课程文献中心 更多>>

关键词： Custom-made post   CAD/CAM   Customized fiber post   Milled post   Fracture resistance   Failure mode  

摘要： BackgroundPost-and-core systems are frequently used when restoring endodontically treated teeth. CAD/CAM technologies contribute to post fabrication by improving material uniformity and adaptation to canal morphology. Additionally, 3D printing presents an emerging approach that constructs posts incrementally and may address certain limitations associated with traditional fabrication methods. The purpose of this in vitro study was to evaluate the fracture resistance of custom-made CAD/CAM posts and cores fabricated using milling and additive techniques compared to customized fiber *** and methodsTwenty-one maxillary central incisors were randomly divided into three groups (n = 7): Group MP (milled posts), Group 3D (3D printed posts), and Group CF (customized fiber posts). The teeth were embedded in acrylic resin and prepared endodontically. Post spaces were created using standardized drills. For Groups MP and 3D, the post spaces were scanned, and posts were fabricated using CAD/CAM technology and 3D printing. In Group CF, posts were customized with resin composite. All posts were cemented with resin cement. The specimens were subjected to thermocycling for 5,000 cycles. Fracture resistance was measured using a universal testing machine, and failure modes were examined with a stereomicroscope. The Kruskal-Wallis test was used to compare the three groups. Failure mode data were reported as frequencies and percentages. Fisher's Exact test was used to compare failure modes among the three *** was no statistically significant difference in fracture resistance among the three groups (p = 0.400) or in failure modes (p = 0.381). The 3D group exhibited the highest percentage of restorable failures (85.7%), while the Milled group showed the highest percentage of non-restorable failures (57.1%).ConclusionsAll post types demonstrated comparable fracture resistance and favorable failure modes when restoring anterior endodontically treated teeth. Milled

关键词： CAD/CAM   retainer   retrospective study   fixed retainer  

摘要： Background/Objectives: The present study aims to assess the success rate of a CAD/CAM nickel-titanium wire (Memotain (R)) used as a fixed orthodontic retainer, over a one-year period. Methods: A retrospective study was conducted on 338 CAD/CAM nickel-titanium (Memotain (R)) fixed retention wires in 205 patients, bonded by a single experienced operator between January 2017 and December 2020. Follow-up visits were scheduled 6 (T1) and 12 months (T2) post-bonding. At each follow-up visit, events (defined as debonding, breakage, retainer loss, or tooth displacement) were classified by tooth, and success or failure of the retainer was determined based on the presence or absence of these events. Results: For the mandibular arch at T1 (6 months), the success rate was 85%, with debonding (n = 46) being the only event observed. At T2 (12 months), the success rate was 77%, with debonding (n = 30), wire breakage (n = 5) and retainer loss (n = 18) having occurred. For the maxillary arch, the overall success rate was 83% at T1 and 78% at T2. Debonding was the most common event observed over the 12-month observation period (n = 29), followed by retainer loss (n = 20) and wire breakage (n = 3). The overall success rates per type of tooth in the upper arch were 86% for the premolars, 96% for the canines, 95% for the lateral incisors and 93% for the central incisors. For the mandibular arch the success rates were 92% for the premolars, 97% for the canines, 96% for the lateral incisors and 94% for the central incisors. Conclusions: CAD/CAM nickel-titanium fixed retainers (Memotain (R)) demonstrated promising 1-year survival rates in both arches, though long-term multicentre studies are needed to confirm their reliability.

