A, Attin T, Mehl A. In vivo precision of conventional and digital methods of obtaining complete-arch dental impressions. J Prosthet Dent 2016;115:313-20. 4. Ender A, Mehl A. Accuracy of complete-arch dental impressions: A new method of measuring trueness and precision. J Prosthet Dent 2013;109:121-8. 5. Flügge TV, Schlager S, Nelson K, Nahles S, Metzger MC. Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a cast scanner. Am J Orthod Dentofacial Orthop 2013;144: 471-8. 6. Goracci C, Franchi L, Vichi A, Ferrari M. Accuracy, reliability, and efficiency of intraoral scanners for full-arch impressions: A systematic review of the clinical evidence. Eur J Orthod 2016;38:422-8. 7. Güth JF, Edelhoff D, Schweiger J, Keul C. A new method for the evaluation of the accuracy of full-arch digital impressions in vitro. Clin Oral Investig 2016;20:1487-94. 8. Güth JF, Runkel C, Beuer F, Stimmelmayr M, Edelhoff D, Keul C. Accuracy of five intraoral scanners compared to indirect digitalization. Clin Oral Investig 2017;21:1445-55. 9. Güth J-F, Keul C, Stimmelmayr M, Beuer F, Edelhoff D. Accuracy of digital casts obtained by direct and indirect data capturing. Clin Oral Investig 2013;17:1201-8. 10. Patzelt SBM, Emmanouilidi A, Stampf S, Strub JR, Att W. Accuracy of full-arch scans using intraoral scanners. Clin Oral Investig 2014;18:1687-94. 11. Schepke U, Meijer HJA, Kerdijk W, Cune MS. Digital versus analog complete-arch impressions for single-unit premolar implant crowns: Operating time and patient preference. J Prosthet Dent 2015;114:403-6. 12. Luthardt RG, Loos R, Quaas S. Accuracy of intraoral data acquisition in comparison to the conventional impression. Int J Comput Dent 2005;8: 283-94. 13. Iwaki Y, Wakabayashi N, Igarashi Y. Dimensional accuracy of optical bite registration in single and multiple unit restorations. Oper Dent 2013;38: 309-15. 14. Solaberrieta E, Arias A, Brizuela A, Garikano X, Pradies G. Determining the requirements, section quantity, and dimension of the virtual occlusal record. J Prosthet Dent 2015;115:52-6. 15. Yee SHX, Esguerra RJ, Chew AAQA, Wong KM, Tan KBC. Three-dimensional static articulation accuracy of virtual models e Part I: System trueness and precision. J Prosthodont 2018;27:129-36. 16. DeLong R, Knorr S, Anderson GC, Hodges J, Pintado MR. Accuracy of contacts calculated from 3D images of occlusal surfaces. J Dent 2007;35:528-34. 17. Nilsson J, Richards RG, Thor A, Kamer L. Virtual bite registration using intraoral digital scanning, CT and CBCT: In vitro evaluation of a new method and its implication for orthognathic surgery. J Craniomaxillofac Surg 2016;44: 1194-200. 18. Nilsson J, Thor A, Kamer L. Development of workflow for recording virtual bite in the planning of orthognathic operations. Br J Oral Maxillofac Surg 2015;53:384-6. 19. Solaberrieta E, Mínguez R, Barrenetxea L, Etxaniz O. Direct transfer of the position of digitized casts to a virtual articulator. J Prosthet Dent 2013;109: 411-4. 20. Solaberrieta E, Otegi JR, Mínguez R, Etxaniz O. Improved digital transfer of the maxillary cast to a virtual articulator. J Prosthet Dent 2014;112:921-4. 21. Solaberrieta E, Garmendia A, Mínguez R, Brizuela A, Pradies G. Virtual facebow technique. J Prosthet Dent 2015;114:751-5. 22. Solaberrieta E, Mínguez R, Barrenetxea L, Otegi JR, Szentpétery A. Comparison of the accuracy of a 3-dimensional virtual method and the conventional method for transferring the maxillary cast to a virtual articulator. J Prosthet Dent 2015;113:191-7. 23. Idris B, Houston F, Claffey N. Comparison of the dimensional accuracy of one- and two-step techniques with the use of putty/wash addition silicone impression materials. J Prosthet Dent 1995;74:535-41. 24. Hung SH, Purk JH, Tira DE, Eick JD. Accuracy of one-step versus two-step putty wash addition silicone impression technique. J Prosthet Dent 1992;67: 583-9.
