International Journal of Computerized Dentistry, 03/2024

Aim: Artificial intelligence (AI)-based systems are used in dentistry to ensure a more accurate and efficient diagnostic process. The objective of the present study was to evaluate the performance of a deep learning (DL) program for the detection and classification of dental structures and treatments on panoramic radiographs of pediatric patients. Materials and methods: In total, 4821 anonymized digital panoramic radiographs of children between 5 and 13 years of age were analyzed by YOLOv4, a CNN (Convolutional Neural Networks)-based object detection model. The ability to make a correct diagnosis was tested on samples from pediatric patients examined within the scope of the study. All statistical analyses were performed using SPSS version 26.0 software. Results: The YOLOv4 model diagnosed the primary teeth, permanent tooth germs, and brackets successfully, with high F1 scores of 0.95, 0.90, and 0.76, respectively. Although this model achieved promising results, there were certain limitations for some dental structures and treatments, including fillings, root canal treatments, and supernumerary teeth. The architecture of the present study achieved reliable results, with some specific limitations for detecting dental structures and treatments. Conclusion: The detection of certain dental structures and previous dental treatments on pediatric panoramic radiographs by using a DL-based approach may provide early diagnosis of some dental anomalies and help dental practitioners to find more accurate treatment options by saving time and effort. Keywords: artificial intelligence, deep learning, pediatric dentistry, panoramic radiography, oral diagnosis

Aim: The aim of the present study was to evaluate the segmentation accuracy of the dentition by testing four open-source semi-automatic software programs. Materials and methods: Twenty CBCT scans were selected to perform semi-automatic segmentation of the maxillary and mandibular dentition. The software programs tested were InVesalius, ITK-SNAP, 3D Slicer, and Seg3D. In addition, each tooth model was manually segmented using Mimics software; this was set as the gold standard (GS) reference of the investigation. A specific 3D imaging technology was used to perform the superimposition between the tooth models obtained with the semi-automatic software and the GS model as well as to perform the surface-to-surface matching analysis. The accuracy of semi-automatic segmentation was evaluated, calculating the volumetric mean differences (mean bias and limits of agreement) and the percentage of matching of the tooth models compared with the manual segmentation (GS). Qualitative assessments were performed using color-coded maps. All data were statistically analyzed to perform comparisons between the investigated software programs. Results: Statistically significant differences were found in the volumetric and matching percentage data (P 0.05). InVesalius was the most accurate software program for 3D rendering of the dentition, with a volumetric bias (Mimics software) ranging from 4.59 to 85.79 mm3, while ITK-SNAP showed the highest volumetric bias, ranging from 30.22 to 319.83 mm3. The mismatched area was mainly located at the radicular tooth region. The volumetric data showed excellent inter-software reliability, with coefficient values ranging from 0.951 to 0.997. Conclusion: Different semi-automatic software algorithms could generate different patterns of inaccuracy error in the segmentation of teeth. Keywords: 3D rendering, cone beam computed tomography (CBCT), digital dentistry, digital orthodontics, oral surgery, orthodontics, segmentation

Aim: New technologies such as tactile robots and artificial intelligence (AI) are finding their way into clinical practice in dentistry and could contribute to the improvement of oral health care in the future. The hypothesis of the present in vitro pilot study trial was that a collaborative tactile robot programmed by a dental student could remove interproximal artificial plaque as effectively as a human operator. Materials and methods: Model teeth were fully covered with artificial plaque and set into phantom jaws. First, a robot was programmed by a dental student to perform interproximal cleaning with an interproximal brush. Second, the teeth were covered with artificial plaque again, and the dental student performed the interproximal cleaning manually. Both experiments were repeated five times. Residual plaque was measured with binary pictures. Surface coverage was reported, and the comparison of methods was performed with significance defined at α = 0.05. Results: No statistically significant difference was found in the cleaning result between the robot and the human operator. Conclusion: The results of the present in vitro pilot study indicate that a tactile robot with integrated AI programmed by a dental student can perform interproximal cleaning as effectively as a dental student. Practical implications: In the future, the use of robot assistants to support oral hygiene, eg, in patients with reduced motor skills or impaired vision, should be fur ther investigated. Keywords: dental robots, dentistry, dentronics, human–machine interaction, interdental brush, interdental cleaning, interproximal, oral hygiene, robotics