关键词： cardiometabolic risk   complication   coronary artery disease   glucagon   type 2 diabetes mellitus  

摘要： Objective To investigate the association between fasting glucagon levels and coronary artery disease (CAD) risk in patients with Type 2 Diabetes Mellitus (T2DM).
Methods This cross-sectional study enrolled 1,739 hospitalized T2DM patients, categorized into T2DM alone and T2DM with CAD (T2DM&CAD) groups. Fasting glucagon levels and clinical characteristics were collected. Multivariable logistic regression models were used to assess this association, with progressive adjustment for confounders.
Results In female patients, fasting glucagon levels were significantly higher in the T2DM&CAD group than in the T2DM alone group (13.33 vs. 11.52 pmol/L, P < 0.01). After full adjustment, each 1-SD increase was associated with a 49.2% higher CAD risk (OR: 1.492;95% CI: 1.099-2.026;P < 0.05). No significant association was found in male patients.
Conclusion Elevated fasting glucagon is an independent risk indicator for CAD in women with T2DM, but not in men. These findings highlight the potential value of glucagon monitoring in T2DM management, especially for women, and support exploring glucagon-pathway-targeted therapies to reduce cardiovascular complications.

关键词： Sign language   Artificial intelligence   Hands   Entropy   Imaging techniques   Preprocessing   Fingers   Learning  

摘要： Sign language recognition is crucial in bridging the communication gaps between hearing and deaf communities. In this study, we build on an existing sign language classification model based on the VGG19 architecture, enhancing its robustness through dataset augmentation and alternative data representations. We introduce a segmentation-based approach that utilizes hand masks generated by the U-Net model, replacing depth images to mitigate noise and improve classification accuracy. We use an Explainable Artificial Intelligence (XAI) approach that incorporates Grad-CAM, LIME, and integrated gradients methods to interpret the model's decision-making process, ensuring transparency and reliability in real-world applications. Our comparative analysis between Red-Green-Blue (RGB) and mask-based models demonstrates that while the RGB model benefits from richer texture and color information, the mask-based model effectively focuses on hand shape and structure. The integration of XAI further validates our results by highlighting key regions of the image that influence the model's predictions and by enabling a multi-perspective analysis that captures complementary aspects of the input, including region-based attention, pixel-level attribution, and structural shape analysis, thereby facilitating a deeper understanding of the model's internal representations and potential failure modes. This work contributes to advancing American Sign Language recognition by improving model generalisation and explainability, ultimately fostering greater trust and usability in assistive technologies.

关键词： accuracy   mechanical testing   removable dental prostheses   surveyed crown  

摘要： Purpose The success of removable partial dentures (RPDs) depends on well-designed abutments or surveyed crowns. Advances in computer-aided design and computer-aided manufacturing (CAD-CAM) technology have made all-ceramic materials viable alternatives to traditional ceramo-metal surveyed crowns, but their performance, particularly regarding internal fit and fracture resistance, requires further studies, especially with different rest seat preparations simulated on the abutment dies. This study evaluated the internal fit and fracture resistance of CAD-CAM ceramic crowns with wide and narrow rest seat designs to assess their potential as surveyed crowns for RPDs. Materials and Methods Sixty mandibular premolar abutment dies were modified to simulate narrow and wide occlusal relief designs and were divided into six groups based on the design (wide/narrow) and material type: force-absorbed hybrid ceramics (HC), zirconium lithium silicate (ZLS), and lithium disilicate glass-ceramic (LDC). CAD-CAM ceramic crowns were fabricated, milled, and analyzed for internal fit using Geomagic Control X software. Fracture resistance in newton (N) was tested using a universal testing machine. Scanning electron microscopy (SEM) was employed to examine fracture morphology. Statistical analyses included two-way analysis of variance (ANOVA) and post hoc Tukey's test (alpha = 0.05). Results The LDC group had the highest fracture resistance, while ZLS and HC showed no significant difference [F(2) = 10.523, p < 0.001]. Internal fit did not significantly differ between materials [F(2) = 0.381, p = 0.686], though LDC had a slightly larger internal gap. Rest seat design significantly affected both fracture resistance and internal fit [F(1) = 68.581, p < 0.001;F(1) = 11.185, p = 0.002], with wide designs showing higher fracture resistance and narrow designs having larger internal gaps. SEM revealed crack propagation originating from rest seat bases, with longer crack lines observed in wide rest d

关键词： accuracy   artificial intelligence   CAD   crown morphology   deep learning   implant-supported crowns  