Úry et al
-
2019
25. Caputi S, Varvara G. Dimensional accuracy of resultant casts made by a monophase, one-step and two-step, and a novel two-step putty/light-body impression technique: An in vitro study. J Prosthet Dent 2008;99:274-81. 26. Nissan J, Laufer BZ, Brosh T, Assif D. Accuracy of three polyvinyl siloxane putty-wash impression techniques. J Prosthet Dent 2000;83:161-5. 27. Dahlström L, Haraldson T. Immediate electromyographic response in masseter and temporal muscles to bite plates and stabilization splints. Scand J Dent Res 1989;97:533-8. 28. Slavicek R. Clinical and instrumental functional analysis for diagnosis and treatment planning. Part 7. Computer-aided axiography. J Clin Orthod 1988;22:776-87. 29. Slavicek R. Clinical and instrumental functional analysis for diagnosis and treatment planning. Part 5. Axiography. J Clin Orthod 1988;22: 656-67. 30. Wilkie ND. The anterior point of reference. J Prosthet Dent 1979;41: 488-96. 31. Besl P, McKay N. A method for registration of 3-D Shapes. IEEE Trans Pattern Anal Mach Intell 1992;14:239-56. 32. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013. Available at: http://www.R-project.org/. Accessed August 15, 2018.
Úry et al
9
33. Egermark-Eriksson I, Carlsson GE, Magnusson T. A long-term epidemiologic study of the relationship between occlusal factors and mandibular dysfunction in children and adolescents. J Dent Res 1987;66:67-71. Corresponding author: Dr El} od Úry ConfiDent Dental Office Füredi sétány 9/1/3 9400 Sopron HUNGARY Email: [email protected]; confident@confident.hu Acknowledgments The authors thank Zsuzsanna Deák Bárdos and András Petyus for data acquisition. Rudolf Slavicek, Fred Bookstein, and Philipp Mitteröcker contributed with valuable discussions. Copyright © 2019 by the Editorial Council for The Journal of Prosthetic Dentistry. https://doi.org/10.1016/j.prosdent.2018.12.019
THE JOURNAL OF PROSTHETIC DENTISTRY
Accuracy of transferring analog dental casts to a virtual articulator }d Úry, DMD, MSc,a Cinzia Fornai, PhD,b and Gerhard W. Weber, PhDc Elo The conventional analog ABSTRACT method of transferring patient Statement of problem. Complex digital workflows have been developed to create virtual dental information to semiadjustable patients. Direct and indirect digital methods are available for transferring analog patient or fully adjustable articulators information to virtual articulators. The direct method consists solely of digital workflows. The involves several clinical and indirect method combines analog steps and digital procedures, representing an intermediate laboratory steps (Fig. 1). These solution between the analog and direct digital approach. Studies that have investigated the overall accuracy of the virtual working space are sparse. steps are followed by a detailed static and dynamic Purpose. The purpose of this clinical study was to investigate the accuracy of the virtual dental occlusal analysis of the gypspace using the indirect digital workflow. sum casts in the programmed Material and methods. Mounted gypsum casts of 18 patients were used for indirect scanning. The articulators, forming part of maxillary casts were mounted in their skull-related position with a kinematic facebow. The the definitive diagnosis. mandibular casts were mounted in centric relation to the maxillary casts. The obtained digitized The widespread use of casts were transferred to a virtual articulator. An occlusal analysis was performed both in the analog and virtual environments, and the coordinates of matching analog and virtual contact digital technology and points were measured. The trueness and precision of the indirect transferring procedure were computer-assisted techniques assessed. in dentistry has led to considerable changes, including the Results. A total of 194 analog points was considered in the reference. Ninety-three percent of all analog points matched a virtual correspondent, and 96% of the analog first contacts between creation of the virtual dental the casts were also present as first contacts in the virtual space. The trueness of the data patient.1 Instead of the contransfer, corresponding to the spatial distance between the matching analog and virtual points, ventional procedure, direct or was 0.55 ±0.31 mm. The maximum recorded deviation was 1.02 mm. indirect digital workflows can Conclusions. The correspondence between the number and position of analog and virtual contacts be used to transfer a patient’s was high. The mean absolute deviation of the matching point-pairs was better than that reported data into a virtual dental space for the direct digital method. Under the conditions described, the virtual dental space created (VDS), where surface models with the indirect digital method can be reliably used for virtual occlusal analysis in clinical from direct digital scans of the practice. (J Prosthet Dent 2019;-:---) dental arches or indirect digital only digital data. The indirect digital workflow represents scans of the gypsum casts are placed in a virtual articuan intermediate option between the analog and direct lator. The protocol of the direct digital workflow is digital methods (Fig. 1). It includes all the conventional comparable with that of the analog procedures but uses
C.F. was financially supported by the Swiss National Science Foundation (grants no. 31003A_156299 and 31003A_176319), the Siegfried Ludwig-Rudolf Slavicek Foundation, Vienna, Austria (FA547016), A.E.R.S. Dental Medicine Organizations GmbH, Vienna, Austria (FA547013). a Postgraduate doctoral student, Department of Evolutionary Anthropology, University of Vienna, Vienna, Austria; and Private practice, ConfiDent Dental Office, Sopron, Hungary. b Postdoctoral Fellow, Institute of Evolutionary Medicine, University of Zurich, Zurich, Switzerland; and Lecturer, Department of Evolutionary Anthropology, University of Vienna, Vienna, Austria. c Associate Professor, Department of Evolutionary Anthropology, University of Vienna, Vienna, Austria; and Head of Core Facility for Micro-Computed Tomography, University of Vienna, Vienna, Austria.