Aim: The objective of the present study was to determine the 3-year clinical and dimensional soft tissue change following implant placement in healed sites before and after loading with either customized or conventional healing abutments. Materials and methods: Premolar/molar implants were immediately loaded with either provisional and customized abutments without finishing lines according to the principles of the biologically oriented preparation technique (BOPT; test group) or conventional healing abutments (control group). Three months later, the definitive crowns were fabricated. Primary outcomes (soft tissue changes) and secondary outcomes (adverse events) were registered. Results: A total of 50 of the 87 subjects who were originally included were finally selected for the present retrospective analysis: 23 in the test group and 27 in the control group. During the first days after surgery, two adverse events of mucositis occurred, one in each group. A few technical complications were recorded such as the unscrewing of four screw-retained crowns. A significant increase in the alveolar width was observed in both groups (test: +2.5 ± 0.5 mm; control: +1.0 ± 0.9 mm). The widths appeared unchanged from 3 months to 3 years in both groups. There were no significant differences regarding the width of the keratinized mucosa measured at baseline and after follow-up. The Jemt papilla index showed a higher increase in the test group compared with the control group. Conclusions: Within the 3-year follow-up period, the peri-implant soft tissue outcomes of single, immediately loaded implants with customized healing abutments showed better results in terms of thickness and width compared with those of the conventional group. The number of side effects (mucositis and dehiscence) appeared to be very similar for both groups. In addition, customized healing abutments led to significant augmentation of the alveolar width, which was more than twice of that registered in the conventional group. Keywords: anatomical cap, customized abutment, dental abutment, dental implant, implant placement

Aim: The objective of the present study was to investigate the effect of using different materials for the fabrication of implant abutments and crowns on the mechanical behavior of implant-supported single crowns after artificial aging. The materials were tested in different combinations to reveal whether using stiff or resilient materials as an abutment or a crown material might influence the fracture strength of the whole structure. Materials and methods: A total of 40 implants (blueSKY) were restored with identical custom-made CAD/CAM abutments milled out of lithium disilicate or ceramic-reinforced polyetheretherketone (PEEK) and divided into five test groups (n = 8). Forty crowns made of three different materials (zirconia, lithium disilicate, and ceramic-reinforced PEEK) were used to restore the abutments. Specimens were subjected to mechanical loading up to 1,200,000 cycles in a chewing simulator (Kausimulator) with additional thermal cycling. The surviving specimens were subjected to quasi-static loading using a universal testing machine (Z010). Results: The PEEK abutments with zirconia crowns showed the highest median failure load (3890.5 N) while the PEEK abutments with lithium disilicate crowns exhibited the lowest (1920 N). Fracture and deformation occurred in both the crowns and abutments. Conclusion: The failure load of the restorations was influenced by the material of the abutment and crown. Restoring PEEK abutments with zirconia crowns showed a high failure load and no screw loosening. Keywords: CAD/CAM, custom-made implant abutment, implant abutment, lithium disilicate, PEEK, zirconia

Aim: The aim of the present in vitro study was to investigate the effects of different sintering procedures on the fit, color parameters, and fracture load of monolithic fixed partial prostheses (FPPs). Materials and methods: A metal master model was scanned and FPPs were designed. Groups were created by fabricating FPPs using four different sintering procedures (n = 10): Prettau-Standard (PST); Prettau-Slow (PSL); Ice-Speed (ISP); Ice-Standard (IST). PST-PSL (Group P; N = 20) and ISP-IST (Group I; N = 20) were colored with different coloring liquids. The marginal and internal fit were measured using the silicone replica method. CIELAB values of the samples were measured using a spectrophotometer. Then, for each sample, the die was obtained from polymethyl methacrylate. The specimens were cemented into dies and tested in a universal testing machine for fracture load. One-way ANOVA was performed to assess the effect of the sintering procedure on the marginal and internal fit; fracture load; and ∆E00, ∆L’, ∆C’, and ∆H’ values of the FPPs. Results: The PSL and PST groups showed significantly smaller internal and marginal fit values compared with the ISP group. Additionally, the internal fit values of the IST group were significantly higher than those of Group P. Sintering time reduction led to a decrease in ∆E00 values. Fracture load values were not statistically significantly affected by the different sintering procedures for both brands. Conclusion: Different sintering procedures did not have a clinically significant effect on fit and fracture load. Different sintering procedures were found to have an impact on the color change of monolithic zirconia restorations. Keywords: color, fit, fracture load, sintering, zirconia

Aim: The objective of the present article is to demonstrate the digital workflow used to manufacture an adjusted oral splint in a patient case. Materials and methods: A 25-year-old female patient presented for management of her bruxism. Therefore, an adjusted oral splint was manufactured. A computer-aided motion analysis of the patient was conducted (JMA Optic). Full-arch scans of the maxilla and mandible, a biocopy of the maxilla with a bite fork, and buccal scans of the centric jaw relation (Primescan) were performed. The jaw relation was determined beforehand by ballistic closing on a chairside-fabricated anterior jig. The digital construction of a Michigan splint took place in the dental laboratory. The design was nested and milled from a polymethyl methacrylate (PMMA)-containing blank (CLEARsplint Disc). Results: The oral splint was inserted into the patient’s mouth and checked to ensure a tension-free fit. The static and dynamic contact relationship was checked. During the follow-up visit, the patient reported an improvement in tension in the masticatory muscles. Conclusion: The described procedure allows for the manufacture of an adjusted oral splint through a purely digital workflow. Keywords: bruxism, centric relation, full-arch impression, hard acrylic oral splint, mandibular movement recordings, occlusal splint, temporomandibular disorders, treatment