Читать книгу Digital Workflow in Reconstructive Dentistry - Группа авторов - Страница 11
ОглавлениеDesktop scanners have been in use in dental laboratories for more than 20 years. The scanners can digitize different types of stone casts and conventional alginate impressions for orthodontics and diagnostics, as well as rubber materials.1 Different fields require distinct aspects of technical capability in assessing surface topography. The general advantage of this lab process is the absence of intraoral difficulties such as saliva, blood, and soft tissue, which can interfere with the outcome while performing an intraoral scan. Additionally, working casts mounted in the articulator are a favorable tool for many lab procedures, especially complex prosthodontics. While early scanning devices utilized touch probe (tactile) scanning, contemporary scanners utilize optical technologies. The first commercially available desktop scanner for the dental laboratory, in the early 1990s, was the Procera “reader” used for an early CAD/CAM (computer-aided design/computer-aided manufacturing) process to fabricate single crown copings by milling and spark erosion for the veneer technique. This device was considered to have a maximum shape-related error of ±10 µm (Fig 3.1a).2 Even today, tactile scanning is recognized as a very precise technology due to the lack of problems which are inherent to optical methods. In dental technology, the accuracy of the tactile systems is used for the scanning of primary crowns or bars for the design and fabrication of retrievable secondary components made of various materials.
Fig 3.1 (a) The first technology for desktop scanners in dental CAD/CAM was a tactile system (Procera, Nobel Biocare, Kloten, Switzerland; formerly Nobelpharma Procera). (b) A desktop scanner with simultaneous scanning of two casts and integrated resolution for the dies in a single scan process (D2000, 3Shape, Copenhagen, Denmark). (c) Original object (left), data of scan with rounded corners (middle), and both overlaid to visualize the difference (right).
Currently available optical desktop scanners have vastly improved in all facets, such as speed, accuracy, and scanning capabilities.3 The software programs which are used to run these devices to generate and calculate the data are also constantly improving. This is noticeable in the decrease of the scanning and calculation times. A present system by 3Shape (D2000, released in 2016) is able to scan two casts and digitize the dies at a high resolution simultaneously (Fig 3.1b).
However, three-dimensional (3D) scanning must generate a highly accurate, if not identical, digital reproduction of the surface topography of the object represented by a point cloud in a system of three spatial coordinates (by convention, x, y, and z). The number and accuracy of the measured points in the spatial space is responsible for the accuracy of the digital reproduction of the scanned surface. Multiple factors, such as the translucency of the object or its reflecting surface (related to the optical scan), the hardware of the scanner regarding its axis or the size of its tactile probe, and the calculating software, influence each scan and the accuracy of the result. Beside these factors, the time required for a single scan or multi-die model is economically significant (Fig 3.1c).
Laser scanning and optical-stripe light projection represent optical measurement technologies primarily deployed for the 3D digitization of the surface of a dental cast or impression. They are both based on the same principle of triangulation; light structures (usually stripes) are projected onto the object, and light sensors acquire the image; using values from the known geometry of the setup, 3D information can be calculated from the imaged data.
The difference between the two methods resides in the way the light structure is projected and imaged. In laser scanning, laser light sources are used to project one or multiple thin, sharp stripes onto the scanned object; in optical-stripe light projection systems, light patterns (usually in the form of a bundle of stripes) are projected onto the entire object being scanned. While laser scanners were previously considered to provide higher resolution and accuracy, both methods today offer a precision of 10 μm or less.3 The spectrum of applications for current desktop scanners varies from scans of individual dies to scans of arch segments, complete arch casts, impressions, bite records, and wax-ups. To improve scan accuracy, some scanners require the application of a nonreflecting agent, while others require the use of casts made of scan-compatible gypsum. Newly introduced scanners include high-resolution cameras that allow for adaptive impression scanning, as well as texture-capture capabilities and surface coloring (Figs 3.2 and 3.3).
Fig 3.2 Scan of a conventional dental impression.
Fig 3.3 Image of a scanned dental impression.
Usually, the scanner features at least two capture devices (two cameras or two laser sensors) while the holder for the model (object to be scanned) is attached to a multiple axis rotation mechanism, allowing the object to be rotated and tilted during the scanning process. This motion feature enables the capture devices to effectively detect the object from any viewpoint; scan all details of the object, including undercuts; increase the size of the measuring field; and improve the overall accuracy of the scan data. Recently introduced desktop scanners can scan two models at the same time and do not require an extra scanning process for the acquisition of the prepared dies, provided there is full optical access of the projected light to the object (Fig 3.4a).
