课程文献中心 更多>>

关键词： CNN   ViT-E   fusion features   Grad-CAM   XGBoost   ANN   thyroid cancer  

摘要： Background/Objectives: Accurate detection of thyroid cancer using ultrasound remains a challenge, as malignant nodules can be microscopic and heterogeneous, easily confused with point clusters and borderline-featured tissues. Current studies in deep learning demonstrate good performance with convolutional neural networks (CNNs) and clustering;however, many approaches focus on local tissue and provide limited, non-quantitative interpretation, reducing clinical confidence. This study proposes an integrated framework combining enhanced convolutional feature encoders (DenseNet169 and VGG19) with an enhanced vision transformer (ViT-E) to integrate local feature and global relational context during learning, rather than delayed integration. Methods: The proposed framework integrates enhanced convolutional feature encoders (DenseNet169 and VGG19) with an enhanced vision transformer (ViT-E), enabling simultaneous learning of local feature representations and global relational context. This design allows feature fusion during the learning stage instead of delayed integration, aiming to improve diagnostic performance and interpretability in thyroid ultrasound image analysis. Results: The best-performing model, ViT-E-DenseNet169, achieved 98.5% accuracy, 98.9% sensitivity, 99.15% specificity, and 97.35% AUC, surpassing the robust basic hybrid model (CNN-XGBoost/ANN) and existing systems. A second contribution is improved interpretability, moving from mere illustration to validation. Gradient-weighted class activation mapping (Grad-CAM) maps demonstrated distinct and clinically understandable concentration patterns across various thyroid cancers: precise intralesional concentration for high-confidence malignancies (PTC = 0.968), edge/interface concentration for capsule risk patterns (PTC = 0.957), and broader-field activation consistent with infiltration concerns (PTC = 0.984), while benign scans showed low and diffuse activation (PTC = 0.002). Spatial audits reinforced this be

关键词： 3R principle   Cancer modelling   CAM   C. elegans   Drosophila   Zebrafish  

摘要： Comprehensively understanding cancer biology requires complex model systems, typically animal models, to replicate tumor behavior, microenvironment interactions, and cell migration/metastatic processes. Traditionally, genetically engineered and xenograft mouse models have been the cornerstone of cancer research. In recent years, the 3R principle-Replacement, Reduction, and Refinement-has reshaped ethical cancer research by promoting alternative animal models. Here we critically examine in vitro and in silico as well as four 3R-compliant animal models - the chicken chorioallantoic membrane (CAM) model, Danio rerio, Drosophila melanogaster, and Caenorhabditis elegans - and assess their advantages, limitations, and translational relevance for cancer research. Special emphasis is placed on practical considerations to inform optimal decision-making, including reproducibility and model-specific advantages and challenges. This review aims to support researchers in selecting ethical and effective preclinical models to advance cancer research.

关键词： 引纬机构   碰撞动力学   投梭压力角余弦耦合系数   多体动力学仿真   刚柔耦合  

摘要： 为了解决有梭织机引纬机构长期存在的高冲击、高磨损以及能量传递效率与冲击强度难以兼得的问题。文章基于经典撞击理论进行引纬机构动力学分析,并提出了一个核心设计参数——投梭压力角余弦耦合系数;进而,建立了以该系数为核心的引纬机构碰撞动力学模型,推导出撞击冲量、运动响应、有效冲击力及动能传递效率的显式表达式。最后,通过ADAMS平台构建了相应的刚柔耦合多体动力学仿真模型,对理论模型进行了多工况验证。结果表明,理论模型与仿真结果吻合良好;该系数的增大将单调加剧速度方向上的冲击载荷,并使动能传递效率单调下降,而该系数存在特定值使得运动响应速度获得最佳的动能输出。文章将传统依赖经验的参数调整转化为对该系数的调控与多目标权衡,可为引纬机构的优化设计提供理论工具与可行的工程路径。

关键词： Distributed CAD system   Simulation efficiency   Simulation framework   Thermal performance  

