HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dalstra, M. Articles by Melsen, B. Search for Related Content PubMed PubMed Citation Articles by Dalstra, M. Articles by Melsen, B.

From alginate impressions to digital virtual models: accuracy and reproducibility

School of Dentistry, University of Aarhus, Denmark

Address for correspondence: Michel Dalstra, Vennelyst Boulevard, 9, School of Dentistry, Aarhus University, DK-8000 Aarhus C, Denmark, Email: mdalstra{at}odont.au.dk

Received 29 October 2007; accepted 11 November 2008

	Abstract

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

 
Objective: To compare the accuracy and reproducibility of measurements performed on digital virtual models with those taken on plaster casts from models poured immediately after the impression was taken, the ‘gold standard’, and from plaster models poured following a 3–5 day shipping procedure of the alginate impression.

Design: Direct comparison of two measuring techniques.

Setting: The study was conducted at the Department of Orthodontics, School of Dentistry, University of Aarhus, Denmark in 2006/2007.

Participants: Twelve randomly selected orthodontic graduate students with informed consent.

Methods: Three sets of alginate impressions were taken from the participants within 1 hour. Plaster models were poured immediately from two of the sets, while the third set was kept in transit in the mail for 3–5 days. Upon return a plaster model was poured as well. Finally digital models were made from the plaster models. A number of measurements were performed on the plaster casts with a digital calliper and on the corresponding digital models using the virtual measuring tool of the accompanying software. Afterwards these measurements were compared statistically.

Results: No statistical differences were found between the three sets of plaster models. The intraand inter-observer variability are smaller for the measurements performed on the digital models.

Conclusions: Sending alginate impressions by mail does not affect the quality and accuracy of plaster casts poured from them afterwards. Virtual measurements performed on digital models display less variability than the corresponding measurements performed with a calliper on the actual models.

Key words: Orthodontics, plaster models, digital models, alginate impressions, virtual models

	Introduction

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

 
Traditional plaster casts are among the last clinical record in the orthodontic office to be converted into digital media, but virtual dental models are gradually becoming more prevalent.^1–4 This change, although meeting opposition from conservative orthodontists who want to ‘feel’ the plaster models in their hands, has considerable advantages, particularly in obviating the need for extensive storage facilities, reducing the risk of physical damage and/or the disappearance of the casts stored in the wrong location. In addition, there is the possibility of sharing the models with colleagues, other specialists involved in the treatment, and even with the patient.

The questions asked by the ‘less computer-enthusiastic’ orthodontists are; can a virtual dental model actually replace the plaster cast as a basis for treatment planning? And whether measurements carried out on virtual models can replace those performed on the study casts?

OrthoCAD^TM (Cadent, Carlstadt, NJ, USA) introduced virtual models in 1999, followed by E-models^TM (Geodigm Corp., Chanhassen, MN, USA) in 2001. Both these products have been evaluated and found to be useful in the treatment planning process. Measurements carried out in relation to the Bolton analysis were not significantly different from those carried out on the ‘gold standard’ whether this was the original plaster model from which the virtual model was developed⁵ or a dentoform model.⁶ Although linear measurements with a digital calliper on a physical model have been reported to be more accurate than their virtual counterparts,⁷ the accuracy of the digital measurements was considered to be clinically acceptable. Another consideration will be the ease of measuring using the different dedicated software programs.

Virtual models can be produced by several methods. The most direct method is by using an intra-oral laser-scanner (Orametrix Inc., Richardson, TX, USA). This method makes the impression superfluous, but the clinical chair time may be increased. Digital virtual models can also be produced by a negative surface model technique generated by laser-scanning the inside of an impression; however this method might encounter difficulties in relation to undercuts and the limited space inside the impression. The most frequently applied method seems to be to pour a plaster model as a halfway step. This plaster model is either non-destructively digitized using stereophotogrammetry,^8–10 a surface laser-scanner^11,12 or industrial computer tomography or by using a destructive sequential slicing technique.

