Evaluation of innovative techniques for dental crowns manufacturing.

Salmi, Alessandro ; Atzeni, Eleonora ; Iuliano, Luca 等

1. INTRODUCTION

Recently, Reverse Engineering (RE) and Additive Manufacturing (AM) have proven to be interesting techniques alternative to traditional procedures in dental applications, such as crown manufacturing. The geometry of a stone replica can be digitized by a three-dimensional scanner. From the points clouds a Virtual Model (VM) is generated and tooth caps could be designed on it and manufactured by Computer Aided Manufacturing (CAM) or employing AM techniques.

The sequence from replica digitization (VM) to cap fabrication involves many production steps, each one inevitably inducing dimensional errors. However, the information given by systems suppliers or present in literature is lacking and does not allow comparison among different manufacturing systems. Therefore, it becomes important to investigate new technologies to quantify their accuracy and precision. Currently, there is not a standard methodology for evaluating the error introduced by each step of the manufacturing cycle, from the scanning of the replica to the final dental device. Moreover, to have comparable data, a suitable benchmark must be defined (Atzeni et al., 2009). In literature a few benchmarks based on standard geometries were proposed for dental applications (Brosky et al., 2002; DeLong et al., 2003), but they appear lacking because differences from posterior and anterior teeth shapes, tooth position along the dental arch, and three dimensional arrangement of teeth are disregarded.

In this work a standard procedure for evaluating the accuracy at each step of the manufacturing sequence for dental caps is proposed. This analysis is based on the definition of an innovative benchmark. Deviations introduced by benchmark manufacturing, scanning operation, cap design, and fabrication are studied.

2. MATERIALS AND METHODS

The innovative benchmark, illustrated in Figure 1, is specifically designed through classical features resembling real prepared teeth. By this way shapes inspection and reproduction is made easier, thus allowing the comparison among different techniques. The benchmark includes tooth orientation and provides oblique surfaces similar to those of a real prepared tooth: this is a significant innovation. The starting point for benchmark geometry definition is the analysis of physical artificial teeth used for partial (or complete) denture fabrication. Three different teeth are selected as reference: lateral incisor (32 and 42), second premolar (35 and 45), and second molar (37 and 47). Prepared teeth shapes and dimensions are identified by simulating the preparation of the corresponding artificial tooth by considering the following rules related to prosthodontics (Goodacre et al., 2001):

* the tooth reduction is set to 1 mm, common average value for metal-ceramic restorations;

* a tapering of 5 degrees is made around the circumference of the prepared tooth, appropriate to allow crown fitting while providing enough grip;

* the tooth is prepared with a chamfer finish line.

The first step of the manufacturing sequence of crown restorations consists in the achievement of the virtual model from the stone replica and errors in this phase will highly affect next prosthesis design and fabrication. Thus, the benchmark is first produced by selective laser sintering (SLS) of polyamide and then inspected and digitized to obtain its virtual model. The inspection of the physical benchmark with a coordinate measuring machine (CMM) is required to evaluate the deviation contribute of the scanning operation. In fact, it is not possible to compare the point-cloud directly to the original 3D CAD model, because results will include errors from the benchmark fabrication. In this way, a Reference Model (ReM), consisting of actual dimensions of the benchmark, is yielded for next comparisons. The point-cloud from digitization process is inspected as well by using a specific software to evaluate the same dimensions. The comparison of results from both inspections allows to identify the accuracy of the digitization step. From the VM, caps are designed disregarding the manufacturing process and installation (i.e. gap for cement is set to zero, while actually it is variable as a function as the fabrication technique). By this way, caps cannot fit the original studs, but the use of a unique geometry makes possible to compare different processes. Compared techniques are milling of Zirconia, selective laser sintering (SLS) of a dental CrCo-alloy, and investment casting of CrCo-alloy from Drop-on-Demand (DoD) wax patterns. The coupling surface of each manufactured cap is inspected and compared with its CAD model to give information about fabrication processes accuracy.

3. RESULTS AND DISCUSSION

The results of the CMM inspection of the physical benchmark are listed in Table 1, where the average error (Av.) and standard deviation (SD) are detailed for dimensions grouped into basic sizes (0/3, 3/6, and 6/10 mm) accordingly to ISO 286-1 (1988).

[FIGURE 1 OMITTED]

Results confirm that the discrepancies of the physical benchmark with respect to its original three-dimensional CAD model meet common SLS tolerances for plastic parts (Silva et al., 2008). In fact, the obtained deviation of about 0.2 mm is in agreement with the declared laser spot size of about 0.4 mm and the accuracy of laser positioning of about 50 [micro]m. The average error of 0.12 mm obtained for basic sizes 6/10 mm, that is smaller than the error for basic sizes 0/3 mm and 3/6 mm, is justified by the fact that the measured dimensions belonging to this group are along the building direction (Z-axis) of the SLS machine. The precision of the SLS system along Z direction is very high because it depends on the (high) precision of the building platform elevator that is an electro-mechanical device, while the influence of the laser system is very low.

