Accuracy of Dental and Orthodontic Scanners used for Invisalign

Have you ever wondered about the orthodontic research that Dr. Mack and Dr. Hansen conducted in their residency programs for their Master's Thesis?

A few weeks back we posted Dr. Hansen's thesis regarding the differences in lip position as part of braces and a specific type of orthodontic treatment in adults! Today we are posting Dr. Mack's thesis which is based on the digital intra-oral scanner that we use for our invisalign patients! Her research was actually published in the American Journal of Orthodontics and Dentofacial Orthopedics which is sent to orthodontists all over the world! Check it out and let us know what you think!

ACCURACY OF CHAIR SIDE SCANNERS FOR SURFACE MEASUREMENTS ALONG A CURVED LINE

By Spencer M. Mack, DDS

INTRODUCTION:

Outstanding orthodontic care is based upon the correct diagnosis and creation of a proper treatment plan.  Correct diagnosis rests heavily on the use of photographs, radiographs, study models, and a clinical exam.  The proper diagnosis can easily be missed if any of these components are missing or inaccurate.  Therefore, it is crucial that all of these components be as descriptive and precise as possible in order to aid the clinician during this important phase of treatment.

Over the past few centuries, may advances have been made concerning the process of acquiring orthodontic records.  With the arrival of the computer age, the ability to move beyond plaster models became an actual reality.  Plaster models are a key component of orthodontic records, and any replacement method must be accurate.  Advancements in technology have allowed for digital photography to replace film, new sensors allowing for digital radiography, and maintenance of health records digitally.  An alternative to plaster models has been the last form of dental records to make the move to the digital world, but with advances in intraoral scanners, this technology has become more available to clinicians.  Digital models have many advantages, including elimination of storage space for plaster models, elimination of the need to search and retrieve these models, and durability.

There are currently many ways to acquire digital models, with many intraoral scanners to choose from.  Plaster models and impressions can be scanned to obtain digital models.  It is also possible to extrapolate models from cone beam computed tomography (CBCT) images.  There have been limited studies performed comparing the accuracies of these scanners to plaster models, but there is still limited information.  The purpose of this study is to evaluate and compare the digital dental models acquired from two commercial intraoral scanners with manual measurements when performing curvilinear measurements.

 

REVIEW OF LITERATURE

In order to diagnose a malocclusion and develop an appropriate treatment plan in orthodontics, a number of diagnostic aids are needed: study models, photographs, radiographs, and a clinical examination. ¹ In fact, the American Association of Orthodontists (AAO) updated their Clinical Practice Guidelines in 2003 to reflect that pre-treatment and post-treatment records should include both extraoral and intraoral photographs, dental models, intraoral and/or panoramic radiographs, and cephalometric radiographs. ²

Each of the elements of orthodontic records gives the provider a different insight into each case. However, it has been argued that study models hold the most value. In fact, a group published an article in 2004 reporting that in the majority of cases, adequate information for making treatment decisions can come from study models alone. ³ Han et al. looked at the consistency of orthodontic treatment decisions in relation to diagnostic records. In this study, it was shown that 55% of treatment plans formulated from study models alone did not change when information from photographs and radiographs became available. ⁴ Traditional radiographs and photographs provide only two-dimensional (2D) images, while study models are able to provide a three-dimensional (3D) view of the dentition. Study models also allow for more extensive measurements that may not be possible in a clinical examination where other factors (e.g., soft tissues) can get in the way.

 

Throughout the years, orthodontists have used a variety of methods to measure plaster casts: from methods as simple as overlaying the teeth with brass wire to measure arch perimeter to sectioning out the teeth for the creation of a diagnostic wax up. Today, the standard measuring techniques that new methods are evaluated against include dividers, calipers, and Boley gauges. ⁵

 

Plaster study models have many benefits and are crucial to a proper diagnosis; however, there are drawbacks and limitations to these models. While they are very accurate if done properly, this accuracy can be quickly jeopardized. The first step to a good model is a good impression. Without a properly made impression, the model holds no chance of being accurate. The first dental impression was made in 1684, and the process of impression making has since made drastic improvements. The most commonly used impression material in orthodontics today is the irreversible hydrocolloid alginate material, which first became available in the early 1900’s, along with reversible hydrocolloid materials. Alginate was the first material to allow details such as undercuts to be accurately captured in a single impression. Materials such as polyvinylsiloxanes (PVS) have since been introduced to the market and further improve impression accuracy while minimizing distortion – a problem that alginate can have if not used properly.  Despite polyvinylsiloxane’s advantages, alginate continues to be the predominate material used due to its ease of use and cost effectiveness. ²

In addition to the potential for inaccuracy, another downfall of plaster models is that they are subject to wear and breakage. Measuring devices can be the cause of this wear, which in turn decreases accuracy and may lead to increased chance of fracture. Careful handling of the casts is also vital, as one drop of the casts can lead to chipping or complete fracture. Another problem or nuisance of dental casts is storage space and convenience of retrieval. Dental casts are typically stored in special boxes and in a designated room. For liability purposes, these casts must be kept for a number of years, and for practices starting several hundred cases per year, space limitations may prevent keeping initial, progress, and final casts on all patients. Many offices do not have this type of space available for models, particularly in large cities. Additionally, many orthodontists work out of multiple locations, which further complicates finding appropriate storage.

 

As mentioned above, liability purposes add another element. For fear of being sued, many orthodontists keep models for a length of time determined by their state’s statute of limitations. This range is from 5-15 years for most states and may not begin until after the patient has turned 18. This means that although a Phase I treatment may have been completed when the patient was ten years old, the orthodontist may still need to keep modes on hand for an additional 23 years after their work has been completed. Therefore, time is a factor that must not be ignored.  In addition to keeping models for legal reasons, many clinicians chose to keep models for academic purposes as well. Taking serial sets of models or retaining casts for growth and retention studies demands even more space.

