课程文献中心 更多>>

关键词： Boundary representation   CAD model   Classification   Graph convolutional network   Segmentation  

摘要： The widespread adoption of 3D CAD models in intelligent manufacturing and digital twin systems demands efficient and accurate classification and segmentation. While effective approaches typically model geometry and topology jointly using B-rep face-edge graphs, prevailing methods rely heavily on auxiliary information (e.g., supplemental geometry/semantics) to compensate for perceived shortcomings in raw B-rep data. This dependency incurs high preprocessing overhead, limits cross-platform generalization, and hinders industrial deployment. Furthermore, inherent redundancies within B-rep graphs can obscure essential semantics during encoding. To overcome these limitations, we propose UVConv, an end-to-end dual-path framework that operates exclusively on the original UV grid and face adjacency graph, eliminating auxiliary inputs. UVConv employs a serial geometry-topology processing: the geometric encoder utilizes bottleneck compression to amplify discriminative features, while the topological encoder integrates a Bidirectional Interaction Module (BIM) and an Attention Filtering Module (AFM) to extract critical structural semantics and suppress noise. Comprehensive experiments on multiple benchmark CAD datasets demonstrate that UVConv matches or surpasses state-of-the-art performance without auxiliary data, exhibiting superior representational capacity, robust generalization, and enhanced industrial deployability.

关键词： 窄间隙焊接技术   TC4钛合金   热循环   显微组织   力学性能  

摘要： 热循环对钛合金厚板窄间隙焊接接头的组织和性能具有重要影响,本工作采用摆动送丝磁控电弧窄间隙焊接技术完成60 mm厚TC4钛合金焊接接头,借助数值模拟研究了典型热循环下的显微组织演变(α相、β相和晶界α_(GB))、显微硬度变化及拉伸性能变化。结果表明,高温热循环发生完全的α→β转变,将有利于α相择优生长,形成粗大且方向一致的α群落;相比之下,后续低温热循环发生不完全α→β转变,较低的冷却速率及温度梯度将削弱择优取向,板条α相得以细化。经过整个热循环后,板条α相的厚度由0.91±0.09μm增加至1.02±0.18μm,残余β相比例由0.02%增加至1.89%。得益于细晶强化作用,焊接接头打底层显微硬度最高;打底层拉伸强度最高,盖面层延伸率最高,接头均表现为韧性断裂。

关键词： Mathematical chemistry   QSPR analysis   Random forest regression   Topological indices   Xgboost regression  

摘要： In this study, a quantitative structure–property relationship (QSPR) framework is presented that combines simple degree-based topological descriptors with ensemble machine learning methods to examine selected physicochemical properties of coronary artery disease (CAD) and cardiovascular therapy drugs. A data set of 100 clinically used compounds was considered, for which six classical degree-based indices, the Atom Bond Connectivity $\left(ABC\right)$, Atom-Bond Sum Connectivity $\left(ABS\right)$, Sum Connectivity $\left(SC\right)$, Harmonic $\left(H\right)$, Randić $\left(R\right)$, and Geometric–Arithmetic $\left(GA\right)$ were computed, and compared to four key properties: topological polar surface area ($tPSA$), calculated lipophilicity ($CLogP$), calculated molar refractivity ($CMR$), and aqueous solubility (

摘要： This digital clinical report presents a novel CAD-CAM workflow for fabricating a customized monolithic zirconia post-and-core restoration. A 26-year-old female patient required prosthetic rehabilitation of the maxillary left first and second premolars, as well as an implant screw-retained crown (ISRC) for the maxillary left first molar. After post space preparation, intraoral scans (IOS) were obtained both with and without a prefabricated fiber post in place. The fiber post was also scanned separately using a desktop scanner. These datasets were superimposed using digital software to design a precise monolithic zirconia post-and-core, which was then milled using CAD-CAM technology. The structure was cemented with self-adhesive resin cement, and a lithium disilicate crown was digitally designed and fabricated. This digital approach minimized the errors associated with conventional techniques, improved fit and adaptation, and enhanced clinical efficiency. The definitive restoration achieved excellent aesthetic and functional outcomes, fulfilling the patient's expectations without complications. This report highlights the value of digital workflows in achieving precision and predictability in post-and-core restorations.

关键词： Aeonium   Canary Islands   Crassulaceae   crassulacean acid metabolism   cuticle   functional traits   minimum conductance   succulence  

摘要： Discovering functional and evolutionary interdependencies of hydraulic traits and crassulacean acid metabolism (CAM) is crucial to understand CAM phenotype diversity and convergence. In complex traits such as CAM, the co-option of associated traits strongly impacts the evolutionary outcome. Here we study Aeonium (Crassulaceae), a diverse Macaronesian genus that exhibits a broad array of CAM expression, focusing on two CAM-associated traits, minimum conductance (gmin) and succulence. At the heart of the study, there are two experiments: a comparative cultivation experiment to monitor nocturnal acidification (Delta H+) under drought and heat treatments and a leaf-drying curve experiment to quantify gmin. Our study group was comprehensively sampled to cover its phylogenetic and ecological diversity. We found a consistently negative correlation of gmin and Delta H+, indicating a critical role for the cuticle in the function of CAM. Although Delta H+ and succulence were overall not positively correlated, we found evidence that more succulent species remain in the CAM mode when stress is relaxed. We conclude that there is a tight evolutionary link between cuticular transpiration barrier properties and CAM performance. Thereby, the stronger CAM plants express diurnal stomatal closure typical of CAM, the more they may optimise water-use efficiency through reduced gmin.