The outputs of the digitization are imported into Rapidform software to inspect the same entities previously analyzed. The comparison between scan data and CMM measurements in terms of dimensional deviations (Table 1) shows that the point-cloud is an accurate description of the real benchmark, with errors ranging from minus 0.02 to 0.07 mm. The accuracy of the digitization accomplished on the benchmark is good and adequate for medical purposes. Bigger deviations are found on the smaller dimensions (basic size 0/3 mm), among which there are the dimensions of the incisor. The deviations range is consistent with the declared accuracy of the scanning device (DentalWings). The rough aspect of the benchmark surface, amplified by the digitization (noise), leads on the Virtual Model to standard deviations (SD) of the same order of magnitude of the error. After this validation, the Virtual Model is proven to be a valid replica and consequently it becomes the base for techniques evaluation.

Caps are designed using a DentalWings proprietary application that full integrates the scanning environment and the "Crown & Bridge" design module, by setting parameters as listed in Table 2. Caps produced by the three selected techniques are shown in Figure 2. Results of the inspection of the inner surfaces of the caps are listed in Table 3, where the deviations are clustered for dimension groups with respect to the Virtual Model. In fact, because of the null gap, the inner surface of the cap copies exacltly the outer surface of the stud of the virtual model. By this way, the measured deviation, only due to the manufacturing process, give information on the ability of the process to reproduct the designed geometry. This indication is of paramount impontance for the definition of the ideal gap as a function of the process.

Results show that the accuracy of caps of CrCo-alloy, produced by SLS and by investment casting from DoD wax pattern, is comparable, the absolute deviation ranging from 0.02 mm to 0.19 mm. Analysing data, a scale effect could be appreciated. Moreover, most deviations are positive in sign, meaning that the produced cap is larger than the corresponding stud. This is an interesting result that could allow the installation of the cap instead of the null gap. Worst results are obtained for zirconia milled caps. It could be observed that among the studied techniques, the sintered caps could be preferred in terms of lead time.

[FIGURE 2 OMITTED]

4. CONCLUSIONS

An evaluation procedure based on an innovative benchmark is defined to assess errors introduced by each step of up-to-date dental restorations production methods. Two main outcomes are obtained: the availability of a validated virtual model of the benchmark, and accuracy information about the three most widely used cap fabrication techniques.

5. ACKNOWLEDGEMENTS

The authors are grateful to Dr. A. Lazzaro (WisilDent S.r.l., Torino, Italy), Mr. A. Sandi (3Dfast S.r.l. Padova, Italy), and Mr. G. Chiauzzi (Dentalabor S.r.l., Torino, Italy) for supporting the experimental part of the research.

6. REFERENCES

Brosky, M.E.; Pesun, I.J.; Lowder, P.D.; DeLong, R. & Hodges, J.S. (2002). Laser digitization of casts to determine the effect of tray selection and cast formation technique on accuracy. J Prosthet Dent, Vol. 87, pp. 204-209

DeLong, R.; Heinzen, M.; Hodges, J.S. & Douglas, W.H. (2003). Accuracy of a system for creating 3D computer models of dental arches. J Dent Res, Vol. 82, pp-438-442

Goodacre, C.J.; Campagni, W.V. & Aquilino, S.A. (2001). Tooth preparations for complete crowns: An art form based on scientific principles. J Prosthet Dent, Vol. 85, pp. 363-376

ISO 286-1: 1988 ISO system of limits and fits--Part 1: Bases of tolerances, deviations and fits

Atzeni, E.; Gatto, A.; Iuliano, L.; Minetola, P. & Salmi, A. (2009). A benchmark for accuracy evaluation of dental crowns up-to-date manufacturing, In: Innovative Developments in Design and Manufacturing, Bartolo, P.J. et al. (Eds.), CRC Press, London (UK)

Silva, D.N.; Gerhardt de Oliveira, M.; Meurer, E.; Meurer, M.I.; Lopes da Silva, J.V. & Santa-Barbara, A.; (2008). Dimensional error in selective laser sintering and 3D-printing of models for craniomaxillary anatomy reconstruction. J Craniomaxillofac Surg, Vol. 36, pp. 443-449

Tab. 1. Benchmark: dimensional deviations for basic sizes

 Fabrication Digitization
 (Physical Model) (Virtual Model)

Basic sizes Dimensional deviation (mm)
(mm) Av. SD Av. SD

0 - 3 -0.19 0.04 0.07 0.03
3 - 6 -0.20 0.07 0.01 0.03
6 - 10 -0.12 0.04 -0.02 0.01

Tab. 2. Cap design parameters

Angle 65 deg
Cement gap 0 mm
Collar position 1.5 mm
Extra horizontal gap 0 mm
Extra vertical gap 0 mm
Margin thickness 0.2 mm
Minimum thickness 0.6 mm

Tab. 3. Caps: dimensional deviations for basic sizes

 CrCo
 (laser sintered)

Basic sizes Dimensional deviation(mm)

(mm) Av. SD

0 - 3 -0.02 0.01
3 - 6 0.02 0.09
6 - 10 0.19 0.04

 CrCo
 (investment cast)

Basic sizes Dimensional deviation(mm)

(mm) Av. SD

0 - 3 0.02 0.01
3 - 6 0.07 0.13
6 - 10 0.14 0.10

 Zirconia
 (milled)

Basic sizes Dimensional deviation(mm)

(mm) Av. SD

0 - 3 0.05 0.06
3 - 6 0.16 0.08
6 - 10 0.33 0.06