Fig 3.4 (a) The object (model) is scanned in different positions. The object can be rotated and tilted on the holder while the light is projected. (b) The digital working model contains information from multiple scans. The individual files are differentially visualized after meshing to indicate this.
Acquisition and Processing of Scan Data
The developed point clouds produced by 3D scanners and their integrated software represent the digital conversion of the physical geometrical surface. The analog data of the surface is converted into digital data by image processing.4 These point clouds are stored in one or more files, mostly in proprietary (closed) formats prepared for the next data processing step of providing a unencrypted (open) file that can be transferred to the design software. In this data processing step, the software cleans and filters the points of the acquisition process, which can contain aberrant and highly concentrated points due to multiple, overlapping scan procedures. Software algorithms then clean and align the points of the captured physical surface. This is a critical phase and can lead to divergences from the reality of the teeth and soft tissue. The developed point cloud can be compiled by multiple scans at several resolutions. Thus, it is possible to rescan and exchange single data files, such as dies with high resolution or neighboring teeth with less concentrated points. Advanced systems also allow the cut-out of areas of the point cloud and enable a rescan and insert of the new situation.
This process produces a single file containing a list of the individual clouds and the coordinates to relate them to each other. In scan systems with an open-format output, the data can be transferred and interpreted by the following post-processing and design utilizing the unencrypted data. These data can be used directly for measurement and visualization in the design (CAD) and later in the production (CAM) processes (Table 3.1).
Table 3.1 Data acquisition and processing steps in the digital workflow to generate a data file for the design process
| Step | Action | Digital result and output |
| Acquisition | Scanning of surface with hardware | Virtual discrete geometrical data of multiple scans with multiple clouds |
| Processing | Software algorithms for treating point clouds | Virtual discrete geometrical data with one cloud |
| Post-processing | Software algorithms for meshing point clouds | Virtual continuous data with a CAD model |
While working in the data acquisition process, and later in the design modus, the operator does not visualize the point cloud, but rather a closed and connected surface between the points, on the screen. This surface does not provide the possibility to evaluate the distribution of the points. Measurement errors, such as holes caused by missing points, can be closed by the calculation process of the software. Thus, the scan holes – up to a preset degree – are automatically closed and therefore invisible to the user. The same phenomenon is present at the sharp edges and corners of a surface. To reproduce these fine surface characteristics of the object, it is necessary to have a very fine and precise distribution of the points and a customized algorithm for dental application in the processing phase of the point cloud. Thus, the accuracy of the scan and developed cloud depends on the hardware in combination with a system and dentistry-specific software.
Post-processing of Scan Data for CAD
The processed digitized data from the scan software is then transferred to the CAD software. Prior to the design phase, the digital geometrical data must be converted (post-processing) into a virtual representation with continuous geometrical data.4 This will provide a visualization of the scan data in one piece with a solid surface to commence the design process within the CAD software (Fig 3.4b).
This point-meshing process by the software is based on triangulation to automatically create a polygonal three-dimensional model (e.g., mesh model), which enables the design of a dental restoration on this surface. The polygonal representation of a curved shape is modeled as many small, faceted, flat surfaces. Reconstruction to a polygonal model involves finding and connecting adjacent points with straight lines in order to create a continuous surface.5 This surface model was first developed for the fabrication of solid parts by additive stereolithographic fabrication; this common file format is also referred to as Standard Tessellation Language (STL).
After defining the surface, the preparation line of the prepared teeth can be verified and marked as an outline for the restoration (design). Also, additional patient information (e.g., the jaw relationship, number of prepared teeth, restoration type, name, and tooth shade) can be added to the file. Transferring this complex information to the next step in the workflow (e.g., CAD), a referenced and calibrated digital interface is required to fully understand all relevant recipient details. Regarding this aspect, a closed (proprietary) data format from a single provider for all hardware and software components in the workflow is more maintenance-friendly compared to an open workflow involving several manufacturers. However, even if a proprietary file format is utilized, the meshing of points in clouds and their coordinates are represented as lists in the file. The advantage of a closed workflow is that the single provider will be responsible for the complete digital solution and often offers support at every step of the digital workflow. In an open workflow, the data is processed by different software and machines provided by different manufacturers. This can create problems in identifying the cause for possible errors and may raise compatibility issues. To overcome such issues in an open workflow, clinicians and technicians often rely on a “geek solution” rather than an open workflow.
