Search

Abstract Background The success of cephalometric analysis depends on the accurate detection of cephalometric landmarks on scanned lateral cephalograms. However, manual cephalometric analysis is time-consuming and can cause inter- and intra-observer variability. The purpose of this study was to automatically detect cephalometric landmarks on scanned lateral cephalograms with low contrast and resolution using an attention-based stacked regression network (Ceph-Net). Methods The main body of Ceph-Net compromised stacked fully convolutional networks (FCN) which progressively refined the detection of cephalometric landmarks on each FCN. By embedding dual attention and multi-path convolution modules in Ceph-Net, the network learned local and global context and semantic relationships between cephalometric landmarks. Additionally, the intermediate deep supervision in each FCN further boosted the training stability and the detection performance of cephalometric landmarks. Results Ceph-Net showed a superior detection performance in mean radial error and successful detection rate, including accuracy improvements in cephalometric landmark detection located in low-contrast soft tissues compared with other detection networks. Moreover, Ceph-Net presented superior detection performance on the test dataset split by age from 8 to 16 years old. Conclusions Ceph-Net demonstrated an automatic and superior detection of cephalometric landmarks by successfully learning local and global context and semantic relationships between cephalometric landmarks in scanned lateral cephalograms with low contrast and resolutions.

Abstract Background Among interdental cleaning aids (ICAs), interdental brushes (IDBs) are in the spotlight because they can effectively remove plaque from interdental surfaces. Guidance on the correct use of ICAs, such as IDBs, is required to prevent dental plaque accumulation. Since it is impossible to confirm the interdental proximal surface unless extracted, it is difficult to conduct quantitative experiments. This study presented an efficient way to evaluate IDBs by realizing dental structures and embrasures using a Dental computer-aided design (CAD) software and a 3D printer. Methods Two different sizes of embrasure (0.7 and 1.2 mm) crown models were prepared with CAD software and a 3D printer. To evaluate the cleaning efficacy of IDBs of each size (0.6, 0.7, 0.8, 1.0, 1.2, and 1.5 mm diameters), the 9th cycle of brush move was performed where artificial plaque was spread and a digital camera was used to record the process. The pixels and percentage of cleaning from the recorded digital images were analyzed. Results The plateau was formed after the 5th brushing cycle under all conditions—after the 5th cycle, the cleaning efficacy of the two crown models was 69.3–86.4% and 49.8–75.4%. In these results, the optimal diameters for the IDB were 1.2 and 1.5 mm for embrasure sizes of 0.7 and 1.2 mm, respectively. Moreover, the cleaning efficacy was the highest at 86.4% and 75.4% after the 9th cycle. Conclusions The 3D-printed model base for the human oral embrasure structure is an adequate model to test artificial plaque removal using IDB. The use of IDBs for more than five cycles does not support the conventional idea that a greater number of IDB brushing moves is more effective in a statistically substantial manner.