THE JOURNAL OF PROSTHETIC DENTISTRY
1
2
Volume
Clinical Implications The indirect digital workflow for transferring analog data to the virtual dental space is accurate and can be used confidently for occlusal analysis.
steps before the gypsum casts are mounted in centric relation (CR) and maximal intercuspal position (ICP). Maxillary and mandibular casts are then digitized with a laboratory scanner, and their skull-related position and spatial correlation are recorded. The resulting 3D surfaces are registered in the virtual coordinate system according to their CR and ICP. The accuracy of the various phases and procedures of the digital workflow have been evaluated. The accuracy of the complete-arch digital scans has been compared with that of conventional impressions.2-12 Ender et al3 reported that highly accurate conventional impression materials provided significantly higher precision than current intraoral scanners. Güth et al8 found that the accuracy of the indirect digitalization was in the mid-range of the various direct digitalization systems tested. Ahlholm et al2 recommended the conventional impression techniques for large fixed dental prostheses, and in a literature review, Goracci et al6 concluded that up-to-date clinical evidence collected on complete-arch intraoral scanning is lacking. Virtual occlusal records in computer-aided design and computer-aided manufacturing (CAD-CAM) systems align the virtual mandibular cast to the maxillary cast. Iwaki et al13 described the dimensional stability of optical and conventional occlusal records. They reported that, for multiple restorations in a quadrant, interarch distances increased significantly from reference values with an optical interocclusal record. Solaberrieta et al14 reported that the combination of left and right lateral virtual interocclusal records (VIRs) is the most convenient for virtual articulation. The virtual articulation accuracy of indirect scanner systems was also evaluated by Yee et al.15 They compared interarch, interocclusal linear, and global virtual distances with reference values obtained from conventionally articulated gypsum casts and concluded that the different levels of distortion from reference values were due to differences in scanner accuracy and virtual articulation algorithms. Intraoral scanners provide an accurate quantitative reproduction of analog contacts in a virtual environment for casts mounted in ICP.1,16 In contrast, VIRs in CR or different therapeutic positions (TPs) are difficult to capture. From the few in vitro protocols published,16,17 it can be concluded that virtual articulation is more accurate if a VIR is registered with the conventional CR record placed intraorally rather than scanning the entire CR record extraorally.17,18
THE JOURNAL OF PROSTHETIC DENTISTRY
-
Issue
-
The virtual facebow transfer locates the maxillary surface model in the virtual articulator in its skull-related position, creating a standard coordinate system between patient data and the VDS. In this coordinate system, electronically recorded mandibular movements can be displayed for a dynamic occlusal analysis. Virtual facebow transfer protocols have been described.19-21 Based on a single patient and 6 matching points, Solaberietta et al22 reported a mean deviation of 0.752 ±0.456 mm between the virtual occlusal points and their digitized analog references using direct digital workflows. The purpose of the present study was to evaluate the global accuracy of the VDS by using the indirect digital workflow. For this purpose, spatial coordinates of analog and virtual contact points obtained from a static occlusal protocol performed in both environments were compared. The null hypothesis was that the accuracy of the indirect digital workflow would be sufficiently precise to be applied for virtual occlusal analysis in clinical practice. MATERIAL AND METHODS Mounted maxillary and mandibular gypsum casts of 18 nonedentulous patients with temporomandibular disorders attending the ConfiDent Dental Office (Sopron, Hungary) were used for indirect digitalization. All included patients received a detailed diagnosis before extensive treatment. Written consent was obtained from the patients. Factors such as sex, age, and craniomandibular relation were not considered relevant to the outcome of this study. Patient data were collected over 2 appointments. First, conventional polyvinyl siloxane impressions were made of both dental arches with a 2-step putty-wash technique (Aquasil Ultra DECA Heavy, Aquasil Ultra LV; Dentsply Sirona).23-26 Two pairs of casts were prepared per patient. The definitive casts poured in Type IV stone (GC Fujirock EP Classic; GC Corp) were equipped with a Zeiser base plate system (Zeiser Dentalgeräte GmbH). Each tooth of the definitive mandibular cast, except the central and lateral incisors, was double pinned, but not sectioned. The second pair of casts was used to prepare the acrylic resin base plate for the centric interocclusal record. In a second appointment, after a 10-minute deprogramming of the masticatory muscles,27 a CR record was made (Primobyte base plate; Primobyte detail paste; Primotec, Joachim Mosch e.K.). Then, electronic axiography was performed (CADIAX 4; GAMMA Medizinisch-wissenschaftliche Fortbildungs-GmbH), and the kinematic axis of the mandible was identified.28,29 The kinematic axis and the left infraorbital point defined an individual axis-orbital plane.30 This plane served as a reference for constructing a unitary coordinate system of the patient’s axiography data, lateral
Úry et al
-
2019
3
ANALOG METHOD
INDIRECT DIGITAL METHOD
DIRECT DIGITAL METHOD
Direct digital scan
Conventional impression
Gypsum casts
Virtual interocclusal record
ICP
Interocclusal record
Virtual surface models
CR, TP
ICP CR? TP?
