30796 research outputs found
Sort by
The ‘OECD machine’ – Using a negative universality gaze to examine the OECD and its positive universal engineering fantasy
Get PDFRecently, there has been a greater emphasis, especially inspired by the Organisation for Economic Co-operation and Development (OECD), on the creation of educational policy that appears to want to control the future. This is evidenced by promoting an engineering fantasy in/for education that embodies positive universalism. We critically examine such positive universalism by drawing on the notion of negative universality (Kapoor & Zalloua, 2022ab) along with concepts of fantasy, desire and sublime objects (Žižek, 1989), and Rosa (2020) cultural criticism. We illustrate our concepts through the story of a skiing holiday where the fantasy of the perfect snowscape always fails to deliver what it promises. Here, travellers who desire the experience of skiing on ‘perfect snow’ are seduced by powerful advertising campaigns. Due to the unpredictability of nature, travellers are often faced with intrusive snow machines that noisily – and in a ‘vulgar’ way – engineer and manufacture the snowscape which spoils and punctures the fantasy of the perfect skiing conditions. Our paper critically examines the OECD’s (2019b) Learning Compass 2030 document, discussing the universal engineering fantasy that promises to produce certainty, moral improvement and control in/with education. We also analyse the accompanying OECD’s attitudes and values document (OECD, 2019a) that identifies a list of sublime objects such as respect, justice and Bildung to which all countries must aspire if they wish to succeed. We conclude that the policy documents of the OECD present a positive universal engineering fantasy that promises a non-antagonistic and harmonious future. However, such a future will be impossible to achieve. Hence, we call for educators to critically engage with negative universality to expose the lacks and contradictions always inherent in global policies. This would provide educators with an opportunity to reflect on and critically confront seductive policy and its engineering fantasy that captures their desires
MiKAD: Memory-Infused Knowledge Networks for Manufacturing Anomaly Detection
Get PDFMany semiconductor industries have adopted smart manufacturing systems for defect detection, while others still rely on manual inspection methods that can compromise product quality and assurance. To address the limitations of manual inspection, we propose Memory-infused Knowledge Networks for Manufacturing Anomaly Detection (MiKAD), an unsupervised learning technique that combines multiscale knowledge distillation and a dynamic memory bank to detect anomalies of varying sizes and shapes in real industrial image datasets. The knowledge distillation framework consists of a teacher-student architecture, where the teacher is a pretrained network and the student is a trainable network that leverages EfficientNet-B7 as the multiscale backbone. A dynamic memory bank is integrated to support the student network during training by enhancing its ability to learn normal features by updating with new features and suppressing outdated ones. Discrepancy loss at multiple scales between the teacher and student ensures accurate detection and localisation of the anomalies across different sizes and shapes using real industrial datasets. Experiments on two real-world datasets, namely a Seagate Write Pole (WP) and BTAD, demonstrate that MiKAD achieves strong performance in both anomaly detection and localisation, with image level ROC_AUC scores of 97.77% and 96.85% respectively
A Virtual Simulator to Improve Weight-Related Communication Skills for Health Care Professionals: Mixed Methods Pre-Post Pilot Feasibility Study
Get PDFBackground:Discussing weight remains a sensitive and often avoided topic in health care, despite rising prevalence of obesity and calls for earlier, more compassionate interventions. Many health care professionals report inadequate training and low confidence to discuss weight, while patients often describe feeling stigmatized or dismissed. Digital simulation offers a promising route to build communication skills through supporting repeatable and reflective practice in a safe space. VITAL-COMS (Virtual Training and Assessment for Communication Skills) is a novel simulation tool designed to support health care professionals in navigating weight-related conversations with greater understanding and skill.Objective:This study aimed to assess the potential of VITAL-COMS as a digital simulation training tool to improve weight-related communication skills among health care professionals.Methods:A mixed-method feasibility study was conducted online via Zoom (Zoom Video Communications) between January to July 2021, with UK-based nurses, doctors, and dietitians. The intervention comprised educational videos and 2 simulated patient scenarios with real-time verbal interaction. Pre- and posttraining self-assessments of communication skills and conversation length were collected. Participants also completed a feasibility questionnaire. Descriptive statistics were used to analyze the feasibility questionnaire, and open-ended feedback was analyzed using content analysis. Paired-samples t tests were used to assess changes in communication skills and conversation length before and post training.Results:In total, 31 participants completed the study. There was a statistically