Resins, Synthetic

This study involved the following figure 1 processes. Frasaco's denture model (ANA-4, Frasaco, Tettnang, Germany) was used in the current study. After a negative mold was fabricated using dental silicone (Deguform, Degudent GmbH, Hanau-Wolfgang, Germany) on a synthetic resin denture model, a model was cast using hard plaster (Fujirock EP, GC Europe NV, Leuven, Belgium).
The reference plaster model was scanned using an intraoral scanner Trios (3Shape, Copenhagen, Denmark). The resulting scan data were designated as the CAD reference model (CRM). Trios is a confocal scanner with a real time rendering mode, which allows the practitioner to scan the target area while viewing it on the screen. We chose to use the intraoral scanner in this study to mimic an actual clinical setting. When scanning was completed, a reference STL file of the 3D shape of the plaster model was created (Fig. 2).
Based upon the reference STL file, a total of 10 milling models were manufactured using the milling equipment (ARUM 5X-200, Doowon, Korea). Polymethyl methacrylate (PMMA) blocks (Yamahachi PMMA Disk, Yamahachi Dental MFG, Aichi-Pref, Japan) were used as the material for the models. Burs with a maximum diameter of 2.5 mm and a minimum diameter of 1 mm were used. To maintain the same condition during milling, a single set of milling burs were used for a single block. Next, an additive manufacturing 3D printer (ZENITH, Dentis, Korea) was used to manufacture another 10 models, with the same condition, using a 16 µm layer. Models manufactured by milling were classified as group A, while those manufactured by 3D printing were group B. Each model was numbered in their respective group (i.e., A1–A10 and B1–B10) (Fig. 3).
The 20 models manufactured by milling and 3D printing were scanned using a desktop scanner (Ceramill MAP 400, Amann Girrbach, Austria). Data of these scanned models were saved as test STL files (Fig. 4). A desktop scanner, rather than an intraoral scanner, was used in this case, since the former generated fewer errors during scanning, compared with those generated with the intraoral scanner, which correlated with the skill level and technique of the practitioner.
When the scan was completed, reference STL files were designated as a control group, while the test STL files were the test group. Each test STL file was superimposed on the reference STL file using specialized software (Geomagic Control X, 2017.0.3, 3D Systems, Cary, NC, USA). For superimposition, the test STL file was converted into point cloud data. Then, the CAD-reference-model (CRM), surface date, CAD-test-model (CTM), and the point cloud data, were initially aligned and subsequently rearranged to the best fit alignment. Finally, point cloud data was projected onto the surface of the CRM data. The sampling rate was set at 100%, with a maximum repetition index of 30. The distances between surface data and all points were converted to root mean square (RMS) values. The RMS is a general method to assess the mean value of errors, by directly comparing two data groups with an identical coordinate system. The accuracy of a corresponding data group can be calculated using a single scale. A higher calculated RMS value indicated a large error, i.e., the difference in the attributes between reference and measurement data. The RMS is typically used as a criterion to measure the similarity of two sets of N-dimensional vector sets after optimal superimposition. The equation used for the RMS calculation is as follows12 (link):
$\mathrm{R}\mathrm{M}\mathrm{S}=\frac{1}{\sqrt{n}}·\sqrt{\sum _{i=1}^n{(x_{1,i}-x_{2,i})}^2}$
Here, χ_1,i is the data point of the CRM, and χ_2,i is that of the CTM; and N is the number of all measurement points.
Unnecessary and inaccurate parts of the 3D shape data of all models were eliminated.13 The superimposition results were illustrated as a color difference map (Fig. 5). The maximum and minimum values of the color difference map were +100 µm and −100 µm, respectively. The effective range indicated in green was set from −30 µm to +30 µm. Although these values were not acceptable for prosthesis restoration, they were set to allow the easy comparison of the accuracy of the models produced using the milling and 3D printing methods. In addition, the two models were measured by using the fixed measuring points (e.g. cusp tip and fossa) in order to evaluate the clinical correlation of the discrepancies observed in between the models through RMS methods.14 (link) The 42 assigned points were located as follows: incisal angle of the incisors (4 points), the tip of cusp of the canines (1 point), the cusp of the molars (10 points), lingual fossa of the incisors (3 points), fossa(pit) of the molars (10 points), undercut in the facial aspect (7 points), and undercut in the lingual aspect (7 points) (Fig. 6). For the location of these points, the divergences in the x-, y-, and z-axis to each reference and the test data were measured.15 (link)
A Shapiro-Wilk test was initially performed before the comparison of the mean values between the reference STL files and the test STL files of the scanned samples from each group. A Mann Whitney U test was conducted to determine significant difference between the groups. All statistical process and analysis were performed using IBM SPSS Statistics 23 (SPSS Inc., Chicago, IL, USA). The significance level was set at P <.05.

All peptides were
synthesized by
using the standard Fmoc solid-phase synthesis strategy on MBHA resins.
The conjugated peptides LK-LE were synthesized according to our previous
method.^{37 (link)} The crude peptides were purified
and identified by reversed-phase high-performance liquid chromatography
(RP-HPLC) and electrospray ionization mass spectrometry (ESI-MS),
respectively. The purity of the peptides was above 95% confirmed by
HPLC.