Virtual articulation
Electronic condylography Kinematic facebow
Maxillary cast mounting
Virtual facebow
Face scan
ICP
Mandibular cast mounting
CR, TP Indirect digital scan digitized casts Virtual articulation
Transferring files to a virtual articulator
General medical and dental anamnesis
Clinical functional analysis
Chronic pain anamnesis
Standardized radiological analysis
Instrumental functional analysis Static and dynamic occlusal protocol
Virtual static and dynamic occlusal protocol
Diagnosis
Figure 1. Flowchart illustrating analog and digital approaches for transferring patient information to articulator. CR, centric relation; ICP, intercuspal position; TP, therapeutic position.
radiographs, cone beam computed tomography (CBCT) data, and mounted casts. Definitive casts of each patient were mounted in a fully adjustable articulator (Reference SL; GAMMA Medizinisch-wissenschaftliche Fortbildungs-GmbH) by means of the kinematic facebow and the CR record. The mounting gypsum (Artifix; Amann Girrbach AG) was allowed to harden for 24 hours. Immediately after, the casts were digitized with a high-resolution laboratory Úry et al
scanner (Activity 885; smart optics Sensortechnik GmbH) with an accuracy of 10 mm according to the manufacturer. The scanner was calibrated as indicated by the manufacturer. To register the VIR, an object holder was used to transfer the articulator-related position of the casts to the scanner (FINOSCAN RELATION Scan Fixator Gamma; Fino GmbH). The Scan Fixator had been previously synchronized with the articulator (Fig. 2). In the CAD software (dental Scan; smart optics Sensortechnik THE JOURNAL OF PROSTHETIC DENTISTRY
4
Volume
-
Issue
-
Figure 3. Virtual maxillary and mandibular surface models aligned to virtual interocclusal record.
Figure 2. Synchronization process between articulator and scan fixator. A, Calibration key in articulator. B, Calibration key in scan fixator, resetting systems to zero. C, Scan fixator holding articulator-related position of casts in scanner.
GmbH), the maxillary and mandibular casts were virtually aligned with the VIR using a best-fit algorithm (Fig. 3).31 On the same day, the mandibular cast was sectioned, and a step-by-step static occlusal analysis was performed. The first contact point between the casts was identified by rotating them against each other and was marked with 8-mm articulating paper (Arti-Fol; Dr Jean Bausch GmbH & Co KG). The contacting mandibular tooth was numbered and removed. Rotation continued THE JOURNAL OF PROSTHETIC DENTISTRY
to the next contact following the same procedure. The procedure ended when the whole sequence of contacts was registered, namely until 1 or more maxillary anterior teeth contacted the mandibular anterior teeth. The contact sequences were ordered numerically, and the contacting maxillary teeth were listed. Multiple contact points on the same tooth were differentiated according to their position on the occlusal surface. The spatial coordinates of the maxillary contact points were recorded with a calibrated mechanical 3D Digitizer (GAMMA Medizinisch-wissenschaftliche Fortbildungs-GmbH) and reported in a spreadsheet (Excel; Microsoft Corp). The sampling was repeated in 3 different sessions by 2 dentists (E.U., ZS.D.B.). The measurements were carried out with magnifying loupes and a headlight. Samplings by the same examiner on the same cast were at least 1 week apart. In each session, the maxillary contacts of a cast were measured 3 times consecutively. Thus, every contact was sampled 9 times by each operator over the 3 sampling sessions. Contacts were of different shapes and sizes and could appear in isolation or as close clusters. For standardization, each operator measured the estimated center of each contact area, regardless of its shape. When the centers of contact areas were less than 1 mm apart, the central point of the cluster rather than the central point of each area was collected. The reoriented surface models were loaded in the virtual articulator (CADIAS 3D; GAMMA Medizinischwissenschaftliche Fortbildungs-GmbH), where a virtual segmentation was performed to separate each tooth from its neighbors. As for the analog casts, a virtual static occlusal protocol was carried out with the segmented surface models in CR (Fig. 1). The segmented mandibular surface model was automatically rotated against the maxillary surface model with 0.05-degree increments. If contacts between the surfaces were detected, they were marked on the maxillary teeth as intersecting areas. The contacting antagonists were automatically removed and Úry et al
-
2019
5
Figure 4. Entire sequence of contacts. A, Different colors indicate collisions at different rotational steps. B, Contact points in order of collision. Rotational angles, incisal pin changes between consecutive collisions, as well as colliding maxillary and mandibular teeth were enumerated.