摘要： This study develops a distributed Computer-Aided Design (CAD) system for scalable thermal performance simulation in clothing engineering, overcoming key limitations of conventional single-server architectures. The proposed framework incorporates three core innovations: (1) a multiphysics coupling engine that models the human–clothing–environment system as a thermal system, integrating Stolwijk’s 25-node human thermoregulation model with dynamic fabric transfer equations via finite-volume discretization; (2) a distributed architecture that improves processing efficiency under multi-user concurrent scenarios; and (3) a five-stage scenario definition workflow supporting parametric modeling of activity-dependent metabolic rates, multi-layer garment configurations (across up to six segments), and environmental boundary conditions. Six simulation cases were tested–running at 8, 10, and 12 km/h under 20°C and 60 % relative humidity for 30 minutes each–demonstrating that the distributed CAD system significantly reduces simulation and user waiting times compared to a legacy Browser/Server system. It also enables visualization of human skin temperature and fabric properties on the trunk during simulated running. This distributed approach facilitates faster, large-scale simulations for designing sportswear and protective clothing, thereby reducing reliance on costly physical prototypes and accelerating the development of more comfortable and functional apparel.

关键词： GIS   CAD建模   cosRMC建模   pythonOCC   城市场景  

摘要： 基于GIS-CAD-MC集成建模方法,开发了用于快速生成大规模城市建筑蒙特卡罗模型的程序GIS2MC,用于提高城市建筑辐射场计算过程的建模效率。采用通用格式的地理信息系统(Geographic Information System, GIS)数据作为数据输入,使用GeoPandas包解析数据,通过Python内置库以及pythonOCC几何核心,生成城市建筑模块的CAD模型和cosRMC模型。通过选取国内外城市的GIS数据进行验证与测试,结果表明GIS2MC单次运行可生成万个城市建筑轮廓的CAD模型与蒙特卡罗(Monte Carlo,MC)模型,生成MC模型未出现几何模型错误或建筑体积丢失。此外,GIS2MC实现了基于GIS数据的建模区域选择、建筑轮廓预处理以及cosRMC模型材料和内部结构自动配置的辅助功能。

关键词： 3D garment CAD   Cloth simulation   Historical costumes   Digital heritage   Digital twin adequacy   CAD validation  

摘要： PurposeApparel CAD simulations use two input parameter types: objective parameters measured from source materials, such as patterns and fabrics, and subjective parameters arising from manufacturing processes, such as stitching or steam pressing. The latter rely on designer estimates to a lack of measurement protocols. The resulting subjectivity causes digital twins to deviate from physical garments. This study aims to introduce a metric to quantify this problem and prioritise parameters to improve adequacy of virtual ***/methodology/approachThis study addresses this gap in the context of historical clothing by introducing a four-stage analytical framework based on a novel degree of objective control (DOC) metric, defined as the proportion of a virtual garment's surface governed by objectively measurable parameters. Analysing 20 costumes and 320 parameter-adequacy pairs, this study correlated DOC with adequacy validated using an AI-driven computer vision workflow across three fidelity characteristics, construction, proportions and *** revealed statistically significant correlations, with surface parameters (e.g. Bond Strength) dominating shape accuracy (R = 0.89), seam parameters (e.g. Tension Ratio) governing structural integrity (R = 0.82) and edge parameters (e.g. Elastic Strength) influencing proportional fidelity (R = 0.76). Surface optimisation yielded 3 & times;greater return-on-effort in adequacy gains than other parameter ***/valueThe insights enabled a parameter prioritisation framework for resource-efficient optimisation of CAD workflows, a standardised adequacy audit protocol compliant with FAIR data principles and the first empirical foundation for mitigating subjectivity-induced errors. This helps designers focus efforts where it matters most, saving time and resources in all digital garment domains including virtual prototyping, sustainable design and heritage preservation.

关键词： Coronary artery disease   Excess apolipoprotein B   Long-term Survival   Residual risk   Secondary prevention   Statin therapy  

