Predicting the Debonding of CAD/CAM Composite Resin Crowns with AI

A preventive measure for debonding has not been established and is highly desirable to improve the survival rate of computer-aided design/computer-aided manufacturing (CAD/CAM) composite resin (CR) crowns. The aim of this study was to assess the usefulness of deep learning with a convolution neural network (CNN) method to predict the debonding probability of CAD/CAM CR crowns from 2-dimensional images captured from 3-dimensional (3D) stereolithography models of a die scanned by a 3D oral scanner. All cases of CAD/CAM CR crowns were manufactured from April 2014 to November 2015 at the Division of Prosthodontics, Osaka University Dental Hospital (Ethical Review Board at Osaka University, approval H27-E11). The data set consisted of a total of 24 cases: 12 trouble-free and 12 debonding as known labels. A total of 8,640 images were randomly divided into 6,480 training and validation images and 2,160 test images. Deep learning with a CNN method was conducted to develop a learning model to predict the debonding probability. The prediction accuracy, precision, recall, F-measure, receiver operating characteristic, and area under the curve of the learning model were assessed for the test images. Also, the mean calculation time was measured during the prediction for the test images. The prediction accuracy, precision, recall, and F-measure values of deep learning with a CNN method for the prediction of the debonding probability were 98.5%, 97.0%, 100%, and 0.985, respectively. The mean calculation time was 2 ms/step for 2,160 test images. The area under the curve was 0.998. Artificial intelligence (AI) technology-that is, the deep learning with a CNN method established in this study-demonstrated considerably good performance in terms of predicting the debonding probability of a CAD/CAM CR crown with 3D stereolithography models of a die scanned from patients.

Development of a Cavity Disinfectant Containing Antibacterial Monomer MDPB

An experimental cavity disinfectant (ACC) that is intended to be used for various direct and indirect restorations was prepared by adding an antibacterial monomer 12-methacryloyloxydodecylpyridinum bromide (MDPB) at 5% into 80% ethanol. The antibacterial effectiveness of ACC and its influences on the bonding abilities of resin cements were investigated. To examine the antibacterial activity of unpolymerized MDPB, the minimum inhibitory and bactericidal concentrations (MIC and MBC) were determined for Streptococcus mutans, Lactobacillus casei, Actinomyces naeslundii, Parvimonas micra, Enterococcus faecalis, Fusobacterium nucleatum, and Porphyromonas gingivalis. Antibacterial activities of ACC and the commercial cavity disinfectant containing 2% chlorhexidine and ethanol (CPS) were evaluated by agar disk diffusion tests through 7 bacterial species and by MIC and MBC measurement for S. mutans. The effects of ACC and CPS to kill bacteria in dentinal tubules were compared with an S. mutans–infected dentin model. Shear bond strength tests were used to examine the influences of ACC on the dentin-bonding abilities of a self-adhesive resin cement and a dual-cure resin cement used with a primer. Unpolymerized MDPB showed strong antibacterial activity against 7 oral bacteria. ACC produced inhibition zones against all bacterial species similar to CPS. For ACC and CPS, the MIC value for S. mutans was identical, and the MBC was similar with only a 1-step dilution difference (1:2). Treatment of infected dentin with ACC resulted in significantly greater bactericidal effects than CPS ( P < 0.05, analysis of variance and Tukey’s honest significant difference test). ACC showed no negative influences on the bonding abilities to dentin for both resin cements, while CPS reduced the bond strength of the self-adhesive resin cement ( P < 0.05). This study clarified that the experimental cavity disinfectant containing 5% MDPB is more effective in vitro than the commercially available chlorhexidine solution to eradicate bacteria in dentin, without causing any adverse influences on the bonding abilities of resinous luting cements.

Clinical effectiveness of contemporary adhesives for the restoration of non-carious cervical lesions. A systematic review

OBJECTIVES

The aim of this systematic review was to evaluate the clinical effectiveness of contemporary adhesives for the restoration of non-carious cervical lesions (NCCLs) in terms of restoration retention as a function of time.