significant improvement in self-assessed communication skills following training (mean difference=3.9; 95% CI, 2.54‐5.26; t30=−5.76, P=.001, Cohen d=1.03). Mean conversation length increased significantly in both scenarios: in the female patient scenario, from 3.73 (SD 1.36) to 6.08 (SD 2.26) minutes, with a mean difference of 2.35 minutes (95% CI, 1.71‐2.99; t30=7.49, P=.001, Cohen d=1.34); and in the male scenario, from 3.61 (SD 1.12) to 5.65 (SD 1.76) minutes, a mean difference of 2.03 minutes (95% CI, 1.51‐2.55; t30=8.03, P=.001, Cohen d=1.44). Participants rated the simulation positively, with 97% (95% CI 90%‐100%) supporting wider use in health care and 84% (95% CI 71%‐97%) reporting emotional engagement. Content analysis of feedback generated two themes: (1) adapting to this form of learning and (2) recognizing the potential of simulation to support reflective, skills-based training. A minority, 13% (95% CI 1%‐25%) expressed a preference for alternative learning methods.Conclusions:VITAL-COMS was feasible to implement and acceptable to a diverse group of health care professionals. Participants demonstrated significant improvements in self-assessed communication skills and patient-scenario engagement. The simulation was perceived as realistic, emotionally engaging, and well-suited for training in sensitive conversations. These findings support further development and integration of VITAL-COMS into health education programs. Next steps include the translation of the insights identified in this study to inform a tool supported by generative artificial intelligence
Optimizing Heart Attack Detection with Brown-Bear Optimization Algorithm and CNN in Healthcare 4.0
Get PDFHeart disease is a non-communicable decease that lead to death if not treated. In the world most of the people died due to heart disease because they are not treated on time or their disease is not detected at early stage. Due to this efficient heart disease prediction techniques are important for the development of Healthcare 4.0. However, most of the current heart disease detection algorithms are either complex or not optimizes for efficient hyper-parameters. In this context, we proposed a CNN based lightweight heart disease detection framework (trained in 5 epoch). We also used random forest algorithm to identify the most important feature and Brown-Bear Algorithm for optimization of the hyper-parameter of CNN. We also compared the proposed model with current literature and present the efficiency of our proposed framework
From Followers to Fillers: Exploring Transformative Service Research and Ethical Social Media Marketing in the UK Non-Surgical Cosmetic Market
Get PDFThe UK’s non-surgical cosmetic procedures market is booming with year-on-year growth. While social media marketing (SMM) fuels demand, it also raises concerns about consumer safety and wellbeing, in addition to exposure of unregulated practitioners. Limited research has explored ethical marketing from the practitioners’ perspective. This study aims to address this gap, examining SMM techniques and the impact on consumer wellbeing through the lens of transformative service research, providing insights for safer, more informed SMM practices
Comparing Non-Invasive and Fluorescein Tear Break-Up Time in a Pre-Operative Refractive Surgery Population: Implications for Clinical Diagnosis
Get PDF Objectives: Fluorescein break-up time (FBUT) is commonly used to assess tear film stability. However, the instillation of fluorescein destabilises the tear film, impacting validity and clinical applicability, while the subjective nature and variation in volume and concentration reduces repeatability. Non-invasive break-up time (NIBUT) offers an alternative method with less potential bias. Normal tear break-up time is conventionally accepted as 10 seconds (s); however, FBUT is expected to be lower than NIBUT. This study was designed to compare FBUT and NIBUT values in a pre-operative refractive surgery population, where diagnosis of dry eye disease may alter the risk-benefits ratio and contraindicate surgical procedure(s). Improved understanding of the relationship between these two methods will aid appropriate pre-operative patient counselling and consent. Methods: Data from consecutive participants presenting to a private ophthalmology clinic, for initial refractive surgery pre-operative assessment, were analysed. NIBUT and FBUT were performed. Paired and unpaired comparisons were made using the Wilcoxon signed-rank and Mann-Whitney U tests, respectively, and relationships with demographics were explored using Spearman's rank correlation coefficient. Results: Median and interquartile range (IQR) for the first NIBUT was 12.5 s (7.0-18.0 s) and 14.2 s (9.4-18.0 s) for the right and left eyes, respectively. Median and IQR for the average NIBUT was 14.0 s (6.9-18.0 s) and 14.6 s (10.1-18.0 s) for the right and left eyes, respectively. Median and IQR for FBUT was 7 s (5-8 s) and 6 s (5-8 s) for the right and left eyes, respectively. There was a statistically significant difference between NIBUT and FBUT ( p < 0.001). Conclusions: The findings suggest that the commonly used diagnostic threshold of 10 s cannot be uniformly applied to both FBUT and NIBUT, as FBUT systematically underestimates tear stability. </p
Orientation Prediction for Robotic Manipulation: Angle Encoding Strategies for Linear Regression