For most current brands of digital virtual models, the technology to create the models is outsourced from the orthodontic practice by sending alginate impressions or plaster models to a company specializing in creating digital models (OrthoCAD^TM, Cadent, Carlstadt, NJ, USA; E-models^TM, Geodigm Corp., Chanhassen, MN, USA; DigiModel^TM, OrthoProof, Nieuwegein, The Netherlands; O3DM^TM, OrtoLab, Czstochowa, Poland). After a number of days, the models can be downloaded from the company’s web-site. This approach has the advantage that individual practices do not have to invest in the technology and know-how to produce virtual models; however a potential error may be introduced by the fact that the alginate impressions are sent by mail. OrthoCAD^TM initially required silicon impressions to counter this problem, but is now accepting alginate ones. This is not a trivial issue, because alginate as a dental impression material has been reported to be subject to volume changes during storage.¹³ The influence of shipping has so far not been investigated and the ‘gold standard’ in the above mentioned trial has always been the model from which the virtual model was generated.

The overall aim of the study was to examine the stability of alginate impressions over a period of time. The specific objectives were:

• to detect any significant differences in the measurements taken from models poured immediately and models poured after 3–5 days;
  
• to assess the variability in the measurements performed on plaster and digital models;
  
• to measure inter-observer variability;
  
• to determine the reproducibility of performing measurements directly on plaster models compared with those obtained from digital models.
  

	Materials and methods

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

 
Three sets of alginate (Aroma Fine^TM DF III, GC Corp., Tokyo, Japan) impressions were taken from twelve randomly selected orthodontic graduate students using plastic impression (ASA Dental^TM, Bozzano, Italy). The students provided verbal consent for their discarded practice impressions to be used in this study, as in Denmark, ethics approval is not required for this type of study. The three sets of impressions from each student were taken within one hour. Plaster models were poured immediately from two sets of impressions, while the third set was wrapped in a moist gauze, put in a sealed bag and mailed from Aarhus, Denmark to an address in Copenhagen, Denmark and back, thus being 3 to 5 days in transit in total. Upon return to the School of Dentistry in Aarhus, the plaster models were poured.

The following measurements were obtained from the three sets of plaster models using a digital calliper (Digimatic Calliper: 700–113 MyCal Lite, Mitutoyo America Corp., Plymouth, MI, USA) with an accuracy of 0.01 mm:

• mesio-distal dimension of teeth 11 and 16 (maximal crown width);
  
• maxillary inter-canine width measured from cusp to cusp;
  
• maxillary arch width measured as the maximal distance between the buccal surfaces of the first molars;
  
• maxillary arch length measured as the distance from the contact point of the central incisors to a frontal plane through the most posterior aspect of the first molars;
  
• overbite measured as the largest overlap perpendicular to the occlusal plane;
  
• overjet measured as the largest overlap parallel to the occlusal plane.
  

The plaster models were carefully packed and shipped to Czstochowa, Poland for the production of the digital virtual models (O3DM^TM, OrtoLab, Czstochowa, Poland), using a laser surface scanning technique. Once the digital models became available on-line, they were downloaded and the same measurements were performed using the O3DM^TM software (Figures 1 and 2).

View larger version (76K): [in this window] [in a new window]   Figure 1 Example of a digital model showing the measurement of the maxillary arch width (horizontal arrow) and maxillary arch length (vertical arrow)

View larger version (62K): [in this window] [in a new window]   Figure 2 Example of a digital model cut in a sagital plane to facilitate the measurement of the overbite and overjet

The measurements were practised by performing the measurements on a single plaster model and the corresponding digital virtual model ten times by the inexperienced observer.