摘要： Purpose: To investigate the effectiveness of a novel transformer-based deep learning model for generating monolithic implant-supported crowns (ISCs), comparing its performance to traditional software-generated crowns. Materials and Methods: A total of 311 patients who received single implant placements in the posterior first molar area were included. Digital impressions as well as technician-designed crowns were imported into the AI model. The PoinTr architecture-based AI model was trained on 291 cases and validated using 20 additional cases. Four groups of ISC designs were evaluated: Group 1, Software1-generated crowns;Group 2, Software 2-generated crowns;Group 3, AI-generated crowns;and Group 4, technician-designed crowns. Accuracy was evaluated through overall contour deviation, occlusal morphology, proximal contact, and emergence profile. Results: Relative to the reference (Group 4), AI-generated crowns (Group 3) demonstrated significantly lower overall contour deviation (438.6 +/- 169.4 mu m) compared with both software-generated crowns (Groups 1-2: 679.5-705.1 mu m, p < 0.01). Occlusal discrepancies were smaller and contact distributions more comparable to technician-designed crowns in the AI group, while software-generated crowns showed higher frequencies of premature contacts. For emergence profile, AI crowns predominantly reproduced concave contours and angles approximating technician-designed crowns, whereas software groups often produced convex forms. Across all automated groups (1-3), proximal contact adaptation remained less accurate than in technician-designed crowns. Conclusions: Within the limitations of this proof-of-concept study, AI-generated implant-supported crowns achieved closer approximation to technician designs than software-generated crowns, particularly in contour, occlusal morphology, and emergence profile.

关键词： Speaker verification   Deep learning   Speaker embeddings   Multi-scale   Selective kernel  

摘要： Speaker verification technology plays a crucial role in biometric authentication and security applications. However, the development of systems for low-resource languages such as Tibetan continues to encounter unique challenges. The complex phonology and distinctive acoustics of Tibetan, combined with the limited availability of speech data, make it difficult for conventional models to achieve strong performance. To address these challenges, this paper proposes an improved CAM-SKANF (CAM-Selective Kernel and Kolmogorov-Arnold Networks Fusion) network, built on the CAM++ framework and tailored to Tibetan speaker verification. First, we integrate Selective Kernel 2D Convolution (SK-2D) into the front-end ResBlock, forming the SKResBlend module. This module employs a multi-scale parallel convolution structure with a dynamic kernel selection mechanism, enhancing acoustic feature extraction and enabling more precise modeling of fine-grained variations and long-term dependencies. Second, we replace the original transition layer after the Dense Block with a KANs layer. Using the Kolmogorov-Arnold function decomposition, this layer optimizes high-dimensional feature representation, reduces dependence on large-scale data, and helps prevent overfitting. Finally, we introduce the Multi-scale Attention Normalization Fusion (MANF) module between adjacent KANs layers to fuse shallow local features with deep global semantics, strengthening the model's robustness and representational capacity. Experimental results show that CAM-SKANF achieves an EER (Equal Error Rate) of 5.5476%, an improvement of about 34% over CAM++. The MinDCF (Minimum Detection Cost Function) is reduced to 0.7291, an 11.5% improvement relative to the baseline. These results demonstrate that CAM-SKANF significantly enhances discrimination ability in Tibetan speaker verification, reducing errors and improving system reliability.

关键词： aortic valve stenosis   coronary artery disease   fractional flow reserve   transcatheter aortic valve replacement   AS   aortic valve stenosis   CAD   coronary artery disease   FFR   fractional flow reserve   PCI   percutaneous coronary intervention   RCT   randomized clinical trial   SPECT   single-photon emission computed tomography   TAVR   transcatheter aortic valve replacement  

摘要： The prognostic value of fractional flow reserve (FFR) in assessing coronary stenosis before transcatheter aortic valve replacement (TAVR) is unclear.