METHODS

Medline Ovid and IADR abstracts were reviewed for NCCLs clinical trials from 1950 to 2013. The reference list of all eligible trials and relevant review articles was checked to find additional studies. The review did not have any language restrictions. Only randomized controlled clinical trials that evaluated at least two adhesives for a follow-up period of at least 18 months were included. Materials with adhesive potential were categorized into 6 main classes: 3-step etch&rinse adhesives (3E&Ra's), 2-step etch&rinse adhesives (2E&Ra's), 2-step self-etch adhesives (2SEa's), 1-step self-etch adhesives (1SEa's), glass-ionomers (GI's) and self-adhesive composites (SAC's). The first four can bond restorative composite to tooth tissue. Both 2SEa and 1SEa were further sub-divided in 'mild' and 'intermediately strong (1/2SEa_m), with a pH≥1.5, and 'strong' (1/2SEa_s), with a pH<1.5. From the restoration retention rates as a function of time the average annual failure rate (AFR) per adhesive and adhesive class was calculated.

RESULTS

The lowest AFR scores [mean (SD)] were recorded for GI [2.0 (1.4)] shortly followed by 2SEa_m [2.5 (1.5)], 3E&Ra [3.1 (2)] and 1SEa_m [3.6 (4.3)] (Tukey Contrasts: p>0.05). Significantly higher AFR scores were recorded for 1SEa_s [5.4 (4.8)], 2E&R [5.8 (4.9)], and 2SEa_s [8.4 (7.9)] (p>0.05). In addition, significant differences in AFR were noticed between adhesives of the same class (Kruskal-Wallis sum test: p>0.05), except for GI (p=0.7) and 2SEa_m (p=0.1). Finally, selective enamel etching did not significantly influence the retention rate of SEa (AFR SEa_etch=0.43 (0.49), AFR SEa_non-etch=1.43 (1.77).

SIGNIFICANCE

The adhesive approach significantly influences the clinical effectiveness of adhesives in NCCLs. Within each class, except for GI, there was a wide variation in clinical bonding effectiveness.

Meta-analytical review of parameters involved in dentin bonding

Bond-strength testing is the method most used for the assessment of bonding effectiveness to enamel and dentin. We aimed to disclose general trends in adhesive performance by collecting dentin bond-strength data systematically. The PubMed and EMBASE databases were used to identify 2,157 bond-strength tests in 298 papers. Most used was the micro-tensile test, which appeared to have a larger discriminative power than the traditional macro-shear test. Because of the huge variability in dentin bond-strength data and the high number of co-variables, a neural network statistical model was constructed. Variables like 'research group' and 'adhesive brand' appeared most determining. Weighted means derived from this analysis confirmed the high sensitivity of current adhesive approaches (especially of all-in-one adhesives) to long-term water-storage and substrate variability.

Enamel-Smear Compromises Bonding by Mild Self-Etch Adhesives

In light of the increased popularity of less acidic, so-called ‘ultra-mild’ self-etch adhesives, adhesion to enamel is becoming more critical. It is hypothesized that this compromised enamel bonding should, to a certain extent, be attributed to interference of bur debris smeared across enamel during cavity preparation. High-resolution transmission electron microscopy revealed that the enamel smear layer differed not only in thickness, but also in crystal density and size, depending on the surface-preparation method used. Lab-demineralization of sections clearly disclosed that resin-infiltration of an ultra-mild self-etch adhesive progressed preferentially along micro-cracks that were abundantly present at and underneath the bur-cut enamel surface. The surface-preparation method significantly affected the nature of the smear layer and the interaction with the ultra-mild adhesive, being more uniform and dense for a lab-SiC-prepared surface vs. a clinically relevant bur-prepared surface.

Filler Debonding & Subhybrid-layer Failures in Self-etch Adhesives

The mechanisms behind bond degradation are still largely unknown, in particular with respect to self-etch adhesives. One-step adhesives have been especially documented with problems, such as insufficient polymerization, water-uptake and subsequent plasticization, water- and enzyme-induced nanoleakage, and/or the presence of voids due to phase-separation or osmosis. It was hypothesized that these shortcomings may weaken the adhesive layer and, as such, may jeopardize long-term bonding. In contrast to the control three-step etch & rinse adhesive, the bond strength to dentin of both one-step and two-step self-etch adhesives decreased after six-month water storage. TEM revealed not only that they exhibited filler de-bonding within the adhesive resin layer, due to hydrolysis of the filler-matrix coupling, but also that they failed predominantly directly under the hybrid layer at dentin, in spite of the presence of interfacial droplets and nanoleakage in the adhesive layer. These failures just under the hybrid layer may be attributed to insufficient encapsulation of surface smear.