 

Yet another issue that arises with plaster models is the lack of ease of transferring or portability. Situations often occur when the practitioner would like to share or send plaster models. Whether it is to another provider due to the patient transferring or a learning situation where the provider wants to share the case with his or her local study club, there are times when models need to be transported.  A major problem with this is the potential for fracture or breakage. Even with proper packaging, there is no way to insure that the casts will not be damaged throughout the mailing process. Additionally, the costs of sending plaster models can be a factor as the weight of plaster increases baggage fees and shipping costs. In an effort to have a back-up cast in the case that something goes wrong along the way, models can be duplicated before shipment. This, however, requires both time and money and may not be feasible in a busy practice. ²

 

Many advances have been made in technology in the dawn of the new millennium. Digital radiography, digital photography, and digital records are quickly becoming the norm in orthodontic practices and institutions of higher education. ^1,4 Digital photography and radiography have proven to be more accurate and even more financially beneficial than traditional film. Additionally, new legislative policies seek to make paper charts obsolete.  For these reasons, it seems only fitting that now is the time to put away the dated plaster models and replace them with up-to-date digital models. However, before replacing the trusty stone models, the accuracy, reliability, practicality, and usefulness of these models must first be systematically evaluated. ⁵

 

Since digital scanners were first introduced to the dental field, they have been adapted to be used for a variety of in-office purposes. Initially, they were used solely as an outsourced technology for storage of 3D models. As intra-oral scanners were developed, they began being used to fabricate clear aligners, custom braces, indirect-bonding trays, and laboratory appliances efficiently and accurately. ⁶ A paradigm shift in orthodontics is represented by the replacement of alginate and PVS impressions with intraoral digital scanners. ⁷

 

Seeing the need for traditional models to enter the digital world, companies such as OrthoCAD, eModels, and OrthoSelect are providing the new standard. Such companies now offer services that allow a doctor to receive three-dimensional models of a patient’s dentition without purchasing any additional equipment or subjecting their patient to radiation. To do so, a doctor takes an alginate impression and mails either the impression or a plaster model to the company. After receiving the cast or impression, most companies use a proprietary laser scanner to generate the file. The scanned file is then digitally cleaned up by a technician and emailed back to the submitting dentist, with the entire process taking approximately one week. After the file is made available, the doctor can download and view it in the provided software where many desired measurements can be made. ⁸ The typical file size is less than one megabyte – so small that a single hard drive can easily store over two million sets of models.²

 

While there are benefits to using these digitizing services, there are a number of factors that orthodontists must consider before going forward and losing the plaster models. First of all, there is a fee that is assessed with each model sent in for digitization; this is in addition to the expense of obtaining the required impressions and bite registrations (and plaster models if the doctor chooses to take that route). Another factor is that orthodontists must wait a minimum of a week before their casts are available, assuming that the staff, shipping company, and digitation service all operate at maximum efficiency. If a doctor is interested in digital models, using a service like this is often the path of least resistance. As a result, large amounts of literature have been published on these techniques.

 

Variable conclusions have been shown in the literature when evaluating the accuracy of digitized models compared to their plaster counterparts, and little is known regarding the minimum degree of accuracy of intraoral scanners required for orthodontic treatment planning and patient consent.⁹  In a study performed by Zilberman et al., measurements made on OrthoCAD models were compared to caliper measurements on the original plaster casts.  Twenty sets of models were used, and measurements included tooth sizes and inter-tooth distances. Digital calipers on the physical models proved to have the highest level of accuracy and reproducibility, followed closely by OrthoCAD models. The author concluded that the digital models were clinically acceptable.¹⁰

 

Tomassetti et al. performed a study that found a much larger variance. For this study, Bolton analysis was performed physically using Vernier calipers as well as digitally in QuickCeph, the Hamilton Arch Tooth System, and OrthoCAD software. Researchers found no statistically significant error for any of the methods (with a confidence interval of p<.05), but they did note a significant difference of >1.5 mm for each method. ¹¹ This difference, however, could be attributed to their use of additive measurements that could compound any existing errors in measurements.
Stevens et al. performed Bolton analysis on plaster casts and eModels. Dental casts from 24 patients were included in the study, which used measurements to the nearest 0.01 mm. Paired t-tests revealed no significant difference between the digital and analog method, and even showed the reproducibility to be slightly higher in the digital models. The authors concluded that digital models are not a compromised choice for treatment planning or diagnosis.¹²

 

Costalos et al. used an entirely different technique for evaluating the reliability and accuracy of digital versus plaster models. Plaster and OrthoCAD models were taken on 24 patients and scored using an American Board of Orthodontics (ABO) measuring gauge and the seven criteria of the ABO Cast Radiograph Evaluation previously known as the Objective Grading System (OGS). The results of this study showed interproximal contacts were not significantly different between the two versions of the models. However, the means for alignment and buccolingual inclination were significantly different. The author did note that a study such as theirs is subject to examiner calibration, and that this could account for the significant difference they found.¹³

 

Like Costalos, Okunami et al. also looked at the ABO OGS. In this study, 30 sets of casts were evaluated in plaster and in the OrthoCAD software. Of the seven aspects included in the OGS, the buccolingual inclination could not be calculated due to software limitations (using version 2.2). The remaining six areas of scoring showed no significant differences between digital and plaster components, further supporting that the digital casts were sufficient.¹⁴

 

Santoro et al. used two examiners and 76 pretreatment patients to assess the accuracy of OrthoCAD in comparison to plaster models. The results showed a statistically significant difference between the two groups for tooth size and overbite, with no significant difference in overjet. While corresponding measurements were significantly different between casts, the differences were all less than 0.49 mm, which the author concluded would not make a clinical difference.¹⁵ The study could have been improved if measurements had been made after the crowding was resolved, allowing for easier and more repeatable measurements to be made.

 

A study from Whetten et al. evaluated Class II malocclusion patients and their corresponding treatment plans based on the method of acquiring models. Class II malocclusion cases are widely called the most difficult to treat and treatment plan. This study used 20 orthodontists and 10 Class II malocclusion subjects to determine if there is a difference in the resulting treatment plan if a doctor uses plaster or digital models (eModels). On two separate occasions each orthodontist treatment planned all ten cases with either the plaster or digital models. The researchers found no statistical difference in their treatment plans and concluded that digital orthodontic study models were a valid alternative to traditional plaster models for the treatment planning in Class II malocclusion patients.¹⁶

 

For just over two decades, Cone Beam Computed Tomography (CBCT) has been commercially available. Cone beam scanners involve a cone shaped beam of x-rays rotating around an object to produce a volume of data. Internal area of the object can then be detected by overlaying the series of 2D images.¹⁷ Many orthodontic practioners have begun incorporating CBCTs into their practices, especially as machine prices drop and availability increases. Numerous companies and machines are on the market today, with each machine differing slightly in their field of views (display volumes), voxel sizes, image detectors, and special resolution.¹⁸ The radiation emitted during a CBCT is equivalent to or only slightly greater than the radiation a patient receives from a traditional panoramic and cephalometric film. A great benefit of using a CBCT machine is that the acquired image not only allows for the generation of 3D study models, but also for the creation of a reconstructed cephalogram, which bears no significant difference to traditional cephalograms.¹⁹ However, while these CBCT machines produce significantly less radiation when compared to a traditional spiral CT, they do so at the expense of the image resolution.