关键词： Fiber-reinforced polymer   CAD/CAM   Shear bond strength   Composite resin hybrid ceramic   Thermomechanical aging  

摘要： Background This study tested the null hypothesis (H0) that the type of veneering material would have no effect on the shear bond strength (SBS) to multilayer glass-fiber-reinforced polymer (MLG-FRP) frameworks. Methods Eighty-four specimens were prepared from Trinia and Zantex CAD/CAM discs (10 mm diameter, 1 mm thickness) and divided into six subgroups (n = 14). Each subgroup was veneered with composite resin (Ceramage, CE), milled hybrid ceramic (CeraSmart, CS), or 3D-printed hybrid ceramic (VarseoSmile Crown Plus, VS). All specimens underwent thermomechanical aging (240,000 chewing cycles;5000 thermal cycles at 5-55 degrees C). SBS was measured using a universal testing machine, and failure modes were classified under a stereomicroscope (20x). Premature test failures (PTFs) were coded as 0 MPa in the ITT analysis, with an additional PP analysis excluding PTFs, and data were analyzed by two-way ANOVA with Tukey's post-hoc test (alpha = 0.05) including Shapiro-Wilk, Levene, partial eta(2), and Hedges g (95% CI). Results The null hypothesis that veneering material does not affect bond strength was rejected. Veneering material significantly affected shear bond strength (SBS) (ITT: F = 5.66, p = 0.005;PP: F = 4.30, p = 0.018), while MLG-FRP type and interaction were not significant (p > 0.05). In intention-to-treat analysis (PTFs = 0 MPa), composite resin (CE: 12.87 +/- 2.66 MPa;95% CI: 11.88-13.86) outperformed milled (CS: 9.19 +/- 6.16 MPa) and 3D-printed hybrids (VS: 9.02 +/- 4.74 MPa). Per-protocol analysis confirmed this trend (CE: 12.87 +/- 2.66 MPa > CS: 12.87 +/- 2.04 approximate to VS: 10.99 +/- 2.25). Effect sizes were large (Hedges g up to 1.2). Kaplan-Meier analysis showed 100% survival in CE, significantly higher than CS (p = 0.002) and VS (p = 0.02), with no difference between CS and VS. Failures were mainly adhesive (> 90%). Conclusions The choice of veneering material is a critical determinant of bond strength to MLG-FRP frameworks. Composite resin pro

关键词： Classroom actions   Deep learning   Education   Grad-CAM   Machine learning   Students   Teacher  

摘要： In the era of advanced computer vision technology, it is possible to use automatic methods to detect and classify student and teacher activities in classroom environments, providing novel approaches to study or evaluate the quality of teaching or learning. However, to date, there has been little research developing and testing these methods to work towards an optimal activity recognition system. This paper proposes an automated framework using a 3D-convolutional neural network (CNN) to recognize classroom activities, including teacher and student behaviors, from classroom videos. The 3D-CNN captured spatiotemporal features from the video data. Then, an extreme learning machine (ELM) classifier was trained over the 3D-CNN features to recognize different activities in the classroom. Multi-layer perceptron (MLP) and support vector machine (SVM) classifiers were also examined in comparison to ELM. Gradient-weighted class activation mapping (Grad-CAM) was employed to provide visual explanations of what information the highest performing model learned from videos to classify classroom activities. To evaluate each model, classifications were carried out on the EduNet dataset, containing annotated classroom activities featuring students and teachers. Classroom videos from the internet were also utilized to further evaluate the performance of the proposed frameworks. The proposed 3D-CNN+ELM model achieved a maximum average recognition accuracy of 88.17 % on EduNet, as estimated by 5-fold cross-validation, which is 5.87 % higher than the standard baseline I3D-ResNet-50 model proposed by the EduNet authors. The model also achieved an accuracy of 80.00 % when applied to an independent dataset of videos sourced from the internet, indicating reasonable reliability and generalizability. The Grad-CAM outcomes indicate that the model focuses on valid features to determine its recognition; however, in some cases, the recognition can still be incorrect. With its high level of performa