Figure 5. A, Maxillary contact points after analog static occlusal protocol. B, Automated virtual static occlusal protocol shows almost perfect point-match. Maxillary right first molar had 2 mesial contacts on analog cast, but only 1 detected on virtual surface model (blue arrows).
excluded from further calculations. The rotation continued until the last mandibular contact was detected. Contacting teeth were automatically listed in their order of contact (Fig. 4). The center of each maxillary virtual contact area was sampled twice by 2 operators (E.U. and A.P., the latter an information technology professional). To assess the precision of the 3D Digitizer, the repeatability of measuring a reference point located at the base of the digitizer was first evaluated. Between consecutive measurements (n=50), the measuring stylus was moved from its position along all the 3 spatial axes. In addition, predefined points were measured on a gypsum cast with and without removing the cast from the base of the digitizer between consecutive samplings. For both analog and virtual points, the means and standard deviations were calculated by operator and by operator and sampling session for the total of the samplings at a confidence interval of approximately 95%. Two standard deviations were reported because of their greater practical significance. Precision was divided into Úry et al
Table 1. Trueness of indirect transfer to virtual space (approximately 95% CI) Metric Analogdvirtual difference
Mean ±2 Standard Deviations (mm)
Maximum (mm)
0.55 ±0.31
1.02
CI, confidence interval.
repeatability (same operator, within a short period) and reproducibility (different operators or same operator over a long period). Repeatability is the average spatial distance of all analog points from the global centroid calculated per day and per operator. Reproducibility is the average distance of all analog points from the centroid of all days and operators together. In addition, the trueness of the transferring method was assessed, namely the spatial deviation between the analog and virtual global centroids. Bland-Altman plots were generated to visualize the interoperator error for the analog measurements as well as the analog and virtual mean deviations from the analog centroid. Excel THE JOURNAL OF PROSTHETIC DENTISTRY
6
Volume
-
Issue
-
Mean Analog Deviations Minus Mean Virtual Deviations
0.2
0.0
+1.96 SD –0.16
–0.2 Mean –0.48
–0.4
–0.6 –1.96 SD –0.80 –0.8
–1.0
0.2
0.3
0.4
0.5
Mean Analog and Virtual Deviations From Analog Centroid Figure 6. Bland-Altman plot of deviations from global analog centroid for 181 matching analog and virtual measurements. Scatter along horizontal axis reflects that some points more difficult to identify than others. All points on vertical axis had negative values, indicating that deviations of virtual points from analog centroids always larger than those of analog points to their centroids. Variation of deviation depended on magnitude of deviation itself. SD, standard deviation.
(Microsoft Corp) and R packages32 were used for statistical analysis.
Table 2. Precision of analog and virtual point measurements (approximately 95% CI) Mean ±2 Standard Deviations (mm)
RESULTS A total of 194 analog and 212 virtual points were identified. Analog points were considered as the reference. Ninety-three percent of the analog points (n=181) matched a virtual correspondent (Fig. 5). The first contact point in CR is of particular diagnostic importance as it may be a deflective occlusal interference associated with temporomandibular disorders.33 Ninety-six percent of the analog first contacts (n=47) were also found virtually. Only 2 data sets showed differences between analog and virtual first contacts. In both situations, multiple analog first contacts could only be partially reproduced by the virtual method in the first rotational step (1 of 2, and 4 of 5, respectively). The mean spatial distance between analogs and their corresponding virtual points was 0.55 ±0.31 mm, which corresponds to the trueness of the transferring method. The maximum deviation recorded was 1.02 mm (Table 1). The Bland-Altman plot shows the range of deviations of the analog and virtual mean points from the analog centroid. Values along the vertical axis were always negative, meaning that the deviation of the virtual points from the analog centroid was always higher than the deviation of the corresponding analog points to their centroid. The regular scatter of the points, aligned along a descending diagonal, showed a strong correlation
THE JOURNAL OF PROSTHETIC DENTISTRY
Metric
Operator 1
Operator 2
Average Operator 1+2
Analog reproducibility (all days)
0.073 ±0.108
0.085 ±0.123
0.079 ±0.116
Analog repeatability (all days)
0.040 ±0.085
0.048 ±0.095
Virtual repeatability (all days)
0.044 ±0.090 0.023 ±0.049
CI, confidence interval.