 

CBCT derived images virtually eliminate certain issues with traditional radiography, including problems with magnification, distortion, and superimposition. A study done by Benington et al. claims that 3D images reconstructed from a CBCT are relatively free of these problems. ²⁰ As a result, many clinicians favor using this method. Additionally, Scarfe et al. state that CBCT imaging of the head and neck brings a true paradigm shift from 2D to 3D approach in imaging and that an appropriate diagnostic approach must soon follow. ¹⁷

 

The virtual study models generated from CBCTs allow for advantages previously unavailable with plaster models, such as the ability to see root and crown morphology in a single model, something Kapila et al. call “the anatomical truth.”¹⁹ Kau et al. found that these generated models were as accurate for measuring crowding, overjet, overbite, and making linear measurements as were OrthoCAD models.²¹ Creed also showed linear CBCT measurements to be more than adequate for diagnosis and treatment planning when compared to OrthoCAD measurements.²² Lightheart used surface analysis to show that models generated from CBCTs were “adequate” for initial diagnosis and treatment planning.²³ While Damastra claimed that CBCT scans acquired with 0.25 to 0.4 voxels of resolution show no difference in linear measurements, it must be noted that the amount of discernible tooth anatomy is greatly influenced by the voxel size.²⁴ Thus, not all CBCT scanners will be appropriate for generating study models.

 

Although a decrease in price has allowed for greater access to CBCT technology, issues like radiation exposure and the quality of the resulting image must still be considered before subjecting a patient to a scan, especially if it is done solely to produce a study model. Furthermore, doctors must be trained in interpreting any image they capture and must also make the financial commitment of purchasing the required CBCT machine. The practicality advances in CBCT technology allow for cheaper machines, variable fields of view, higher resolution scans, less radiation to the patient, and less doctor liability.

 

In a study from Akyalcin et al., values comparing virtual dental models were obtained from the Ortho Insight 3D laser scanner, the eModel system, and CBCT scans. The study showed that arch length discrepancy measurements made on digital model files made from these sources had similar patterns of random error when compared with direct caliper measurements. The best surface overlap correlation was observed between the virtual scanned models and the eModels. ²⁵

 

One alternative to the headaches imposed by model services or CBCT imaging is the use of intraoral scanners. Over the past ten years, numerous intraoral scanners have emerged onto the orthodontic scene, some from startup companies, and others from well-known dental supply companies. In their earliest stages, these scanners cost well over $100,000 and were highly technique sensitive. The field to be scanned had to be completely dry, and some companies even required you to lightly dust the surface with a fine powder in order for the enamel to properly register with the scanner. Too much powder would result in lack of detail, while not enough powder would result in the scanner being unable to generate an image. Each company would generate the scans in their own proprietary software format that could only be viewed or manipulated in their software. Furthermore, if a physical copy was desired, it had to be ordered directly from the scanner company.

 

Today, much has changed. Scanners cost significantly less than a new car, thrive in moist environments, and produce files that can be saved in a variety of formats or printed in office with equipment costing less that $1000.

 

The most widely available intraoral scanners are the iTero™ and the OrthoCAD iOC. Both are manufactured by Cadent and use the same laser and CMOS sensors. Since their debut in 2007, general dentists have used the iTero™ scanner to fabricate over 250,0000 restorations.^26,27  Orthodontists have taken notice, and in the summer of 2011, Align Technology, parent company to Invisalign, purchased Cadent to help make the iTero™ technology more readily available to orthodontists and streamline its integration into Invisalign case submission.

 

Unfortunately, despite the advances in scanning technology and all the time and money being invested by dental companies in research and development, few scientific papers have been published to verify scanner accuracy. In fact, more information is available about the financial status of the scanner manufacturers than on the scientific accuracy of the product they produce.   A systematic review by Goracci et al. investigated accuracy, reliability, and efficiency of intraoral scanners for full-arch impressions.   It was considered a remarkable observation that very few studies have evaluated complete-arch scans acquired directly in the patient’s mouth, and of all intraoral scanners available for use in orthodontics, only two (Lava COS and iTero™) have been tested in clinical settings.  Of the sixteen articles deemed relevant to this study, only eight published studies with complete-arch scans and only four studies reported data on validity, repeatability, and reproducibility of digital measurements.²⁸

 

A recent study from Akyalcin et al. showed that measurements taken by the iTero™ were acceptably interchangeable when compared to manual measurements, as well as interchangeable with measurements from a CBCT scan. This tells us that both the iTero™ and CBCT are sufficient to use in the diagnosis and treatment planning of orthodontic cases.²⁹

 

Ender and Mehl have published one of the most scientific articles to date on the accuracy of intraoral scanners. Their results show that after taking into account distortion from the impression material and stone, the digital models acquired using an intraoral scanner are just as accurate as traditional models.³⁰ However, the limitation to their study is that only five scans were performed on a single cobalt-chromium dentiform and not natural dentition. Accurately detecting enamel despite its translucency was a major obstacle that had to be overcome in the scanner development and must not be ignored when evaluating its accuracy.

 

A 2013 study from Flügge compared the precision of digital impressions taken intraorally by the iTero™ vs. the extraoral digitization with the iTero™ and a model scanner. It was shown that the precision of the intraoral iTero™ scan is similar to the values documented in the literature with the conventional polyether impressions for reproduction of the intraoral situation. However, scanning with the iTero™ intraorally is less precise than model scanning with it, which demonstrated that patient-related factors can influence the scanning process. ³¹

 

Grünheid et al. conducted a study in 2014 assessing the accuracy, reproducibility, and time efficiency of dental measurements using different technologies, including eModels, SureSmile, and AnatoModels.  Mesiodistal tooth-width measurements were made on models while being timed.  These measurements were compared with the plaster models, and differences in time efficiency were tested for statistical significance.  The data reported that SureSmile models were the most accurate and reproducible.  SureSmile models and eModels were also significantly faster than those taken on AnatoModels and plaster models.  It was concluded that linear measurements on digital models can be as accurate and possibly more reproducible and significantly faster than those measurements obtained from plaster models.  Another important take-away from this study comes from the conclusion that although plaster models are currently considered the gold standard, this should not suggest that they are measured without errors.  Digital models could result in more valid measurements than plaster models because of the lack of physical barrier dictating point placement for measuring.  Calipers are unlikely to reach the exact interproximal contact point of a tooth in contact with other teeth.  Therefore, neither method can be considered as providing unequivocally correct measurements.⁷

 

Another study by Grünheid et al. evaluated the clinical use of a direct chair side oral scanner with regard to accuracy, time and patient acceptance.  The accuracy of the digital models was compared with the existing standard, alginate impressions. Through the use of mathematical superimposition of the digital models, it was found that there was high relative accuracy of digital models made from intraoral scans, and these models can be as accurate as those from alginate impressions.  However, impressions required less chair side time and were favored over intraoral scans by most patients.  This study concluded that as scanning technologies become faster and more efficient, direct scanning will become more readily accepted in the orthodontic setting.³²

 

