Page 392 - IJB-10-3
P. 392

International Journal of Bioprinting                                    Optimizing 3D-printed mouthguards


































                                            Figure 4. The setting of the cycle-loading fatigue test.




























                                              Figure 5. The setting of the retention force test.


            measurement. Calculations were performed automatically         X  = sqrt( (/ )1  n sumx )      (I)
                                                                                            n
                                                                                              2
            to determine the most appropriate scanning position. The        rms             i=1  i
            root-mean-square (RMS) differences of the model surfaces
            were calculated by 3D software (Artec STUDIO; 3DS Co.,   2.4. Statistical analysis
            Kanagawa, Japan) based on the RMS formula, representing   The normality of data distribution was assessed using the
            the average distances of the surface points at different   Shapiro–Wilk test. Mean ± standard deviation was used to
            stages where the straight lines crossed in two same position   represent average values and dispersion. One-way analysis
            point, respectively. 32,33  The formula for calculating the RMS   of variance (ANOVA) analyzed the shock absorption test
            is as follows:                                     data for MIF and MIF-t. The Fisher test with Bonferroni’s


            Volume 10 Issue 3 (2024)                       384                                doi: 10.36922/ijb.2469
   387   388   389   390   391   392   393   394   395   396   397