between the deviations of the virtual points from the analog centroid and the deviations of the analog points to their centroid (Fig. 6). The spatial difference between analog and correspondent virtual points could result from both analog and virtual sampling approaches. First, the global interoperator reproducibility for analog measurements was examined. The mean was calculated from all measurements of the 2 operators over the 3 sessions (Table 2). Overall reproducibility for both operators was 0.079 ±0.116 mm. Thus, approximately 95% of the measurements were within a sphere of 0.232 mm in diameter. The Bland-Altman plot shown in Figure 7 visualizes reproducibility, namely the deviation of the analog point measurements of the operators from the global analog centroid. Interoperator error is negligible with a mean difference of 0.013 mm. The cone-shaped scatter of the points from left to right indicates that interoperator error along the vertical axis
Úry et al
2019
7
Mean Operator 1 Minus Mean Operator 2
-
0.10
0.05
+1.96 SD 0.040
0.00
Mean –0.013
–0.05
–1.96 SD –0.066
–0.10
0.04
0.06
0.08
0.10
0.12
0.14
Grand Mean Operator 1 and Operator 2 Figure 7. Bland-Altman plot of deviations for 194 analog measurements of both operators from global analog centroid. Minimal shift of operators’ mean from zero (−0.013 mm) implied that interoperator error negligible. Only few measurements outside 2 standard deviationeconfidence interval (blue dashed lines). Rather symmetric cone-shaped scatter of points from left to right indicated lack of systematic errors. Differences between operators (vertical axis) increased with increasing deviations from centroid, meaning that interoperator error increased for points for which intraoperator error higher. SD, standard deviation.
increased for points which were more difficult to identify and thus also showed a higher intraoperator error. Second, the global intraoperator repeatability for analog measurements was calculated, which was, as expected, more precise than reproducibility with a mean of 0.044 ±0.090 mm (Table 2). Third, the error introduced by measuring the assumed center of a virtual contact area was assessed by examining the mean repeatability of the 2 operators (Table 2). Because virtual surfaces do not change over time, reproducibility was not evaluated. For the virtual measurements, the mean repeatability was 0.023 ±0.049, and thus, approximately 95% of the virtual measurements were within a sphere of 0.098 mm in diameter. Finally, the precision of the mechanical 3D Digitizer was 0.025 ±0.030 mm for the reference point and 0.028 ±0.048 mm for the predefined points on a gypsum cast (Table 3). Removing and repositioning the cast between successive measurements led to an error of 0.027 ±0.036 mm with negligible impact on the measures. DISCUSSION This clinical study evaluated the accuracy of transferring gypsum casts from a conventional articulator to a virtual environment using indirect digitalization. The primary requirement for an individualized reconstruction of occlusion with CAD-CAM technology is the accurate transfer of the patient’s analog data to the VDS. Both
Úry et al
Table 3. Precision of 3D digitizer (approximately 95% CI) Mean ±2 Standard Deviations (mm)
Maximum (mm)
Reference point measurement
0.025 ±0.030
0.063
Cast measurement w/o removal
0.028 ±0.048
0.136
Cast measurement w removal
0.027 ±0.036
0.090
Procedure
CI, confidence interval; w, with; w/o, without.