Another study by Patzelt et al. evaluated the time efficiency of intraoral scanners.  Three intraoral scanners were investigated: Cerac AC, Lava COS, and iTero™.  In contrast to the Grünheid study, the implementation of computer aided impression making with intraoral scanners improved work flow and lead to higher patient satisfaction.  Scanning for impression making was up to 23 minutes faster when considering all steps involved with traditional alginate impression material.  However, it was noticed that there are opportunities to reduce the actual chair time for both approaches by sharing several steps among the dental team members.³³

 

The Lythos scanner from Ormco was designed specifically for orthodontic use. As a newer product on the market, there is not a significant amount of research available about its accuracy. Advantages of the Lythos scanner include its compact size and its powder-free use during scanning. Images are captured by video in continuous motion and images are then stitched together. The software first captures a “backbone” from which future scans stem, which allows the program to have memory so that the scanner knows where to stitch the scans together. ⁶

 

Despite these products entering the market and proving to be acceptable for use in diagnosis and treatment planning, a number of factors contribute to the slow entrance into offices across the country. As mentioned earlier, cost is a factor. Over the past decade, however, cost of the scanner continues to decline and is within the manageable range for many practioners to purchase. For digital scans such as eModels, however, there remains the extra cost of storing the models in a database. Another factor is that taking intraoral scans requires more chair side time, particularly before adequate training, than alginate impressions. Older practitioners in particular may not see the benefit of purchasing a scanner at that stage in their career due to comfort in using traditional models and the cost and training that goes into transitioning to using digital scanners.

As technology improves and availability increases, orthodontists will frequently have new tools to assist in diagnosis and treatment planning.  It is the duty of the orthodontic community to consider the advantages and disadvantages of these new technologies to determine their clinical usefulness.  Like most new products, it will take time for scanners to be fully integrated into practices. It is evident that scanners, like other digital technology, are on the horizon of becoming the new gold standard in orthodontics.

 

REFERENCES:

 

1. Rhende et al. An Evaluation of the Use of Digital Study Models in Orthodontic Diagnosis and Treatment Planning. Angle Orthod. 2005;75:300-304.
2. J. Peluso et al. Digital Models: An Introduction. Semin in Orthod. 2004;10:226-238.
3. Joffe. Current Products and Practices OrthoCADTM: digital models for a digital era. J Orthod. 2004;Dec:31(4):344-347.
4. K. Han, K.W. Vig, J.A. Weintraub, P.S. Vig, C.J. Kowalski. Consistency of orthodontic treatment decisions relative to diagnostic records. Am J Orthod Dentofac Orthop. 1991;100:212-219.
5. L. Quimby, K.W. Vig, R.G. Rashid, A.R. Firestone. The Accuracy and Reliability of Measurements Made on Computer-Based Digital Models. Angle Orthod. 2004;74:298-303.
6. D. Kravitz et al. Intraoral Digital Scanners. J of Clin Ortho. 2014;Jun: 48(6):337-347.
7. Grünheid, N. Patel, N.L. De Felippe, A. Wey, P.R. Gaillard, B.E. Larson. Accuracy, reproducibility, and time efficiency of dental measurements using different technologies.  Am J Orthod Dentofac Orthop. 2014; Feb: 145:157-164.
8. Bell et al. Assessment of the accuracy of a three-dimensional imaging system for archiving dental study models. J. Orthod. 2003;Sept:30(3):219-223.
9. B. Martin, E.V. Chalmers, G.T. McIntyre, H. Cochrane, P.A. Mossey. Orthodontic scanners: what’s available? J Orthod. 2015; Jan 42:136-143.
10. Zilberman. Evaluation of the Validity of Tooth Size and Arch Width Measurements Using Conventional and Three-dimensional Virtual Orthodontic Models. Angle Orthod. 2003;73:301-306.
11. Tomassetti, L. Taloumis, J. Denny, J. Fischer. A comparison of 3 computerized Bolton tooth-size analyses with a commonly used method. Angle Orthod. 2001;71:351-357.
12. Stevens et al. 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 Dentofac Orthop. 2006;Jun:129(6):794-803.
13. A. Costalos et al. Evaluation of the accuracy of digital models analysis for the American Board of Orthodontics objective grading system for dental casts. Am J Orthod Dentofac Orthop. 2005;128:624-629.
14. Okunami et al. Assessing the American Board of Orthodontics objective grading system: digital vs plaster dental casts. Am J Orthod Dentofac Orthop. 2007;Jan;131(1):51-56.
15. Santoro et al. Comparison of measurements made on digital and plaster models. Am J Orthod Dentofac Orthop. 2003;124:101-105.
16. Whetten et al. Variations in orthodontic treatment planning decisions of Class II patients between virtual 3-dimensional models and traditional plaster study models. Am J Orthod Dentofac Orthop. 2006;Oct:130(4):485-91.
17. C. Scarfe et al. What is Cone-Beam CT and How Does it Work? Dent Clin North Am. 2008;Oct:52(4):707-30.
18. Kaeppler. Applications of Cone Beam Computed Tomography in Dental and Oral Medicine. Int J Comput Dent. 2010;13(3):203-19. English, German.
19. Kapila et al. The Current Status of Cone Beam Computed Tomography Imaging in Orthodontics. Dentomaxillofac Radiol. 2011;Jan:40(1):24-34. Review.
20. C. Benington et al. An Overview of Three-Dimensional Imaging in Dentistry. Dent Update. 2010;Oct:36(8):494-6, 499-500, 503-4.
21. H. Kau et al. Evaluation of CBCT Digital Models and Traditional Models Using the Little’s Index. Angle Orthod. 2010;May:80(3):435-9.
22. Creed et al. A Comparison of the Accuracy of Linear Measurements Obtained from Cone Beam Computerized Tomography Images and Digital Models. Semin Orthod. 2011;Mar:17(1):49-56.
23. Lightheart et al. Surface analysis of study models generated from OrthoCAD and cone-beam computed tomography imaging. Am J Orthod Dentofac Orthop. 2012;141(6):686-93.
24. Damstra et al. Accuracy of linear measurements from cone-beam computed tomography-derived surface models of different voxel size. Am J Orthod Dentofac Orthop. 2010;Jan:137(1):16.e1-16.e6.
25. Akyalcin, D.J. Dyer, J.D. English, and C. Sar. Comparison of 3-dimensional dental models from different sources: Diagnostic accuracy and surface registration analysis. Am J Orthod Dentofac Orthop. 2013;Dec:144(6):831-837.
26. Garino, B. Garino. The iOC Intraoral Scanner and Invisalign: A New Paradigm. J of Clin Orthod. 2012;Feb:46(2): 115-121.
27. Redmond, et al. The OrthoCAD iOC Intraoral Scanner. A Six Month User Report. J of Clin Orthod. 2011;March:45(3): 161-164.
28. Goracci, L. Franchi, A. Vichi, M. Ferrari. Accuracy, reliability, and efficiency of intraoral scanners for full-arch impressions: a systematic review of clinical evidence. Eur J Orthod.  2015; Oct: 1-7.
29. Akyalcin, B.E. Cozad, J.D. English, C.D. Colville, S. Laman. Diagnostic accuracy of impression-free digital models. Am J Orthod Dentofac Orthop. 2013;Dec:144:6, p916-922.
30. Ender, A. Mehl. Full Arch Scans: Conventional Versus Digital Impressions – an In-vitro Study. Int J Comput Dent. 2011;14(1): 11-21.
31. V. Flügge, S. Schlager, K. Nelson, S. Nahles, M.C. Metzger. Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a model scanner. Am J Orthod Dentofac Orthop. 2013;Sept: 133(3):471-478.
32. Grünheid, S.D. McCarthy, B.E. Larson. Clinical use of a direct chair side oral scanner: An assessment of accuracy, time and patient acceptance. Am J Orthod Dentofac Orthop. 2014; Nov: 146(5): 673-682.
33. B.M. Patzelt, C. Lamprinos, S. Stampf, W. Att. The time efficiency of intraoral scanners: An in vitro comparative study.  JADA. 2014; June: 145(6):542-551

