课程文献中心 更多>>

摘要： Background Computer-aided detection (CAD) systems hold promise for improving tuberculosis (TB) detection on digital chest radiographs. However, data on their performance in exclusively paediatric populations are scarce. Methods We conducted a retrospective diagnostic accuracy study evaluating the performance of CAD4TBv7 (Computer-Aided Detection for Tuberculosis version 7) using digital chest radiographs from well-characterised cohorts of Gambian children aged <15 years with presumed pulmonary TB. The children were consecutively recruited between 2012 and 2022. We measured CAD4TBv7 performance against a microbiological reference standard (MRS) of confirmed TB, and also performed Bayesian latent class analysis (LCA) to address the inherent limitations of the MRS in children. Diagnostic performance was assessed using the area under the receiver operating characteristic curve (AUROC) and point estimates of sensitivity and specificity. Results A total of 724 children were included in the analysis, with confirmed TB in 58 (8%), unconfirmed TB in 145 (20%) and unlikely TB in 521 (72%). Using the MRS, CAD4TBv7 showed an AUROC of 0.70 (95% CI 0.60-0.79), and demonstrated sensitivity and specificity of 19.0% (95% CI 11-31%) and 99.0% (95% CI 98.0-100.0%), respectively. Applying Bayesian LCA with the assumption of conditional independence between tests, sensitivity and specificity estimates for CAD4TBv7 were 42.7% (95% CrI 29.2-57.5%) and 97.9% (95% CrI 96.6-98.8%), respectively. When allowing for conditional dependence between culture and Xpert assay, CAD4TBv7 demonstrated a sensitivity of 50.3% (95% CrI 32.9-70.0%) and specificity of 98.0% (95% CrI 96.7-98.9%). Conclusion Although CAD4TBv7 demonstrated high specificity, its suboptimal sensitivity underscores the crucial need for optimisation of CAD4TBv7 for detecting TB in children.

关键词： life cycle assessment   environmental impact assessment   wooden products   SimaPro   SolidWorks Sustainability   NX Sustainability   Fusion 360 Makersite   sustainable design  

摘要： The study evaluates how the environmental impacts of wooden products could be assessed in the early stages of product development using CAD-integrated life cycle assessment (LCA) tools. Focusing on a wooden chair design, the study compares the environmental impact results derived from LCA tools integrated in SolidWorks, NX and Fusion against a traditional LCA analysis performed using SimaPro. Methods involve analysing a chair model to measure the environmental impacts across different life cycle phases, such as material extraction and manufacturing. The results reveal that manufacturing processes, particularly electricity use, significantly contribute to environmental impacts, especially marine and freshwater ecotoxicity. Comparisons between LCA tools integrated into commercial CAD software and SimaPro 9.***.0.1. showed that while the tools deliver comparable results for global warming potential and other categories, they struggle with certain impact categories. The main distinguishing features of the results were methodological. Overall, the results aligned the most with the impact values calculated in Solidworks Sustainability. The study concludes that CAD-integrated tools are useful for early-stage environmental assessments but have limitations, particularly in their material databases and life cycle scope. For a comprehensive assessment, combining these tools with more detailed analysis methods may be necessary. The research suggests improvements for CAD-based tools to enhance their effectiveness in evaluating the environmental impact of wooden products.

关键词： Solid modeling   Reliability   Medical diagnostic imaging   Chatbots   Visualization   Brain modeling   Databases   Large language models   multi-modality system   medical dialogue   computer-assisted diagnosis  

摘要： The integration of Computer-Aided Diagnosis (CAD) with Large Language Models (LLMs) presents a promising frontier in clinical applications, notably in automating diagnostic processes akin to those performed by radiologists and providing consultations similar to a virtual family doctor. Despite the promising potential of this integration, current works face at least two limitations: (1) From the perspective of a radiologist, existing studies typically have a restricted scope of applicable imaging domains, failing to meet the diagnostic needs of different patients. Also, the insufficient diagnostic capability of LLMs further undermine the quality and reliability of the generated medical reports. (2) Current LLMs lack the requisite depth in medical expertise, rendering them less effective as virtual family doctors due to the potential unreliability of the advice provided during patient consultations. To address these limitations, we introduce ChatCAD+, to be universal and reliable. Specifically, it is featured by two main modules: (1) Reliable Report Generation and (2) Reliable Interaction. The Reliable Report Generation module is capable of interpreting medical images from diverse domains and generate high-quality medical reports via our proposed hierarchical in-context learning. Concurrently, the interaction module leverages up-to-date information from reputable medical websites to provide reliable medical advice. Together, these designed modules synergize to closely align with the expertise of human medical professionals, offering enhanced consistency and reliability for interpretation and advice. The source code is available at GitHub.

