Page 397 - IJB-10-3
P. 397

International Journal of Bioprinting                                    Optimizing 3D-printed mouthguards





























            Figure 11. Relationships between Shore A hardness of soft materials and retention force in double-layer additively manufactured MG samples. As the
            Shore A hardness increased, the retention force of MG samples increased.

            Table 5. Root-mean-square deviation of various MG samples in each fatigue test stage
             Pre-mid           Retention force (N)       95% CI for mean               F-value     P-value
             ERK MGs           200.7 ± 98.5 a            (-44.07,445.41)
             MG21 MGs          140.3 ± 14.2 a            (105.00,175.67)
             D-A30 MGs         117.3 ± 23.1 a            (59.91, 174.76)               2.078       0.159
             D-A50 MGs         119.3 ± 12.7 a            (87.88, 150.79)
             D-A70 MGs         102.0 ± 13.9 a            (67.58, 136.42)
             Mid-post          Retention force (N)       95% CI for mean               F-value     P-value
             ERK MGs           179.3 ± 131.5 a           (-144.03, 502.69)
             MG21 MGs          124.7 ± 24.9 b            (62.80,186.54)
             D-A30 MGs         97.0 ±  18.2 a            (51.81, 142.19)               1.029       0.438
             D-A50 MGs         120.0 ± 9.5 a             (96.30, 143.70)
             D-A70 MGs         89.0 ± 16.5 a             (48.11, 129.89)
            Abbreviation: A, single-layer Agilus or/and Vero composites MG samples; CI, confidence interval; D-A, double-layer MG samples without layer D-ABS,
            inlayer Agilus, or/and Vero composites; ERK, Erkoflex.

            Table 6. Root-mean-square difference in each sample between the pre-mid and pre-post experimental stage
             Samples               Mean and 95% CI                           T-value         P-value
             ERK MGs               75.00 (-153.53, 303.52)                   1.412           0.293
             MG21 MGs              50.33 (27.25, 73.42)                      9.383           0.011
             D-A30 MGs             -2.67 (-22.90,17.57)                      -0.567          0.628
             D-A50 MGs             -10.33 (-37.79,17.03)                     -1.625          0.246
             D-A70 MGs             17.67 (-23.18, 58.51)                     1.861           0.204
            Abbreviation: A, single-layer Agilus or/and Vero composites MG samples; CI, confidence interval; D-A, double-layer MG samples without layer D-ABS,
            inlayer Agilus, or/and Vero composites; ERK, Erkoflex.

                               35
            the lack of FDM process.  In addition, it conducted a series   Recently, Unkovskiy et al. presented the workflow of
                                                                                                  36
            of physical performance tests on the MG materials and   a multi-layer customized sports MG design.  Despite the
            samples in a simulated oral environment.           well-designed shape, its crucial physical property has not

            Volume 10 Issue 3 (2024)                       389                                doi: 10.36922/ijb.2469
   392   393   394   395   396   397   398   399   400   401   402