 

 

ABSTRACT

 

AIM:

Chair side scanners offer the advantage of obtaining digital dental models directly from the patient without the need for dental impressions. To date, no study evaluated their accuracy in obtaining surface measurements along a curved line (curvilinear). The aim of this study was to evaluate and compare the digital dental models generated from two commercial intraoral scanners with manual measurements when performing curvilinear measurements.

MATERIALS AND METHODS:

The study sample was comprised of 61 dry mandibles with intact dentition chosen from a unique collection. Each skull had the mandibular arch scanned with Cadent iTero™ scanner (Align Technology, San Jose, CA) and Lythos™ Digital Impression system scanner (Ormco Corporations, Anaheim, CA). Surface measurements along a curved line were performed digitally in three different directions (anteroposterior, mesiodistal, and buccolingual) on the digital models and manually on the dry skulls. One-sample t test and linear regression analyses were performed. To further graphically examine the accuracy between the different methods, Bland-Altman plots were computed. Level of significance was set at p<0.05.

RESULTS:

There were no significant differences between any of the paired methods, which indicated a certain level of agreement between the methods tested (p>0.05). Bland-Altman analysis showed no fixed bias of one approach vs. the other, and random errors were detected in all comparisons. Although mean bias of iTero™ and Lythos™ scanner measurements when compared to direct measurements were very low, minimum mean bias occurred for the comparison of two intra-oral scanners. However, Lythos™ scanner and direct measurement comparison had the largest confidence interval (agreement level, 0.85 to -1.11) range.  None of the comparisons displayed statistical significance for the t scores, which indicated the absence of proportional bias in these comparisons.

CONCLUSIONS:

Intraoral scanners tested in this study produced digital dental models that were highly accurate when performing direct surface measurements along a curved line in three dimensions of the space.

 

 

INTRODUCTION

 

The last few decades have shown significant advancement in the orthodontic field, especially with regards to technology.  Much of the dental field, including orthodontics, has been steadily transitioning to the digital world.  The digital world provides a means for more efficient and sophisticated systems, and more and more clinicians are pursuing these kinds of digital-based practice models.  Many practitioners have embraced digital photographs, digital records, and digital radiographs. Diagnostic models have been a little slower to make the digital transition; however, over the last decade, more and more orthodontists are now moving in the direction of digital models as well.¹

 

Intraoral scanners were initially used for fabricating restorations in general dentistry.  Their increased use in the dental field led to orthodontists’ utilization of these scanners.^2,3  A number of appliances such as retainers, expanders, etc., as well as clear retainers and indirect bracket set-ups can all be produced directly from an accurate digital model.⁴ Because of the ease and efficiency of these scanners, orthodontists are now using these services daily with the expectation of speeding up treatment time and/or reducing costs.   With a digital model, the time and cost of taking impressions and sending them to the laboratories can be minimized, if not completely eliminated.  Indirect bracket set-ups performed correctly can be extremely accurate, helping to decrease the need for bracket repositioning later in treatment, thus reducing overall treatment time. ⁵

 

A number of recent studies have assessed the accuracy of several different digital model platforms. These studies evaluated digital models produced by intraoral scanners used both extraorally on plaster models and intraorally, from extraoral scanners, and from 3-dimensional (3D) radiographs. Overall, the findings show that digital models are just as accurate as traditional models and are appropriate to use in the diagnosis and treatment planning of orthodontic cases.^6,7,8

 

To date, there have been no studies evaluating the accuracy and validity of the different intraoral scanners along a curved line. The aim of this study was to investigate the curvilinear accuracy of two commercial intra-oral scanners, the Lythos™ scanner and the iTero™ scanner, in comparison to indirect digital caliper measurements when using a nylon monofilament to measure directly on the same dentition.

 

 

 

MATERIALS AND METHODS

 

The study sample was comprised of 61 dry skulls with intact dentition selected from a unique collection at the University of Texas Health Science Center at Houston School of Dentistry. The mandibles from each skull were used for this study. No IRB approval was needed for this study, as cadavers are no longer considered human subjects.

 

Lythos™ Digital Impression system (Ormco Corporations, Anaheim, CA) was used to scan each mandible running software version 1.9.10398. Once the scans were completed, the raw images were available for chair side viewing.  Images were then sent via the Internet to Ormco Corporation where they were immediately made available via a secure webpage for download as stereolithography (STL) files.  STL files are the industry standard for computer-aided design (CAD) and allow the viewer to be able to view and manipulate the object in a number of different applications while preserving the quality of the image.  Next, a Cadent iTero™ scanner (Align Technology, San Jose, CA) was used to scan the 61 dry skull mandibles running software version 5.2.1.290.  These scans were also available for chair side viewing within minutes and were then sent via the internet to Align Technology where they were stored as STL files and made available for viewing. See Figure 1 for a photograph and STL files from both the Lythos™ and iTero™ scans of a mandible used in the study.

 

Digital calipers (Carrera Precision CP5906 0-6-Inch Electronic Digital Caliper, Carrera Precision) were used to indirectly carry out initial measurements on the dry skulls after using nylon monofilament to directly obtain the curvilinear measurements on the skulls.  Vertical curvilinear measurements were obtained by measuring from the tip of the lower right canine tip directly down to the crestal bone.  In cases where permanent canines were not yet erupted, the primary canines were measured.  Sagittal curvilinear measurements were obtained by measuring the crestal bone along the lingual of the lower right first molar.  Transverse curvilinear measurements were acquired by measuring the lingual cusp tip of the lower left second premolar, starting from the center of the mesial marginal ridge and ending on the center of the distal marginal ridge (See Figure 2 for illustrations). In instances where the second premolars had not yet erupted, the primary 2^nd molars were measured.  All measurements were carried out by the same operator to the nearest 0.01 mm, and the digital caliper was zeroed out prior to every measurement to assure increased accuracy.