A significant factor is the created size of the data set which is exported to the CAD program. In dental use, the exported surface file is not editable; only some products permit pulling and sculpturing the surface in order to manipulate the scan situation.
Desktop scanner accuracy can vary between devices. The ISO standard 12836:2015 describes procedures for the standardized comparison and evaluation of dental scanners.6 All available products are compliant with this standard, which is applied by the manufacturers. However, a general and independent evaluation and comparison of the accuracy of different desktop products is missing. Only a few studies cover this point of interest.3 According to the DIN ISO 5725 standard, accuracy is defined by precision and trueness. Reproducibility of results within a group is referred to as precision. Comparison with a reference scan is referred as trueness.7
Multiple factors can influence both aspects. In desktop scanning technology, the trueness to the original prepared tooth or a primary component (bar, double-crown) is significant. By applying optical-stripe light projection, reflection and intrusion of the light onto and into the surface must be eliminated. To accomplish this, the object to be scanned is either made from a nonreflective material, or its surface is conditioned and coated by a scan powder. The first option is favorable because the powdering procedure can be error-prone, leading to measurement mistakes. In addition, all results that are based on an impression and accompanying stone cast are susceptible to the inaccuracy inherent to the conventional workflow.
Using a striped light scanner (D250, 3Shape A/S, Copenhagen, Denmark) without powder, Flügge et al. evaluated the precision of virtual models rendered from repeated cast scanning.8 The average deviation between the virtual models was 10 µm (median 6 µm) with a maximum of 460 µm. Lee et al. compared the trueness of scans (D800, 3Shape A/S) with and without powder (VITA CEREC, VITA, Bad Säckingen, Germany) to stone replicas (master die); the group without powder produced fewer dimensional changes in the coordinates.9 The total deviation values of scans with and without powder relative to the master die were 8.65 and 7.10 µm, respectively. In the same study, the precision of a contact probe scanner (Incise, Renishaw, Wotton-under-Edge, Gloucestershire, UK) and an optical scanner (D800, 3Shape A/S) was evaluated. The contact probe scanner produced fewer measurement errors and limits of agreement for all x-, y-, and z-coordinates relative to the optical scanner, which may be due to the influence of the relatively larger standard deviation (SD) of the optical scanner’s measurement errors and limit of agreements. The total root mean square (SD) of the discrepancy was 1.58 (0.43) and 1.97 (0.12) µm for the master die and stone replica.9 These findings with the latest equipment confirm the results of previous studies using older hardware and software.2,8,10,11 However, in a study evaluating the precision of complete arch scans, DeLong achieved results of 30 µm with an optical scanner (Carl Zeiss Optotechnik, Neubeuern, Germany).12
Beside the technology used and the specimen’s surface condition, the quality of the mathematical triangulation is also important. The manner in which the triangles describe the surface, particularly sharp etches and corners, is quite challenging and requires high-resolution data. This generates, in turn, large files and requires processors with high graphics capacity. The scanning software, the capacity of the computer, and the necessary accuracy need to be balanced to make a system practical and economical (Fig 3.5). To compare clinical outcomes with respect to precision, an in vitro study was performed and the marginal adaptation of three-unit zirconia fixed dental prostheses (FDPs) obtained from intraoral digital scanners (Lava True Definition, Cadent iTero), scanning of a conventional silicone impression, and scan of a master cast with an extraoral scanner (3Shape lab scanner) was evaluated. From most to least precise: the Lava True Definition scanner showed the lowest SD gap among all groups with 26.6 μm (4.7); the master cast scan 50.2 μm (6.1); Cadent iTero scan 62.4 μm (5.0); the group of silicone impression scans produced the largest recorded mean gap with 81.4 μm (6.8). However, all groups were within the limit of clinical acceptance on the basis of a 120 μm gap.13
Fig 3.5 (a) Digital model after post-processing and meshing of the point cloud image of a model with a polygon-based solid surface. (b) Cloud of scan points with 8 scan positions at reduced resolution. (c) Cloud of scan points with 8 scan positions at high resolution. (d) Cloud of scan points and wireframe with 8 scan positions at high resolution. (e) Cloud of scan points and wireframe with 16 scan positions at reduced resolution. (f) Cloud of scan points and wireframe with 16 scan positions at high resolution. (g) Cloud of scan points and wireframe with 16 scan positions at high resolution.
