Background: Dental plaque causes many common oral diseases (e.g., caries, gingivitis, and periodontitis). Therefore, plaque detection and control are extremely important for children’s oral health. The objectives of this study were to design a deep learning-based artificial intelligence (AI) model to detect plaque on primary teeth and to evaluate the diagnostic accuracy of the model. Methods: A conventional neural network (CNN) framework was adopted, and 886 intraoral photos of primary teeth were used for training. To validate clinical feasibility, 98 intraoral photos of primary teeth were assessed by the AI model. Additionally, tooth photos were acquired using a digital camera. One experienced pediatric dentist examined the photos and marked the regions containing plaque. Then, a plaque-disclosing agent was applied, and the areas with plaque were identified. After 1 week, the dentist drew the plaque area on the 98 photos taken by the digital camera again to evaluate the consistency of manual diagnosis. Additionally, 102 intraoral photos of primary teeth were marked to denote the plaque areas obtained by the AI model and the dentist to evaluate the diagnostic capacity of each approach based on lower-resolution photos. The mean intersection-over-union (MIoU) metric was employed to indicate detection accuracy. Results: The MIoU for detecting plaque on the tested tooth photos was 0.726 ± 0.165. The dentist’s MIoU was 0.695 ± 0.269 when first diagnosing the 98 photos taken by the digital camera and 0.689 ± 0.253 after 1 week. Compared to the dentist, the AI model demonstrated a higher MIoU (0.736 ± 0.174), and the results did not change after 1 week. When the dentist and the AI model assessed the 102 intraoral photos, the MIoU was 0.652 ± 0.195 for the dentist and 0.724 ± 0.159 for the model. The results of a paired t-test found no significant difference between the AI model and human specialist (P > .05) in diagnosing dental plaque on primary teeth. Conclusions: The AI model showed clinically acceptable performance in detecting dental plaque on primary teeth compared with an experienced pediatric dentist. This finding illustrates the potential of such AI technology to help improve pediatric oral health.

The objective was to establish and evaluate a method for manufacture of custom trays for edentulous jaws using computer aided design and fused deposition modeling (FDM) technologies. A digital method for design the custom trays for edentulous jaws was established. The tissue surface data of ten standard mandibular edentulous plaster models, which was used to design the digital custom tray in a reverse engineering software, were obtained using a 3D scanner. The designed tray was printed by a 3D FDM printing device. Another ten hand-made custom trays were produced as control. The 3-dimentional surface data of models and custom trays was scanned to evaluate the accuracy of reserved impression space, while the difference between digitally made trays and hand-made trays were analyzed. The digitally made custom trays achieved a good matching with the mandibular model, showing higher accuracy than the hand-made ones. There was no significant difference of the reserved space between different models and its matched digitally made trays. With 3D scanning, CAD and FDM technology, an efficient method of custom tray production was established, which achieved a high reproducibility and accuracy.

To design a hydrogel material containing nano hydroxyapatite particles for three-dimensional (3D) bio-printing of human adipose-derived stem cells (hASCs) and to explore whether nano hydroxyapatite particles can promote osteogenic differentiation of a 3D bio-printing construct consisting of hASCs in vivo and in vitro. A 3D reticular printing structure was designed. Sodium alginate/gelatin/hASCs (AG group) was considered as the control group, and sodium alginate/gelatin/nano hydroxyapatite/hASCs (AGH group) was considered as the experimental group. Immunofluorescence microscopy was used to observe the cell viability and cell adhesion, and cell proliferation was analyzed by comparison of viable cell numbers in printed constructs at 1 day and 7 days after printing. After 14 days of osteogenic induction for the AG group and AGH group, real-time quantitative PCR and immunofluorescence were used to analyse the expression of the osteogenesis-related genes Runt-related transcription factor 2 (RUNX2), osterix (OSX), and osteocalcin (OCN). New bone formation in printed constructs was observed using micro-CT, HE staining, Masson trichrome staining, and OCN immunohistochemical staining 8 weeks after being implanted. The cells in the AG group and AGH group were evenly distributed in the 3D printed constructs. The number of viable cells and cell viability both in the AG group and AGH group at 7 days after printing were higher than those at 1 day after printing (p < 0.05); however, the difference between the AG group and AGH group was not significant. At 14 days after osteogenic induction in vitro, real-time PCR results showed that the expression of osteogenesis-related genes in the AGH OM group was significantly higher than that in the AGH PM group, AG PM group, and AG OM group (p < 0.05). At 8 weeks after bio-printed construct implantation, the results of micro-CT, HE staining, Masson trichrome staining, and OCN immunohistochemical staining showed that the new bone formation in the AGH group was higher than that in the AG group (p < 0.05). The in vivo and in vitro results demonstrated that nano hydroxyapatite particles dispersed in a sodium alginate/gelatin matrix could promote osteogenic differentiation of hASCs in a 3D bio-printed construct, and this scaffold material could be considered to repair large bone tissue defects.

Here, we aimed to investigate osteogenic differentiation of human adipose-derived stem cells (hASCs) in three-dimensional (3D) bioprinted tissue constructs in vitro and in vivo. A 3D Bio-plotter dispensing system was used for building 3D constructs. Cell viability was determined using live/dead cell staining. After 7 and 14 days of culture, real-time quantitative polymerase chain reaction (PCR) was performed to analyze the expression of osteogenesis-related genes (RUNX2, OSX, and OCN). Western blotting for RUNX2 and immunofluorescent staining for OCN and RUNX2 were also performed. At 8 weeks after surgery, osteoids secreted by osteogenically differentiated cells were assessed by hematoxylin-eosin (H&E) staining, Masson trichrome staining, and OCN immunohistochemical staining. Results from live/dead cell staining showed that most of the cells remained alive, with a cell viability of 89%, on day 1 after printing. In vitro osteogenic induction of the 3D construct showed that the expression levels of RUNX2, OSX, and OCN were significantly increased on days 7 and 14 after printing in cells cultured in osteogenic medium (OM) compared with that in normal proliferation medium (PM). Fluorescence microscopy and western blotting showed that the expression of osteogenesis-related proteins was significantly higher in cells cultured in OM than in cells cultured in PM. In vivo studies demonstrated obvious bone matrix formation in the 3D bioprinted constructs. These results indicated that 3D bioprinted constructs consisting of hASCs had the ability to promote mineralized matrix formation and that hASCs could be used in 3D bioprinted constructs for the repair of large bone tissue defects.

To quantitatively evaluate maxillary complete dentures fabricated from polylactic acid (PLA) using fused deposition modelling (FDM) technology. A digital maxillary complete denture was prepared based on a standard maxillary edentulous plaster model. The PLA pattern was printed by a FDM machine, with 5 repetitions, while another 5 wax patterns were printed as control group, using a high accuracy three-dimensional (3D) wax printer. The patterns were scanned with a 3D scanner. A light-body silicone film was made after each denture pattern had been totally seated on the plaster model, and was scanned to determine its thickness, which reflected the 3D space between the plaster model and the tissue surface of the denture pattern. The overall area was separated into four parts: primary stress-bearing area, secondary stress-bearing area, border seal area and relief area, and the average deviation of these four parts were measured. The values were analyzed by independent t-test. The overall mean value and standard deviation of space between PLA denture patterns and plaster model was 0.277 ± 0.021 mm, while that of the wax denture patterns was 0.279 ± 0.045 mm, which showed a good fit overall. No statistically significant (𝑃 > 0.05) difference was observed between the PLA patterns and wax patterns. The adaptation of the PLA pattern of maxillary complete denture printed by FDM technology is comparable to that prepared by wax printer, and can satisfy the accuracy requirements.