 

The files from both the Lythos™ and iTero™ scans were opened as STL files and measured electronically using 3-matic Research 9.0 (x64) by Materialise (Materialise, Belgium).  The same operator carried out all of these measurements, and the same three dimensions that were measured directly on the mandibles were measured electronically on these STL files.  Measurements were carried out on unsectioned, shaded models of the mandibles to the nearest 0.01 mm with the software’s built in ruler tool.  The ruler tool was set to measure the distance over a surface with a curve creation method set on the true shortest path of a curve using the World Coordinate System.

 

Measurements for each tooth were paired in twos for the methods tested. A one-sample t test was used to test the hypothesis that there would be no difference between the two measurements. The test value was set at 0. No significant differences were found for any of the paired measurements. Therefore, measurements from the same model were averaged for the iTero™, Lythos™ and direct measurements. Bland-Altman analysis was performed using XLSTAT Mac (version 2012; Addinsoft, New York, NY) and Bland-Altman plots were computed for the paired comparisons of the three methods. The analysis was used to visually demonstrate the agreement for measurement values between the manual caliper measurements, the Lythos™ scan measurements, and the iTero™ scan measurements.  The within-observer repeatability was evaluated using intraclass correlation analysis (ICC) by repeating all of the measurements from ten randomly selected skulls at a one-month interval.  Furthermore, linear regression analyses were used to investigate whether there was a proportional bias in the data. The level of significance was set at p<0.05 for all tests.

 

RESULTS

 

The intra-observer repeatability for each paired measurement set was excellent with the ICCs ranging between 0.96 and 0.98. Mean bias, SD, confidence interval, and p values for the paired method comparisons are demonstrated in Table I. According to the one-sample t test, there were no significant differences between any of the paired methods, which indicated a certain level of agreement between the methods tested. Bland-Altman plots of the method comparisons are shown in Figures 1-3. Bland-Altman analysis showed no fixed bias of one approach vs. the other, and random errors were detected in all comparisons. Although mean bias of iTero™ and Lythos™ scanner measurements when compared to direct measurements were very low, minimum mean bias occurred for the comparison of two intra-oral scanners.  However, the Lythos™ scanner and direct measurement comparison had the largest confidence interval (agreement level, 0.85 to -1.11) range.

 

Linear regression analyses were summarized in Table II. None of the comparisons displayed statistical significance for the t scores. This indicated that there was no proportional bias in any of these comparisons. Therefore, the null hypothesis was accepted. In other words, no trend was shown for any of the mean differences being above or below the mean bias level shown in Bland-Altman scattergram plots.

 

DISCUSSION

 

The precision and validity of the Lythos™ intraoral scanner and the Cadent iTero™ scanner along a curved line were evaluated in this study.  Results were based on comparison to indirect caliper measurements along a curved line when using a nylon monofilament fiber to directly measure from the mandibular dentition. Based on our findings, it is evident that intraoral scanners, the Lythos™ and iTero™, can be used interchangeably with caliper measurements for diagnosis and treatment planning.

 

Dry skull caliper measurements served as the gold standard reference.  Based on our findings, the confidence intervals for the Lythos™ scanner and direct measurements had the largest confidence interval (agreement level, 0.85 to -1.11) when compared to the intervals for the iTero™ scanner and direct measurements.  This was likely due to the difficulty in repeating the landmark locations, particularly along the crestal bone of the molars and the lingual cusp tip landmarks of the premolars, as it can be somewhat ambiguous.

 

One of the main problems with the direct measurements on the skulls, as well as the digital dental models is the consistent identification of landmarks.  It was also difficult measuring curved lines on the actual mandibular dentition with the nylon monofilament.  The filament was marked from the beginning of a landmark and wrapped carefully along the curvature being measure.  Once the second landmark had been reached, another mark was placed on the monofilament and the calipers were used to obtain the final measurement.  These marks were placed with a 0.2 mm extra precise, ultra fine point black Sharpie pen.  The narrowed tip allowed for improved control during measurements, and the fast drying, smear proof, fade- and water-resistant ink contributed to the consistency of the measurements.  It was much easier using the digital scanners to measure these curved lines since multiple points could be marked along the curved line and measured section by section; there was also increased ability to place points directly in smaller spaces that were difficult to reach when directly measuring the skulls.  The operator tried to minimize human error when directly measuring the skulls as well as digitally attempting to duplicate the same landmarks as much as possible, but inevitably, there will be differences.  As mentioned above, reaching the interproximal edges along the crestal bone was often difficult with the direct measurements on the dry skulls, as the contacts with other teeth precluded it. The calipers were likely mildly inconsistent for these values as well, again, due to the neighboring teeth blocking the points of measurement.  The results reflected that the digital scanners matched almost perfectly in their agreement, which confirms their ease of use when measuring curved lines digitally.  These scanners’ results were closer in agreement than either scanner to the direct measurements, though the Lythos and direct measurement comparison has a larger confidence interval.  Though there is a possibility for deformation of a STL file, or losing information on a millimetric level, it is an insignificant occurrence as this loss is so minimal.

 

One of the goals in evaluating the accuracy of these scanners along a curved line was for future use of digital fabrication of lower fixed canine-to-canine (3-3) retainers.  Lower fixed 3-3 retainers could be fabricated from final digital models with a high level of accuracy and delivered to the patient at the debond appointment, increasing time efficiency and reducing chair time.  Another curvature that could be measured with these scanners is the curve of Spee. The curve of Spee refers to the anterior-posterior curvature of the occlusal surfaces, beginning with the cusp tip of the lower cuspid and following along the buccal cusp tips of the premolars and molars.  This is a very significant and often overlooked curvature of the dentition that is important when leveling and aligning in orthodontic treatment.⁹  Accurately measuring this curve on digital models can aid in treatment planning, progress and final evaluations.