direct and indirect digital solutions use iterative best-fit alignment for superimposing the various surface models. Best-fit alignment is a nonspecific surface alignment method that globally minimizes the distance of each measured point to its reference point, often with an iterative least-square fitting algorithm. Best-fit algorithms allocate the same weight to each point during calculations. Redundant data points (representing noise) or surfaces differing markedly in resolution or margins can disturb the superimposition and considerably reduce the transferring accuracy. On the basis of the results of the present study, the null hypothesis, that the accuracy of the indirect digital workflow would be sufficiently precise to be applied for virtual occlusal analysis in clinical practice, was supported. The main advantage of the indirect method is that the axis-orbital planeerelated maxillary position and differently assigned mandibular positions (ICP, CR, TP) are easily transferable to the VDS. Moreover, the number of surface alignments is reduced to only 2,
THE JOURNAL OF PROSTHETIC DENTISTRY
8
Volume
namely the superimposition of both maxillary and mandibular surface models to the VIR. The fact that some analog steps are still required can be an advantage because it offers desirable control options. Moreover, the indirect technology is readily available in most dental offices and does not require costly chairside technology. The high precision of 0.023 ±0.049 mm achieved in measuring a virtual point in the CAD software was similar to the precision of the 3D Digitizer. In contrast, measuring an analog point on a gypsum cast was less precise (0.044 ±0.090 mm repeatability and 0.079 ±0.116 mm reproducibility). The lower precision of the analog samplings might originate from different sources, such as the digitizer, different skills of the operators, and gradual deterioration of the cast surface after repeated positioning of the stylus tip of the digitizer. The trueness of the indirect transfer (mean spatial distance between an analog point and its virtual counterpart) was 0.55 ±0.31 mm. In a comparable study, Solaberrieta et al22 tested the complete direct digital workflow and reported lower trueness (0.75 ±0.46 mm) for transferring analog points to a VDS, which may reflect greater error owing to multiple surface alignments in their protocol. A further disadvantage of the direct digital method is that transferring CR or TP positions to the virtual space still requires an analog occlusal record. The virtual facebow transfer is also more complicated and demands a costly face scanner. The authors are unaware of available data for estimating a threshold for a clinically acceptable mean deviation between analog and virtual points. From a clinical point of view, the high coincidence of the first contact points in the present study and the results of the virtual static and dynamic occlusal protocol both supported the diagnoses consistently. Accordingly, the protocol applied here yielded results applicable in clinical practice. CONCLUSIONS Based on the protocol and indirect scanning system used, the following conclusions were drawn: 1. A high correspondence was found between analog and virtual environments regarding the number and spatial position of occlusal contacts. The virtual instrumental functional analysis is reliable, contributing to a valid diagnosis. 2. The precision of both analog and virtual point measurements was high. The trueness (0.55 ±0.31 mm) of the indirect transferring method can be used as a current reference, but individual steps of the protocol should be further investigated and
THE JOURNAL OF PROSTHETIC DENTISTRY
-
Issue
-
improved. Based on the limited literature data, the indirect digital transferring method is better than the direct digital method regarding accuracy. REFERENCES 1. DeLong R, Ko CC, Anderson GC, Hodges JS, Douglas WH. Comparing maximum intercuspal contacts of virtual dental patients and mounted dental casts. J Prosthet Dent 2002;88:622-30. 2. Ahlholm P, Sipilä K, Vallittu P, Jakonen M, Kotiranta U. Digital versus conventional impressions in fixed prosthodontics: A review. J Prosthodont 2018;27:35-41. 3. Ender A, Attin T, Mehl A. In vivo precision of conventional and digital methods of obtaining complete-arch dental impressions. J Prosthet Dent 2016;115:313-20. 4. Ender A, Mehl A. Accuracy of complete-arch dental impressions: A new method of measuring trueness and precision. J Prosthet Dent 2013;109:121-8. 5. Flügge TV, Schlager S, Nelson K, Nahles S, Metzger MC. Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a cast scanner. Am J Orthod Dentofacial Orthop 2013;144: 471-8. 6. Goracci C, Franchi L, Vichi A, Ferrari M. Accuracy, reliability, and efficiency of intraoral scanners for full-arch impressions: A systematic review of the clinical evidence. Eur J Orthod 2016;38:422-8. 7. Güth JF, Edelhoff D, Schweiger J, Keul C. A new method for the evaluation of the accuracy of full-arch digital impressions in vitro. Clin Oral Investig 2016;20:1487-94. 