Get PDFAccurately predicting object orientation from visual information is essential for effective robotic manipulation in smart manufacturing environments, yet standard single stage linear regression approaches struggle with the inherent discontinuity of angular data at the 0°/360° boundary. This paper investigates angle encoding strategies for objects within images to address this challenge, evaluating both trigonometric and vector based representations of sine and cosine components within shallow learning frameworks specifically, Support Vector Regression and Random Forest. Using a curated subset of the MetaGraspNet dataset focused on screwdrivers, we augment and validate orientation annotations through geometric analysis of segmentation masks. Comparative experiments demonstrate that these angular encodings substantially reduce angular prediction error, with Support Vector Regression models employing vector encoding achieving a mean absolute angular error of 5.01°. While all models exhibit increased error under severe occlusion, the results confirm that the proposed encoding strategies can accurately predict object orientation, offering a practical alternative to more complex multi-parameter grasp representations or multi-stage prediction models
Manipulating embryogenesis and testing for potential:Two real problems for the regulation of stem cell-based embryo models
Get PDFStem cell-based human embryo models (SCBEMs), generated in vitro from stem cells, currently exist outside the scope of regulatory frameworks that govern in vitro embryo research in most jurisdictions. A widely discussed proposal suggests using a’Turing test’ framework, whereby regulatory oversight is triggered if an SCBEM is found to be’equivalent’ to a human embryo. In this paper, we argue that such a proposal faces two major complications. First, sophisticated laboratory techniques such as trophoblast replacement allow researchers to manipulate normal embryogenesis, obscuring whether a given SCBEM meets embryo-like regulatory thresholds. Second, attempts to assess SCBEMs’ developmental potential—especially through non-human analogues—rest on tenuous epistemic assumptions that may not align with human-specific developmental trajectories. Given SCBEMs’ potential manipulability and uncertain biological and potentiality benchmarks, we argue that reliance on equivalence-based frameworks alone is highly problematic. We conclude by urging a cautious, flexible approach that recognises both the scientific promise of SCBEMs and the normative need to prevent the circumvention of regulatory safeguards.</p
Cu or Fe‐Exchanged Natural Clinoptilolite as Sustainable Light‐Assisted Catalyst for Water Disinfection at Near Neutral pH
Get PDFNatural zeolites can be used to obtain effective catalysts for heterogeneous photocatalytic reactions due to their low cost and favorable physicochemical properties for water treatment. In this work, a natural clinoptilolite is modified by incorporating iron (NZ–Fe) and copper (NZ–Cu) as compensation cations through ion exchange processes. Metals incorporation and structural stability are demonstrated through X‐ray diffraction, Fourier transform infrared spectroscopy, and scanning electron microscopy. DR‐UV–Vis measurements are used to estimate the bandgap and predict the photocatalytic performance of both materials. Their effectiviness in heterogeneous photocatalytic systems is confirmed by evaluating the inactivation of E. coli as a model pathogen in water. The bacterial detection limit (initial ≈106 CFU/mL) is reached using 1 gL−1 of both catalysts, 100 ppm of H2O2 under visible light (410–710 nm) and near neutral pH in 2 h, with no post‐treatment regrowth observed. Experimental data are analyzed according to the Chick–Watson, Weibull, and Hom disinfection kinetic models. Although more hydroxyl radicals are generated (trapping tests) and less iron leachate is observed for NZ–Fe, good reusability is attained for three disinfection cycles when NZ–Cu is used. This makes copper‐exchanged clinoptilolite a suitable and low‐cost photocatalyst for water disinfection through heterogeneous photo‐Fenton‐type processes
Enhancing Lateral Flow Device Design for Multiplexed Plasmonic Biosensing of Inflammatory Biomarkers
Get PDFThis study investigates the enhancement of plasmonic nanoparticle-based multiplexed inflammatory biomarker detection sensitivity using a novel 3×3 microarray-based lateral flow device (μALFD) design, compared to a traditional single test line format. The biomarkers C-reactive protein (CRP), procalcitonin (PCT), and serum amyloid A (SAA) are critical indicators of inflammation and infection. Our μALFD design demonstrated superior performance, achieving limits of detection (LODs) of 3.25 μg/mL, 0.24 ng/mL, and 1.79 μg/mL for CRP, PCT, and SAA, respectively, representing a significant improvement over the conventional test line design. By incorporating three replicates for each biomarker per μALFD strip, the microarray design improved assay reliability while reducing the risk of non-specific binding through an individual assay flow path for each test spot. These advancements highlight the potential of μALFD technology for developing more sensitive and reliable point-of-care (POC) diagnostic devices, that can provide faster and more accurate clinical decision-making, particularly in resource-limited settings. Future work will explore integrating artificial intelligence (AI) readers for automated quantitative analysis to further enhance diagnostic accuracy and usability