 

Intraoral scanners were first used by restorative dentists for the production of fixed prosthodontic restorations.  These intraoral scanners hold promise since they only emit optical radiation.   A 2013 study¹⁰ proved that three of the leading intraoral scanners (Lava C.O.S. [3M Espe], CEREC [Sirconia], and iTero™ [Strauman]) were able to create highly accurate restorations with as much accuracy as a 2-step putty-and-wash technique.  Many previous studies were performed on plastic or chromium cobalt dentoforms, which have different refractive indexes than tooth structure.  This led researchers to perform a study later in the year testing the accuracy of the iTero™ on natural dentition to ensure that the prismatic properties of enamel and its effect on the accuracy of the scanner would not be overlooked.  Aforementioned study confirmed that measurements acquired from iTero™ scans are clinically acceptable for orthodontic diagnosis and treatment planning. ⁶

 

Another study by Nedelcu and Persson evaluated the accuracy and precision of four intraoral scanners (3M Lava COS, Cerec AC/Bluecam, E4D, and iTero™) to assess the influence of different test materials and coating thicknesses.  Models were fabricated with three materials (polymethyl methacrylate [Telio CAD], titanium and zirconia) and were scanned with an industrial optical scanner.  Telio CAD was used to simulate tooth substance because of the similarities of its refractive index to enamel and dentin.  These models were then scanned with the intraoral scanners; a thick layer of coating was applied when scanning with the 3M Lava COS.  These scanners displayed similar results when comparing deviations, while maximum deviations were noted in the noncoating scanners.  The iTero™ scanner displayed consistent material-specific, localized errors on the Telio CAD translucent material, while the E4D displayed the largest deviations.  Excessive coating showed no negative affect.   The use of 3D analysis software has become a norm within reverse engineering and has been adopted into the field of dentistry.¹¹  Reverse engineering is the process of determining the technological principles of an object or component through analysis of its structure and function.  This analysis can then be used to redesign the object rapidly using computer-aided design (CAD) in concert with rapid-manufacturing processes to produce small numbers of components adapted to the needs of the customer.  This has huge benefits of speed and flexibility over traditional mass-production-based design and manufacturing processes.¹² In this study, scanners using still-image acquisition indicated errors due to pattern recognition, where stitching of overlapping images could not be combined into larger segments.  It was also noted that the visual evaluation of high-resolution images found that triangle density and mesh topology varied greatly among the scanners, contradicting previous thoughts that scanners with higher-definition sensors would produce higher accuracy.¹¹

 

The use of natural dentition is considered as one of the strengths of this current study.  Evidence shows that both reflection and refraction occur when light moves between materials.¹³   As mentioned above, a scanner’s diagnostic accuracy can present differences when used on plastic and chromium-cobalt dentoforms versus natural dentition.  Plastics exhibit a variety of different refractive indexes, all of which are different from the refractive index of tooth structure. Differences in light reflected back to the scanner can affect the scanner’s ability to determine the actual depth of an object accurately.⁶   Because of these differences,  natural dentitions from the 61 dry skulls were used in this study in order to eliminate the chance for refractive indices to influence the results.  Our current study was not done in vivo because of the difficulty involved when obtaining direct caliper measurements from a patient’s mouth.  Impression materials would be necessary to obtain plaster model replications of the patient’s dentition, and as reported in previous studies, impression materials and stone are prone to deformation.¹⁴  It was important when designing our study to be able to compare digital measurements directly to the main object to eliminate the above mentioned inconsistencies with materials. Results may also differ in an in vivo setting due to the presence of blood, saliva and patient movements,¹⁵ and because we were testing the accuracy of these scanners along a curved line, we wanted to eliminate these unpredictable variables.

 

Flügge et al. compared virtual models produced from stone models scanned by an extraoral scanner (D250, 3Shape, Copenhagen, Denmark) versus those same models scanned by the iTero™, versus intraoral scans taken by the iTero™.  The lowest precision was demonstrated with the intraoral scans; this was likely due to the condition of a wet and unpredictable intraoral environment versus a dry extraoral environment.  The extraoral scanner used on plaster models showed the greatest precision of the group. The mean difference between the two groups was 27 micrometers, a very small difference. ⁸ This study also helped support an in vitro study design by demonstrating the lower precision of the intraoral scans when subjected to the wet intraoral environment.

 

A study by Cuperus in 2012 evaluated the validity and reproducibility of measurements using the Lava scanner from 3M ESPE versus caliper measurements taken on skulls. However, they used only ten dry human skulls in their study, and they printed their STL files into 3D models. Statistically significant differences were noted between the measurements of the STL and digital models, but these differences were considered to be clinically insignificant.  It was concluded that the measurements made with the Lava scanner were valid and reproducible for measuring distances in dentition.¹⁶ Because this study investigated the accuracy between digital files and 3D printed models and concluded the differences to be clinically insignificant, it was decided to proceed with our current study without printing models and without making impressions of the dry skull mandibular dentition.

 

A study by Campanelli et al. implied that the accuracy of laser scanners depends on the shape, texture, and material reflectivity of the object being scanned.  A complex object, the distal femur, and a simple object, a gage block, were scanned by five laser scanners (Nikon, Laser Design Inc., Creaform, Northern Digital, and NextEngine).  The repeatability in the 3D models generated by the laser scanner was less repeatable when scanning the more complex distal femur than when scanning the gage block.  In other words, the repeatability of the 3D models generated by the laser scanners tested was worse on average when scanning a complex freeform surface.  Little difference between the scanners, aside from bias, was also noted.¹⁷

 

A study published in January of 2016 compared digitally fabricated custom edentulous mandibular trays with handmade trays.  The scanned data was imported into a reverse engineering software, and after statistical analysis, it was discovered that the digitally made custom trays achieved good matching with the mandibular model and also showed higher accuracy than the handmade trays.  An efficient method for custom tray fabrication was accomplished using 3D scanning, computer-aided design and fused deposition modeling, and this method resulted in high reproducibility and accuracy.¹⁸

 

The importance of measuring lines along the curve and the accuracy of these measurements lie within the ability of these scanners to properly stitch together these images and transform the geometric information into 3D models.  In our current study, we attempted to evaluate whether digital models are prone to geometric distortion during this transformation, and if so, how much distortion existed.  In a 2012 study by Van der Meer et al., the accuracy of three intraoral scanners was compared.  A master model was scanned by the CEREC (Sirona), iTero™ (Cadent) and Lava COS (3M) scanners, and the distance between the centers of the cylinders and the angulation between the cylinders were measured and evaluated against a high accuracy 3D scan of the master model.  Small differences were noted between these scanners with none being statistically significant.  It was noted that differences in the results may be attributed to the registration of the 3D images and post-processing procedures.  How the registration takes place and what algorithms were used in the different scanners was not shared knowledge, but algorithms involving registration based on surface overlap was most likely.  Registration errors will always occur in the registration procedures, however minute, but one would expect this to have an additive effect over the length of an arch.  This explained how there might be an accumulation of registration errors of the patched 3D surfaces with increased length in the distance and angular measurements, but the effects of the accumulation were deemed statistically insignificant. ¹⁹

 

The ease of measuring along a curve on the digital models is an important point to note in this current study.  Digital models can be manipulated easily on the computer screen to reach many different angles.  Placing points on the digital skulls with the computer mouse was much easier than manually trying to use nylon monofilament to reach the difficult and small interproximal areas on the mandibular dentition.  The mean bias between the iTero™ and direct measurements (-0.17) was slightly more than between the Lythos™ and direct measurements (-0.13); however, the smallest mean bias, noted at -0.03, existed between the two intraoral scanners.  This would confirm the consistency and ease of use when measuring these scanned models digitally rather than manually.  As noted in the above results, the Lythos™ and direct measurement comparison did have the largest confidence interval (agreement level, 0.85 to -1.11) range.   None of the comparisons exhibited statistically significant p values, indicating everything was homogenous and no proportional bias existed between the clusters of points in one area on all three Bland-Altman plots.