	Statistical methods

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

 
Comparison of possible changes in the plaster models obtained from the three sets of alginate impressions was undertaken using a one-way ANOVA with a significance level of P=0.05. Intra- and inter-observer variability as well as inter-model variability was determined by calculating the error of the method from double measurements using Dahlberg’s formula.¹⁴ The validity of the measurements performed on the virtual models was assessed by comparing results obtained by the measurements of the ‘gold standard’ and the results obtained from the digital models. The differences between the 10 repeated measurements on the plaster and digital model were assessed by an independent sample t-test.

	Results

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

 
A delay of 3 to 5 days in pouring a plaster model from an alginate impression was found not to affect the accuracy of the model as no statistically significant differences were observed between the measurements performed on the plaster models obtained from the three sets of alginate impressions (Table 1).

	Discussion

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

The intra-observer variation was generally lower for the measurements on the digital models than on the plaster models, although there was a considerable variation in the error related to the different parameters. The error of the method was 0.09 to 0.38 mm for the measurements on the plaster models, while 0.05 to 0.28 mm for the measurements taken on the digital models (Table 2). This corroborated the findings of Bell et al. in a similar study.¹⁰ The difference in the intra-observer variation indicated the existence of a learning curve, both in relation to measuring on plaster and virtual models.

In spite of providing clear definitions for the individual parameters, some disagreement in the measurements between the observers occurred. Apart from the overjet measurement, the agreement between the observers when measuring on the digital virtual models was better (Table 3), thus supporting the findings of Costalos et al.¹⁵ This variation can be ascribed to a more precise definition and execution of the measuring protocol when using the measuring tools in the dedicated software.

Inter-observer agreement for measurements performed on the plaster and digital models was worse than the intra-observer agreement for the plaster models. This may be a somewhat surprising result as the same observer should have the same set of definitions for the measuring protocol in mind when executing them.

Santoro et al. found smaller values when measuring tooth widths on digital models.¹⁶ A similar consistent ‘width’ bias was not found in the present results. The variation in all measurements was clearly larger for the plaster models than for the digital models, in particular for the point-to-plane measurements (arch length, over-jet and overbite) the plaster models displayed variations 6 to 10 times larger than the ones in the digital models (Table 4).

Overall, a better accuracy and reproducibility was found for measurements taken from the digital virtual models. Quimby et al. and Zilberman et al. concluded that measurements obtained with a calliper were slightly superior to those obtained using the virtual measurement tools;^5,7 however the present study could not corroborate this. It should be noted here that a different brand of digital models and its associated visualization and analysis software had been employed and it could be that the new software measuring tools are easier to handle.

Based on the evaluation performed O3DM^TM digital models can fully replace the traditional plaster models as no clinically relevant difference could be established between the measurements obtained from the virtual model and the ‘gold standard’ and the reproducibility was better in the case of the virtual models. The typical size of O3DM^TM digital models is around 4 MB, which is more than OrthoCAD^TM files (~3 MB) and considerably more than E-models^TM files (~0.8 MB). The reason for this difference was partly the method, but also the resolution. Since digital storage is not a problem, 120 models can be stored on a CD and 1500 models can be stored on a DVD. The legal aspects related to replacing plaster models by digital models seem to have been solved with European law accepting the validity of virtual models provided the producer delivers a digital signature ensuring that the original data files have not and cannot be tampered with.

Copies of digital model files can easily be shared with colleagues for consultation and/or with patients as an extra stimulus for their motivation to comply with their treatment. The result of this study supports the replacement of plaster casts for diagnosis and planning within all fields of dentistry. In addition to replacing the plaster casts the digital models offer a long list of additional tools including: the possibility for cutting the model (Figure 2) in any plane of space to allow for assessment of the third order alignment of the individual teeth and superimposition on stable structures (Figure 3) making it possible to evaluate the changes generated during treatment. Finally the old storage spaces will surely find a better use in the future....

View larger version (83K): [in this window] [in a new window]   Figure 3 Example of a superimposition of a post- and pre-treatment digital model to visualize the tooth movements that have occurred during treatment

	Conclusions

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

• There was no statistically significant difference in the measurements taken from the alginate impressions that were cast immediately and those that had been in transit for several days.
  