The effect of clinical experience on dentine bonding effectiveness: students versus trained dentists

Clinical successful application of dentine adhesives depends not only on material-related but also on operator-related factors. The purpose of this study was to evaluate the dentine bonding effectiveness of a self-etch composite cement applied by operators with or without clinical experience under well-standardized, randomized and blind conditions. Forty-eight bovine dentine surfaces were randomly divided into two groups. The first group consisted of eight dental students with no clinical experience at all, and the second group consisted of eight dentists with extensive experience in adhesive dentistry (mean experience of 11.4 years). Next, a 4-mm-diameter stainless steel rod (SUS-304) was bonded to the dentine surface using Panavia Fluoro cement (Kuraray Medical Inc., Tokyo, Japan). After application procedures, the specimens were randomized and shear bond-strength measurements were performed by a single blinded operator. Mann-Whitney U test was used to determine statistical differences in bond strength between the two groups, and Kruskal-Wallis was used to determine statistical difference between the student and dentist groups. The means and standard deviations of bond strength were 11.5 +/- 8.1 MPa for the student group and 7.1 +/- 4.3 MPa for the dentist group, respectively. The bond strength of the student group was significantly higher than that of the dentist group. However, the variability in bond strength was significantly higher in the student group, and some specimens failed prior to actual testing (included as 0 MPa). Clinical experience did not have a positive effect on the bonding effectiveness of the self-etch composite cement to dentine.

Impact of implant number, distribution and prosthesis material on loading on implants supporting fixed prostheses

The purpose of this study is to evaluate axial forces and bending moments (BMs) on implants supporting a complete arch fixed implant supported prosthesis with respect to number and distribution of the implants and type of prosthesis material. Seven oral Brånemark implants with a diameter of 3.75 mm and a length of 13 and 7 mm (short distal implant) were placed in an edentulous composite mandible used as the experimental model. One all-acrylic, one fibre-reinforced acrylic, and one milled titanium framework prosthesis were made. A 50 N vertical load was applied on the extension 10 mm distal from the most posterior implant. Axial forces and BMs were measured by calculating signals from three strain gauges attached to each of the abutments. The load was measured using three different models with varying numbers of supporting implants (3, 4 and 5), three models with different implant distribution conditions (small, medium and large) and three models with different prosthesis materials (titanium, acrylic and fibre-reinforced acrylic). Maximum BMs were highest when prostheses were supported by three implants compared to four and five implants (P < 0.001). The BMs were significantly influenced by the implant distribution, in that the smallest distribution induced the highest BMs (P < 0.001). Maximum BMs were lowest with the titanium prosthesis (P < 0.01). The resultant forces on implants were significantly associated with the implant number and distribution and the prosthesis material.

Inhibition of Enzymatic Degradation of Adhesive-Dentin Interfaces

Adhesive procedures activate dentin-associated matrix metalloproteinases (MMPs), and so iatrogenically initiate bond degradation. We hypothesized that adding MMP inhibitors to adhesive primers may prevent this endogenous enzymatic degradation, thereby improving bond durability. A non-specific MMP inhibitor (chlorhexidine) and a MMP-2/9-specific inhibitor (SB-3CT) were admixed to the primers of an etch & rinse and a self-etch adhesive, both considered as gold-standard adhesives within their respective categories. For dentin powder exposed to the adhesives under clinical application conditions, gelatin zymography revealed the release of MMP-2 (not of MMP-9) by the etch & rinse adhesive, while no release of enzymes could be detected for the mild self-etch adhesive, most likely because of its limited dentin demineralization effect. The built-in MMP inhibitors appeared effective in reducing bond degradation only for the etch & rinse adhesive, and not for the self-etch adhesive. Water sorption of adhesive interfaces most likely remains the principal mechanism of bond degradation, while endogenous enzymes appear to contribute to bond degradation of only etch & rinse adhesives.

Migratory responses of hemocytes to Vibrio parahaemolyticus in the alimentary tract of an estuarine neritid gastropod, Clithon retropictus

Migratory responses of hemocytes to Vibrio parahaemolyticus strain D3 in the alimentary tracts of an estuarine neritid gastropod, Clithon retropictus, and a related marine neritid, Nerita albicilla, were examined under the scanning electron microscope. After ingesting the strain, active responses were seen at the esophagus, stomach and anterior intestine of adult C. retropictus and at the middle and posterior intestines of adult N. albicilla. When the alimentary tracts were isolated from the gastropod and incubated in vitro with strain D3, active response was induced at the most parts of the tract of the adult gastropods and at the stomach and the anterior intestine of juvenile C. retropictus. The responding hemocytes were confirmed to be granulocytes in the semi-thin sections of the tract of adult C. retropictus. The poor hemocyte responses at the middle and posterior intestines of juvenile C. retropictus might support the colonization of the organism there.