Automated identification of cephalometric landmarks: Part 2-Might it be better than human?

OBJECTIVES: To compare detection patterns of 80 cephalometric landmarks identified by an automated identification system (AI) based on a recently proposed deep-learning method, the You-Only-Look-Once version 3 (YOLOv3), with those identified by human examiners. MATERIALS AND METHODS: The YOLOv3 algorithm was implemented with custom modifications and trained on 1028 cephalograms. A total of 80 landmarks comprising two vertical reference points and 46 hard tissue and 32 soft tissue landmarks were identified. On the 283 test images, the same 80 landmarks were identified by AI and human examiners twice. Statistical analyses were conducted to detect whether any significant differences between AI and human examiners existed. Influence of image factors on those differences was also investigated. RESULTS: Upon repeated trials, AI always detected identical positions on each landmark, while the human intraexaminer variability of repeated manual detections demonstrated a detection error of 0.97 ± 1.03 mm. The mean detection error between AI and human was 1.46 ± 2.97 mm. The mean difference between human examiners was 1.50 ± 1.48 mm. In general, comparisons in the detection errors between AI and human examiners were less than 0.9 mm, which did not seem to be clinically significant. CONCLUSIONS: AI showed as accurate an identification of cephalometric landmarks as did human examiners. AI might be a viable option for repeatedly identifying multiple cephalometric landmarks.

Automated identification of cephalometric landmarks: Part 1-Comparisons between the latest deep-learning methods YOLOV3 and SSD.

OBJECTIVE: To compare the accuracy and computational efficiency of two of the latest deep-learning algorithms for automatic identification of cephalometric landmarks. MATERIALS AND METHODS: A total of 1028 cephalometric radiographic images were selected as learning data that trained You-Only-Look-Once version 3 (YOLOv3) and Single Shot Multibox Detector (SSD) methods. The number of target labeling was 80 landmarks. After the deep-learning process, the algorithms were tested using a new test data set composed of 283 images. Accuracy was determined by measuring the point-to-point error and success detection rate and was visualized by drawing scattergrams. The computational time of both algorithms was also recorded. RESULTS: The YOLOv3 algorithm outperformed SSD in accuracy for 38 of 80 landmarks. The other 42 of 80 landmarks did not show a statistically significant difference between YOLOv3 and SSD. Error plots of YOLOv3 showed not only a smaller error range but also a more isotropic tendency. The mean computational time spent per image was 0.05 seconds and 2.89 seconds for YOLOv3 and SSD, respectively. YOLOv3 showed approximately 5% higher accuracy compared with the top benchmarks in the literature. CONCLUSIONS: Between the two latest deep-learning methods applied, YOLOv3 seemed to be more promising as a fully automated cephalometric landmark identification system for use in clinical practice.

The effect of zirconia surface architecturing technique on the zirconia/veneer interfacial bond strength.

PURPOSE: The purpose of this study was to evaluate the effect of the zirconia surface architecturing technique (ZSAT) on the bond strength between veneering porcelain and zirconia ceramic. MATERIALS AND METHODS: 20 sintered zirconia ceramic specimens were used to determine the optimal surface treatment time, and were randomly divided into 4 groups based on treatment times of 0, 1, 2, and 3 hours. After etching with a special solution, the surface was observed under scanning electron microscope, and then the porcelain was veneered for scratch testing. Sixty 3 mol% yttria-stabilized tetragonal zirconia polycrystal ceramic blocks were used for tensile strength testing; 30 of these blocks were surface treated and the rest were not. Statistical analysis was performed using ANOVA, the Tukey post-hoc test, and independent t-test, and the level of significance was set at α=.05. RESULTS: The surface treatment of the zirconia using ZSAT increased the surface roughness, and tensile strength test results showed that the ZSAT group significantly increased the bond strength between zirconia and veneering porcelain compared to the untreated group (36 MPa vs. 30 MPa). Optimal etching time was determined to be 2 hours based on the scratch test results. CONCLUSION: ZSAT increases the surface roughness of zirconia, and this might contribute to the increased interfacial bond strength between zirconia and veneering porcelain.

Effects of surface treatments on the susceptibilities of low temperature degradation by autoclaving in zirconia.

The aim of this study was to investigate the effect of surface treatments on the aging susceptibilities by autoclaving in zirconia ceramics. Four commercially available tetragonal zirconia polycrystals and one zirconia-alumina composite were tested. Disk-shaped specimens were prepared and the grain sizes were analyzed using a scanning electron microscope and image analyzer. The specimens were divided into three groups based on surface treatments including heat treatment subsequent to mirror polishing, grinding, and sandblasting. Specimens in each group were autoclaved at 134°C for 1, 3, 5, 10, and 15 h. The phases of the specimens were analyzed using an X-ray diffractometer, and the relative amount of the monoclinic phase was calculated and analyzed using Student's t-test and Newman-Keuls multiple comparisons test. Single routine autoclave treatment for sterilization did not promote the phase transformation in zirconia. The phase transformations of all specimens by autoclaving were correlated with grain size, except for the zirconia-alumina composite. Grinding or sandblasting treatments gave rise to increased formation of the monoclinic phase, especially for the zirconia-alumina composite, which showed the highest fraction for the monoclinic phase. The effects of surface treatments on the aging susceptibilities by autoclaving were different in the experimental groups. It is notable that not all zirconia ceramics show similar phase transformation by autoclaving after surface treatments.