摘要： BACKGROUND:Residual cardiovascular risk persists in statin-treated patients with coronary artery disease (CAD), even when low-density lipoprotein cholesterol (LDL-C) targets are met. Excess apolipoprotein B (apoB), defined as measured apoB minus LDL-C-predicted apoB, may capture atherogenic particle burden beyond LDL-C, but its prognostic value for long-term mortality in secondary prevention remains uncertain. METHODS:We conducted a pooled analysis of two nationwide Chinese cohorts (CIN-II and RED-CARPET) comprising 68,616 statin-treated CAD patients. Excess apoB was calculated using an internal reference population (triglycerides ≤ 1.0 mmol/L). Associations with all-cause and cardiovascular mortality were assessed using multivariable Cox models, with adjustment for clinical covariates including nutritional status. External validation was performed in 13,702 participants from the UK Biobank. RESULTS:Over a median follow-up of 5.2 years, 10,835 deaths occurred (5,090 cardiovascular). Each 1-standard deviation (15.4 mg/dL) increase in excess apoB was associated with a 12% higher risk of all-cause mortality (adjusted hazard ratio [aHR] 1.12, 95% CI 1.06-1.18) and a 24% higher risk of cardiovascular mortality (aHR 1.24, 95% CI 1.15-1.34). Patients in the highest excess apoB quartile (≥ 11.5 mg/dL) had significantly worse survival. Notably, these associations persisted consistently across all achieved LDL-C strata (< 2.0 to > 4.0 mmol/L). These findings were robustly confirmed in the external validation cohort. CONCLUSIONS:Excess apoB is an independent predictor of long-term mortality in statin-treated CAD patients, even among those with well-controlled LDL-C. Its incorporation into risk assessment could improve prognostic stratification and guide personalized management in secondary prevention. TRIAL REGISTRATION:CIN-II: ***, NCT05050877 (Retrospectively registered, 21 September 2021); RED-CARPET: Chinese Clinical Trial Registry, ChiCTR2000039901 (Prospec

关键词： CAD/CAM restorative materials   Color stability   Surface roughness   Coffee staining   Toothbrushing abrasion  

摘要： Objectives This laboratory study aimed to investigate the effects of toothbrushing with different toothpaste formulations on the color stability and surface roughness of CAD/CAM restorative materials following coffee-staining thermocycling.
Materials and methods A total of 160 specimens were prepared from four restorative materials: IPS *** CAD, CEREC Tessera, Shofu Block HC, and VITA Enamic (n = 40 each). Following initial measurements, specimens underwent ther-mocycling in a coffee solution (5000 cycles) to simulate staining. Each group was subsequently divided and brushed with either a whitening toothpaste containing hydrated silica and hydrogen peroxide or a non-whitening toothpaste containing hydrated silica only. A total of 10,000 brushing cycles were performed under standardized conditions. Color stability was assessed using Delta E-0(0) values, and surface roughness was quantified using the Ra parameter at baseline, after thermocycling, and after brushing. Data were analyzed using Kruskal-Wallis and Mann-Whitney U tests with the significance threshold set at p < 0.05. SEM analysis was conducted on one randomly selected specimen per group (x1000).
Results Statistically significant differences in color stability, as assessed by Delta E-0(0) values, were observed among the tested materials (p < 0.001). Resin-based groups, particularly Shofu Block HC, exhibited the greatest reduction in color stability after coffee thermocycling and subsequent brushing (Delta E-0(0) = 1.069, p < 0.001), whereas glass-ceramics (IPS *** CAD and CEREC Tessera) demonstrated superior color stability (Delta E-0(0) <= 0.520). Surface roughness increased significantly in resin-containing materials (VITA Enamic Delta Ra = 0.036;Shofu Block HC Delta Ra = 0.020) compared with glass-ceramics (p < 0.001). No significant differences were detected between whitening and non-whitening toothpastes for either parameter across all groups (p > 0.05). Ra changes were also concluded by

关键词： cardiovascular remodeling   cognitive impairment   coronary artery disease   heart-brain axis   multimodal magnetic resonance imaging  

摘要： Coronary artery disease (CAD) and cognitive impairment (CI) are becoming more and more a public health challenge, and the heart-brain axis theory can describe their correlation. In this review, we highlight recent research and future directions for brain magnetic resonance imaging (MRI) based on CAD. Recent studies show that CAD damages the brain's white matter microstructure, disrupts brain functional network connectivity, and causes gray matter atrophy through multiple pathological mechanisms, including chronic cerebral hypoperfusion, cerebral small vessel disease, neuroinflammation, and oxidative stress, which in turn cause cognitive decline. Recently, multimodal brain MRI has been used to assess damage to the heart-brain axis. Graph-theoretic analysis and artificial intelligence can be used to learn the relationships between abnormal brain network topology, cardiovascular risk phenotypes, and brain imaging features in CAD patients, and have been proven to be a promising biomarker. Notably, this review is the first to systematically explore brain MRI biomarkers as objective tools for validating the scientific basis of traditional Chinese medicine (TCM) heart-brain theory in CAD-related CI. By elucidating the convergence of TCM holistic concepts with modern neuroimaging, we provide a novel imaging assessment framework to guide the development of integrated TCM-Western medicine strategies for the synergistic treatment of heart and brain.