• Measuring distances on plaster models gives rise to more intra- and inter-observer variability than measuring the same distances on digital models using virtual measuring tools.
  

	References

Top Abstract Introduction Materials and methods Statistical methods Results Discussion Conclusions References

 
1 Redmond WR. Digital models: a new diagnostic tool. J Clin Orthod 2001, 6: 386–7.

2 Redmond WR, Redmond WJ, Redmond MJ. Clinical implications of digital orthodontics. Am J Orthod Dentofacial Orthop 2002; 117: 240–1.[CrossRef]

3 Rheude B, Sadowsky PL, Ferriera A, Jacobson A. An evaluation of the use of digital study models in orthodontic diagnosis and treatment planning. Angle Orthod 2005; 75: 300–4.[Medline]

4 Paredes V, Gandia JL, Cibrian R. Digital diagnosis records in orthodontics: an overview. Med Oral Patol Oral Cir Bucal 2006; 11: E88–E93.[Medline]

5 Quimby ML, Vig KWL, Rashid RG, Firestone AR. The accuracy and reliability of measurements made on computer-based digital models. Angle Orthod 2004; 74: 298–303.[Medline]

6 Stevens DR, Flores-Mir C, Nebbe B, Raboud DW, Heo G, Major PW. Validity, reliability, and reproducibility of plaster vs digital study models: comparison of peer assessment rating and Bolton analysis and their constituent measurements. Am J Orthod Dentofacial Orthop 2006; 129: 794–803.[CrossRef][Medline]

7 Zilberman O, Huggare JA, Parikakis KA. Evaluation of the validity of tooth size and arch width measurements using conventional and three-dimensional virtual orthodontic models. Angle Orthod 2003; 73: 301–6.[Medline]

8 Ayoub AF, Wray D, Moos KF, et al. A three-dimensional imaging system for archiving dental study casts: a preliminary report. Int J Adult Orthod Orthognath Surg 1997; 12: 79–84.

9 Dirksen D, Diederichs S, Runte C, von Bally G, Bollmann F. Three-dimensional acquisition and visualization of dental arch features from optically digitized models. J Orofac Orthop 1999; 60: 152–9.[CrossRef][Medline]

10 Bell A, Ayoub AF, Siebert P. Assessment of the accuracy of a three-dimensional imaging system for archiving dental study models. J Orthod 2003; 30: 219–223.[Abstract/Free Full Text]

11 Kuroda T, Motohashi N, Tominaga R, Iwata K. Three-dimensional dental cast analyzing system using laser scanning. Am J Orthod Dentofacial Orthop 1996; 110: 365–9.[CrossRef][Medline]

12 Sohmura T, Kojima T, Wakabayashi K, Takahashi T. Use of an ultrahigh-speed laser scanner for constructing three-dimensional shapes of dentition and occlusion. J Prosthet Dent 2000; 84: 345–52.[CrossRef][Medline]

13 Chen SY, Liang WM, Chen FN. Factors affecting the accuracy of elastometric impression materials. J Dent 2004; 32 : 603–9.[CrossRef][Medline]

14 Dahlberg G. Statistical Methods for Medical and Biological Students. New York: Interscience Publications, 1940.

15 Costalos PA, Sarraf K, Cangialosi TJ, Efstratiadis S. Evaluation of the accuracy of digital model analysis for the American Board of Orthodontics objective grading system for dental casts. Am J Orthod Dentofacial Orthop 2005; 128: 624–9.[CrossRef][Medline]

16 Santoro M, Galkin S, Teredesai M, Nicolay OF, Cangialosi TJ. Comparison of measurements made on digital and plaster models. Am J Orthod Dentofacial Orthop 2003; 124: 101–5.[CrossRef][Medline]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dalstra, M. Articles by Melsen, B. Search for Related Content PubMed PubMed Citation Articles by Dalstra, M. Articles by Melsen, B.