关键词： cervical artery dissection   chiropractic manipulation   STOP-CAD registry   ischemic stroke   neck pain  

摘要： Objectives: Cervical artery dissection (CeAD) is an important cause of ischemic stroke in young adults. Nearly 100 million annual chiropractic cervical manipulations are performed in the United States. The relationship between manipulation and CeAD remains controversial. Methods: We analyzed patients in the multicenter STOP-CAD registry (n=4023) to identify CeAD cases diagnosed after chiropractic cervical manipulation. Demographics and clinical features were compared between manipulation-associated and nonmanipulation-associated cases using chi 2 and t tests. Multivariable logistic regression identified key factors associated with manipulation-related CeAD. Results: About 1 in 20 CeAD cases in this registry reported antecedent cervical manipulation. In multivariable binary logistic regression, compared with patients without prior manipulation, those with prior manipulation were younger (OR per year 0.98, 95% CI: 0.97-0.99, P=0.014), more often female (OR: 1.64, 95% CI: 1.21-2.23, P=0.001), less often diabetic (OR: 0.24, 95% CI: 0.08-0.78, P=0.018), presented with neck pain (OR: 2.80, 95% CI: 2.08-3.77, P<0.001), and had higher odds of isolated vertebral artery dissection (OR: 2.15, 95% CI: 1.57-2.94, P<0.001). Recurrent ischemic stroke rates were similar between groups. Conclusions: Given the very high number of manipulations performed annually, the absolute risk of secondary CeAD is extremely low. Manipulation-associated cases have distinct clinical features, occurring more often in younger women with vertebral dissections. Whether manipulation acts as a precipitating trigger or patients with early CeAD symptoms seek manipulation remains unresolved.

关键词： ACS   CAD   SYNTAX   CYP2C19  

摘要： Background Coronary angiography is the gold standard for assessing coronary artery disease (CAD), while the SYNTAX score (SxS) quantifies lesion complexity. Its prognostic value across clinical settings remains uncertain. This study compared angiographic findings in stable angina and acute coronary syndrome (ACS) using the SxS. Methods A prospective study was conducted at Shaikh Zayed Hospital, Lahore, over 12 months, including 156 patients undergoing coronary angiography: stable angina (n = 78) and ACS (n = 78). Patients aged 18-80 years undergoing primary PCI or elective angiography were included, excluding prior PCI or CABG. Demographic and clinical data were collected, and SxS was applied post-angiography. Data were analyzed using SPSS v.25, with p <= 0.05 considered significant. Results A total of 156 patients were enrolled (71.2% male;mean age 59.2 +/- 10.6 years, BMI 27.2 +/- 3.5 kg/m(2)), equally divided into stable angina (n = 78) and ACS (n = 78). The overall mean SxS was 26.1 +/- 11.3. Comparison between groups showed no significant differences in age, BMI, sex distribution, smoking, hypertension, or diabetes (all p > 0.05). However, dyslipidemia was more prevalent in ACS patients (p < 0.05), and the mean SxS was significantly higher in ACS versus stable angina (28.2 +/- 9.9 vs. 23.8 +/- 12.4;p = 0.01), indicating greater angiographic complexity. Discussion The higher SxS in ACS patients indicated greater angiographic complexity, consistent with prior studies. Scores were also significantly elevated in patients with diabetes and hypertension, supporting the score's prognostic value in CAD stratification. However, the absence of clinical variables limits risk prediction. The single-center design, small sample size, lack of adjusted analyses, and absence of long-term outcomes further constrain generalizability, warranting validation in larger multicenter cohorts. Conclusion The SxS was significantly higher in ACS than stable angina, supporting its role in g

关键词： coronary CT angiography   CAD-RADS 2.0   large language models  

摘要： Objective: To assess whether few-shot prompting improves the performance of 2 popular large language models (LLMs) (ChatGPT o1 and DeepSeek-R1) in assigning Coronary Artery Disease Reporting and Data System (CAD-RADS (TM) 2.0) categories Methods: A detailed few-shot prompt based on CAD-RADS (TM) 2.0 framework was developed using 20 reports from the MIMIC-IV database. Subsequently, 100 modified reports from the same database were categorized using zero-shot and few-shot prompts through the models' user interface. Model accuracy was evaluated by comparing assignments to a reference radiologist's classifications, including stenosis categories and modifiers. To assess reproducibility, 50 reports were reclassified using the same few-shot prompt. McNemar tests and Cohen kappa were used for statistical analysis. Results: Using zero-shot prompting, accuracy was low for both models (ChatGPT: 14%, DeepSeek: 8%), with correct assignments occurring almost exclusively in CAD-RADS 0 cases. Hallucinations occurred frequently (ChatGPT: 19%, DeepSeek: 54%). Few-shot prompting significantly improved accuracy to 98% for ChatGPT and 93% for DeepSeek (both P<0.001) and eliminated hallucinations. Kappa values for agreement between model-generated and radiologist-assigned classifications were 0.979 (0.950, 1.000) (P<0.001) for ChatGPT and 0.916 (0.859, 0.973) (P<0.001) for DeepSeek, indicating almost perfect agreement for both models without a significant difference between the models (P=0.180). Reproducibility analysis yielded kappa values of 0.957 (0.900, 1.000) (P<0.001) for ChatGPT and 0.873 [0.779, 0.967] (P<0.001) for DeepSeek, indicating almost perfect and strong agreement between repeated assignments, respectively, with no significant difference between the models (P=0.125). Conclusion: Few-shot prompting substantially enhances LLMs' accuracy in assigning CAD-RADS (TM) 2.0 categories, suggesting potential for clinical application and facilitating system adoption.