 

There is no current literature evaluating intraoral scanners’ accuracy in measuring along a curved line.  Future projects could use intraoral scanners on actual patients rather than skulls to give more realistic results as patient factors such as saliva and soft tissue may play a role in affecting the scanning process and accuracy.  Data may also be more consistent if one mandible was measured with all methods in the same sitting to insure more consistency with landmark identification of points.  It would also be beneficial to mark points on the skulls prior to scanning them so that landmark identification would be more precise throughout all measuring procedures.

 

CONCLUSIONS

 

Our study shows that the intraoral scanners tested in this study, the Lythos™ and iTero™, produced digital dental models that were highly accurate when comparing to direct surface measurements along a curved line in three dimensions of the space.  Both scanners are interchangeable with direct caliper measurements, if not more accurate and require a more effortless measuring approach.

 

REFERENCES

 

1. J. Peluso et al. Digital models: An introduction. Semin in Orthod. 2004;10:226-238.
2. Garino, B. Garino. The iOC intraoral scanner and invisalign: a new paradigm. J of Clin Orthod. 2012;Feb:46(2): 115-121.
3. Redmond, et al. The orthoCAD iOC intraoral scanner. A six month user report. J of Clin Orthod. 2011;March:45(3): 161-164.
4. D. Kravitz et al. Intraoral digital scanners. J of Clin Orthod. 2014;Jun: 48(6):337-347.
5. J. Redmond, M.J. Redmond, W.R. Redmond. The OrthoCAD bracket placement solution. Am J Orthod Dentofac Orthop. 2004;May:125(5):645-646.
6. Akyalcin , B.E. Cozad, J.D. English, C.D. Colville, S. Laman. Diagnostic accuracy of impression-free digital models. Am J Orthod Dentofac Orthop. 2013;Dec:144:6, p916-922.
7. Ender, A. Mehl. Full arch scans: Conventional versus digital impressions – an in-vitro study. Int J Comput Dent. 2011;14(1): 11-21.
8. V. Flügge, S. Schlager, K. Nelson, S. Nahles, M.C. Metzger. Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a model scanner. Am J Orthod Dentofac Orthop. 2013;Sept: 133(3):471-478.
9. G. Spee. Prosthetic dentistry.  Medico-Dent Pub Co. 1928.
10. Seelbach, C. Brueckel, B. Wostmann. Accuracy of digital and conventional impression techniques and workflow. Clin Oral Investig. 2013; 17:1759-1764.
11. G. Nedelcu, A.S.K. Persson. Scanning accuracy and precision in 4 intraoral scanners: an in vitro comparison based on 3-dimensional analysis. J of Pros Dent. 2014; Dec:112(6): 1461-1471.
12. Raja, K. J. Fernandes. Reverse engineering: an industrial perspective. London: Springer; 2008: 1-69.
13. Crew. The wave theory of light – Memoirs by Huygens, Young, and Fresnel. American Book Company, New York (1900).
14. Torassian, C.H. Kau, J.D. English, J. Powers, H.I. Bussa, M.A. Salas-Lopez, J.A. Corbett. Digital models vs plaster models using alginate and alginate substitute materials. Ang Orthod. 2010;Jul;80(4): 474-481.
15. D. Hack, S.B.M. Patzelt. Evaluation of the accuracy of six intraoral scanning devices: an in-vitro investigation. The ADA Prof Prod Rev. 2015; Sept:10(4).
16. M.R. Cuperus, M.C. Harms, F.A. Rangel, E.M. Bronkhorst, J.G.J.H. Schols, K.H. Breuning. Dental models made with an intraoral scanner: A validation study. Am J Orthod Dentofac Orthop. 2012;Sept:142(3):308-313.
17. Campanelli, S.M. Howell, M.L. Hull. Accuracy evaluation of a lower-cost and four higher-cost laser scanners. J of Biomech. 2015; Nov: http://dx.doi.org/10.1016/j.jbiomech.2015.11.015.
18. Chen, X. Yang, L. Chen, Y. Wang, Y. Sun. Application of FDM three-dimensional printing technology in the digital manufacture of custom edentulous mandible trays. Sci Rep. 2016; Jan:6 (19207).
19. J. Van der Meer, F.S. Andriessen, D. Wismeijer, Y. Ren. Application of intra-oral dental scanners in the digital workflow of implantology. PLoS One. 2012; 7:e43312.

 

 

Figure 1. Example case as seen in photos, Lythos™ scan, and iTero™ scan.

Figure 2. Diagrams of measurements taken, demonstrated on iTero™ digital models. (From left to right: An overall view of measurements, the facial curve of the canine, the lingual cusp tip of the premolar, and the crestal bone along the molar).

 

Table I. Mean bias, SD, confidence interval, and p values for the paired method comparisons.

	Mean Bias	SD	+1.96 SD	-1.96 SD	p
iTero – Direct Measurements	-0.17	0.4	0.60	-0.95	0.057
Lythos – Direct Measurements	-0.13	0.5	0.85	-1.11	0.061
iTero – Lythos	-0.03	0.4	0.75	-0.81	0.468

 

 

Table II. Summary of the linear regression analysis.

	t	p
iTero – Direct Measurements	0.13	0.89
Lythos – Direct Measurements	0.10	0.91
iTero – Lythos	0.26	0.79

 

Figure 3. Bland-Altman plot for iTero™ and direct measurements

 

Figure 4. Bland-Altman plot for Lythos™ and direct measurements

 

Figure 5. Bland-Altman plot for iTero™ and Lythos™ measurements

 

If you have any additional questions about Mack and Hansen Orthodontics or want to discover how Dr. Spencer Mack and Dr. Andrew Hansen can help you gain straighter teeth and a healthier smile with braces or Invisalign please contact us today! You can also visit www.etxortho.com to learn more about our orthodontists or dental braces,. We’re conveniently located in Longview, TX right by Longview Regional Hospital and are just a quick drive from Marshall, Carthage, Henderson, Kilgore, and Gilmer.