关键词： String insulators   Dirty insulators   Synthetic dataset   CAD   Inventor   Unity 3D  

摘要： String insulators are interconnected disks that insulate the electrical conductors of high-voltage transmission lines from the supporting towers, withstanding both mechanical stress and electrical current. Regular cleaning is crucial for string insulators, as different types of pollutants can build up on the insulators, impairing their ability to prevent electrical arcing and short circuits. However, visual observation, the primary method for verifying cleaning needs, can be misleading. Developing robust algorithms for autonomous system-based inspection leveraging artificial intelligence presents a promising solution to address the limitations of visual inspection in preventive insulator maintenance. However, machine learning algorithms require considerable training data to learn from patterns, avoid overfitting, and account for outliers and data variation within the training dataset, especially considering the limited availability of images depicting dirty string insulators, as reported in the literature. In response to the limitations of existing training data for machine learning algorithms in insulator inspection, this work proposes developing a synthetic dataset for clean and dirty string insulators using software tools, such as SolidWorks, Inventor, and Unity 3D to model the towers, insulators, backgrounds, and various pollutant types. The clean string insulator dataset contains 47,286 synthetic images featuring diverse string insulator materials (glass, polymer, and porcelain) situated within various realistic backgrounds (mountain, forest, desert, city, river, and plantation). Similarly, the dirty insulator dataset comprises 14,424 distinct images, including clean insulators and simulated contaminants, such as soot, salt, and bird droppings. To test the proposed dataset, first, a classical neural semantic segmentation network was used to segment the clean string insulators dataset images, and an average dice coefficient of 0.95 was achieved by using only syn

关键词： Inactivated SARS-CoV-2 vaccine   CAD   Thrombosis   Platelet activation   Prothrombotic antibody  

摘要： Association of Coronavirus disease 2019 vaccines with thrombosis has raised concerns among patients with coronary atherosclerosis disease (CAD).

关键词： FTIR spectral image   Chenpi   2D-CNN   Grad-CAM   Feature visualization  

摘要： Dried tangerine peel ("Chenpi"), has numerous clinical and nutritional benefits, with its quality being significantly influenced by its storage age, referred to as "Chen Jiu Zhe Liang" in Chinese. Concequently, the rapid and accurate identification of Chenpi's age is important for consumers. In this study, Fourier transform infrared spectroscopy (FTIR) was employed to capture spectral images of Chenpi. These FTIR images were then analyzed using a two-dimensional convolutional neural networks (2D-CNN) model, achieving a discrimination accuracy of 97.92%. To address the "black box" nature of the 2D-CNN, Gradient-weighted Class Activation Mapping Plus Plus (Grad-CAM++) was utilized to highlight the important regions contributing to the model's performance. Additionally, six other machine learning models were developped using features identified by the 2D-CNN to validate their effectiveness. The results demonstrated that the combination of FTIR spectral images and 2D-CNN provides a highly effective method for accurately determining the age of Chenpi.

关键词： Modulator   dynamics   vibrational characteristics   cam profile errors   numerical simulation   stability   heald frame  

摘要： This research endeavors to analyze the dynamic characteristics of the dobby, particularly the influence of a modulator on the dynamic response of the heald frame. By constructing precise kinematic and dynamic mathematical models and employing efficient computational algorithms, this study delves into the motion behavior of dobby. Component elastic deformation is considered, and a detailed analysis of the heald frame's vibrational properties is conducted. With the establishment of dynamic and kinematic equations, coupled with Taylor series expansion and partial derivative techniques, the displacement errors were accurately calculated. The implementation of the fourth-order Runge-Kutta method facilitated numerical simulation, allowing for the analysis of the effect of cam profile errors on vibrational characteristics. The findings indicate that cam profile machining and system errors significantly influence the vibration amplitude under high-speed conditions. Consequently, this research highlights the imperative to control operating speeds in the design and operation of dobby and enhance vibration stability through precise machining of cam profiles, using high-precision equipment, and regular maintenance, which are essential for optimized design and improved stability and reliability.

关键词： antiplatelet therapy   aspirin   desensitization   cardiovascular disease   hypersensitivity   ACS   acute coronary syndrome   AH   aspirin hypersensitivity   ASCVD   atherosclerotic cardiovascular disease   CAD   coronary artery disease   CCS   chronic coronary syndrome   COX   cyclo-oxygenase   CV   cardiovascular   DAPT   dual antiplatelet therapy   DPI   dual pathway inhibition   L-ASA   lysine acetylsalicylate   LDAC   low-dose aspirin challenge   LOE   Level of Evidence   NSAID   nonsteroidal anti-inflammatory drug   PCI   percutaneous coronary intervention   PD   pharmacodynamic   PG   prostaglandin   STEMI   ST-segment elevation myocardial infarction  

摘要： Low-dose aspirin remains the most commonly used antiplatelet agent among patients with atherosclerotic cardiovascular disease. Aspirin hypersensitivity occurs in 1% to 5% of patients and is among the most frequent causes for prohibiting the use of aspirin, posing a significant dilemma on how to manage these patients in clinical practice. Aspirin hypersensitivity is often misinterpreted and confused with aspirin intolerance, with treatment approaches being often unclear and lacking specific recommendations. Aspirin desensitization and low-dose aspirin challenge have emerged as pragmatic, effective, and safe approaches in patients with suspected or confirmed aspirin hypersensitivity who require aspirin therapy, but they are underused systematically in clinical practice. Furthermore, there is confusion over alternative antiplatelet agents to be used in these patients. The pathophysiological mechanisms and classification of aspirin hypersensitivity, as well as alternative strategies and practical algorithms to overcome the need for aspirin use in patients with atherosclerotic cardiovascular disease with suspected aspirin hypersensitivity, are discussed.

摘要： Background The aim of the present research was to validate 2 pivotal pretest probability (PTP)–CAD models (CAD Consortium model and IJC-CAD model). Methods Two traditional PTP models were generated under the CAD Consortium framework, namely CAD1 a...

摘要： Background The yield of coronary angiography is low; an inordinate number of patients sent to catheterization do not have disease. To combat this, we have developed an artificial intelligence model—ECGio, designed to be deployed at the point-of-ca...