8. Güth JF, Runkel C, Beuer F, Stimmelmayr M, Edelhoff D, Keul C. Accuracy of five intraoral scanners compared to indirect digitalization. Clin Oral Investig 2017;21:1445-55. 9. Güth J-F, Keul C, Stimmelmayr M, Beuer F, Edelhoff D. Accuracy of digital casts obtained by direct and indirect data capturing. Clin Oral Investig 2013;17:1201-8. 10. Patzelt SBM, Emmanouilidi A, Stampf S, Strub JR, Att W. Accuracy of full-arch scans using intraoral scanners. Clin Oral Investig 2014;18:1687-94. 11. Schepke U, Meijer HJA, Kerdijk W, Cune MS. Digital versus analog complete-arch impressions for single-unit premolar implant crowns: Operating time and patient preference. J Prosthet Dent 2015;114:403-6. 12. Luthardt RG, Loos R, Quaas S. Accuracy of intraoral data acquisition in comparison to the conventional impression. Int J Comput Dent 2005;8: 283-94. 13. Iwaki Y, Wakabayashi N, Igarashi Y. Dimensional accuracy of optical bite registration in single and multiple unit restorations. Oper Dent 2013;38: 309-15. 14. Solaberrieta E, Arias A, Brizuela A, Garikano X, Pradies G. Determining the requirements, section quantity, and dimension of the virtual occlusal record. J Prosthet Dent 2015;115:52-6. 15. Yee SHX, Esguerra RJ, Chew AAQA, Wong KM, Tan KBC. Three-dimensional static articulation accuracy of virtual models e Part I: System trueness and precision. J Prosthodont 2018;27:129-36. 16. DeLong R, Knorr S, Anderson GC, Hodges J, Pintado MR. Accuracy of contacts calculated from 3D images of occlusal surfaces. J Dent 2007;35:528-34. 17. Nilsson J, Richards RG, Thor A, Kamer L. Virtual bite registration using intraoral digital scanning, CT and CBCT: In vitro evaluation of a new method and its implication for orthognathic surgery. J Craniomaxillofac Surg 2016;44: 1194-200. 18. Nilsson J, Thor A, Kamer L. Development of workflow for recording virtual bite in the planning of orthognathic operations. Br J Oral Maxillofac Surg 2015;53:384-6. 19. Solaberrieta E, Mínguez R, Barrenetxea L, Etxaniz O. Direct transfer of the position of digitized casts to a virtual articulator. J Prosthet Dent 2013;109: 411-4. 20. Solaberrieta E, Otegi JR, Mínguez R, Etxaniz O. Improved digital transfer of the maxillary cast to a virtual articulator. J Prosthet Dent 2014;112:921-4. 21. Solaberrieta E, Garmendia A, Mínguez R, Brizuela A, Pradies G. Virtual facebow technique. J Prosthet Dent 2015;114:751-5. 22. Solaberrieta E, Mínguez R, Barrenetxea L, Otegi JR, Szentpétery A. Comparison of the accuracy of a 3-dimensional virtual method and the conventional method for transferring the maxillary cast to a virtual articulator. J Prosthet Dent 2015;113:191-7. 23. Idris B, Houston F, Claffey N. Comparison of the dimensional accuracy of one- and two-step techniques with the use of putty/wash addition silicone impression materials. J Prosthet Dent 1995;74:535-41. 24. Hung SH, Purk JH, Tira DE, Eick JD. Accuracy of one-step versus two-step putty wash addition silicone impression technique. J Prosthet Dent 1992;67: 583-9.
Úry et al
-
2019
25. Caputi S, Varvara G. Dimensional accuracy of resultant casts made by a monophase, one-step and two-step, and a novel two-step putty/light-body impression technique: An in vitro study. J Prosthet Dent 2008;99:274-81. 26. Nissan J, Laufer BZ, Brosh T, Assif D. Accuracy of three polyvinyl siloxane putty-wash impression techniques. J Prosthet Dent 2000;83:161-5. 27. Dahlström L, Haraldson T. Immediate electromyographic response in masseter and temporal muscles to bite plates and stabilization splints. Scand J Dent Res 1989;97:533-8. 28. Slavicek R. Clinical and instrumental functional analysis for diagnosis and treatment planning. Part 7. Computer-aided axiography. J Clin Orthod 1988;22:776-87. 29. Slavicek R. Clinical and instrumental functional analysis for diagnosis and treatment planning. Part 5. Axiography. J Clin Orthod 1988;22: 656-67. 30. Wilkie ND. The anterior point of reference. J Prosthet Dent 1979;41: 488-96. 31. Besl P, McKay N. A method for registration of 3-D Shapes. IEEE Trans Pattern Anal Mach Intell 1992;14:239-56. 32. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013. Available at: http://www.R-project.org/. Accessed August 15, 2018.
Úry et al
9
33. Egermark-Eriksson I, Carlsson GE, Magnusson T. A long-term epidemiologic study of the relationship between occlusal factors and mandibular dysfunction in children and adolescents. J Dent Res 1987;66:67-71. Corresponding author: Dr El} od Úry ConfiDent Dental Office Füredi sétány 9/1/3 9400 Sopron HUNGARY Email: [email protected]; confident@confident.hu Acknowledgments The authors thank Zsuzsanna Deák Bárdos and András Petyus for data acquisition. Rudolf Slavicek, Fred Bookstein, and Philipp Mitteröcker contributed with valuable discussions. Copyright © 2019 by the Editorial Council for The Journal of Prosthetic Dentistry. https://doi.org/10.1016/j.prosdent.2018.12.019
THE JOURNAL OF PROSTHETIC DENTISTRY