Although optical-stripe light projection is the most common technology used in dental lab scanning today, laser scanning has a long tradition in scanning dental casts and impressions. These devices have been used to capture the surface topography of dental casts and alginate impressions in the orthodontic field. A laser beam scanner (R 700, 3Shape) with two cameras and three axes introduced for this application was evaluated by Lemos et al.14 The authors described an applicability of the scanner for performing measurements and analyses in orthodontics. The findings are comparable with other studies in this field14; the same application of scanning impressions was evaluated by Kim et al.1 Scans of rubber (A-silicone) and alginate impressions were performed by a currently available device (7 series, Dental Wings, Montreal, Canada). The point clouds of the impression groups were compared with the data of the master dies. The rubber material revealed –3.0 ± 3.2 µm and the alginate 9.8 ± 6.9 µm in difference.
Desktop scanners are offered in a wide range of products regarding hard- and software, mostly included in a specific business model by the provider. The decision for certain equipment should be based on the individual needs of the used fields. With increased scanner capability, the necessary times for scan and calculation increases too and can be an issue when operating. Flexible scan software should allow adaptation for individual use of the multiple indications. Some scanners are already set up with the possibility to easily link with additional information, such as individual 3D articulation, facial 3D scans, and 3D diagnostic information. When purchasing a product, the respective software version should be up to date, as should the computer hardware.
1.Kim S-R, Lee W-S, Kim W-C, Kim H-Y, Kim J-H. Digitization of dental alginate impression: three-dimensional evaluation of point cloud. Dent Mater J 2015;34:835–840.
2.Persson M, Andersson M, Bergman B. The accuracy of a high-precision digitizer for CAD/CAM of crowns. J Prosthet Dent 1995;74:223–229.
3.Alghazzawi T. Advancements in CAD/CAM technology: options for practical implementation. J Prosthodont Res 2016;60:72–84.
4.Tapie L, Lebon N, Mawussi B, Fron Chabouis H, Duret F, Attal JP. Understanding dental CAD/CAM for restorations – the digital workflow from a mechanical engineering viewpoint. Int J Comput Dent 2015;18:21–44.
5.Wikipedia: 3D scanning. Available at: https://enwikipediaorg/wiki/3D_scanning [last accessed November 5, 2018].
6.Standard ISO 12836:2015. Dentistry – digitizing devices for CAD/CAM systems for indirect dental restorations – test methods for assessing accuracy. Berlin, Germany: Beuth Publishing, 2015.
7.Standard DIN ISO 5725-1 (1997-11). Accuracy (trueness and precision) of measurement methods and results – part 1: general principles and definitions (ISO 5725-1:1994). Berlin, Germany: Beuth Publishing, 1997.
8.Flügge TV, Nelson K, Nahles S, Metzger MC. Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a model scanner. Am Orthod Dentofacial Orthop 2013;144:471–478.
9.Lee W, Park J, Kim J, Kim H, Kim W, Yu C. New approach to accuracy verification of 3D surface models: an analysis of point cloud coordinates. J Prosthodont Res 2016;60:98–105.
10.Persson A, Andersson M, Oden A, Sandborgh-Englund G. A three-dimensional evaluation of a laser scanner and a touch-probe scanner. J Prosthet Dent 2006;95:194–200.
11.Persson AS, Andersson M, Oden A, Sandborgh-Englund G. Computer aided analysis of digitized dental stone replicas by dental CAD/CAM technology. Dent Mater 2008;24:1123–1130.
12.DeLong R, Heinzen M, Hodges JS, Ko CC, Douglas WH. Accuracy of a system for creating 3D computer models of dental arches. J Dent Res 2003;82:438–442.
13.Shembesh M, Ali A, Finkelman M, Weber H, Zandparsa R. An in vitro comparison of the marginal adaptation accuracy of CAD/CAM restorations using different impression systems. J Prosthodont 2017;26:581–586.
14.Lemos LS, Rebello IM, Vogel CJ, Barbosa MC. Reliability of measurements made on scanned cast models using the 3 Shape R 700 scanner. Dentomaxillofac Radiol 2015;44:20140337.
