30793 research outputs found
Sort by
SIMAC: A Semantic-Driven Integrated Multimodal Sensing And Communication Framework
Get PDFTraditional unimodal sensing faces limitations in accuracy and capability, and its decoupled implementation with communication systems increases latency in bandwidth-constrained environments. Additionally, single-task-oriented sensing systems fail to address users’ diverse demands. To overcome these challenges, we propose a semantic-driven integrated multimodal sensing and communication (SIMAC) framework. This framework leverages a joint source-channel coding architecture to achieve simultaneous sensing, decoding, and transmission of sensing results. Specifically, SIMAC first introduces a multimodal semantic fusion (MSF) network, which employs two extractors to extract semantic information from radar signals and images, respectively. MSF then applies cross-attention mechanisms to fuse these unimodal features and generate multimodal semantic representations. Secondly, we present a large language model (LLM)-based semantic encoder (LSE), where relevant communication parameters and multimodal semantics are mapped into a unified latent space and input to the LLM, enabling channel-adaptive semantic encoding. Thirdly, a task-oriented sensing semantic decoder (SSD) is proposed, in which different decoded heads are designed according to the specific needs of tasks. Simultaneously, a multi-task learning strategy is introduced to train the SIMAC framework, achieving diverse sensing services. Finally, experimental simulations demonstrate that the proposed framework achieves diverse and higher-accuracy sensing services.The paper was partly funded by Jiangsu Major Project on Fundamental
Research (Grant No.: BK20243059), Gusu Innovation Project (Grant No.:
ZXL2024360), High-Tech District of Suzhou City (Grant No.: RC2025001)
and Natural Science Foundation of China (Grant No. 62132004 and
62301122), the Major Program Project of Xiangjiang Laboratory (Grant No.
XJ2023001 and XJ2022001), and Qiyuan Lab Innovation Fund (Grant No.
2022-JCJQ-LA-001-088)
A method to rapidly match environmental impact data to > 60 dietary datasets
Get PDFThere is growing interest in assessing the environmental impacts of diets due to the awareness of food consumption consequences on human health and ecological systems. A limiting factor in this field has been linking information on the environmental impacts of foods to detailed individual-level dietary data that reflect the food consumption habits of people. Here we present i) a method of linking environmental impact data from a meta-analysis to a food description and classification system; ii) a resulting dataset of environmental impact values matched to 4089 food descriptors; and iii) an example of applying this data to assess the environmental footprints of the Brazilian and USA diets. Our methodology accelerates the assignment of environmental impact values to foods reported in dietary surveys from different countries, and can be used by researchers, policy makers, practitioners and consumers to inform the promotion of healthy diets from sustainable food systems.This research project arose from the N8 AgriFood-funded project, Greenhouse Gas and Dietary choices Open-source Toolkit (GGDOT) hacknights, and was funded through the Science and Technology Facilities Council Global Challenges Research Fund project, Trends in greenhouse gas emissions from Brazilian foods using GGDOT, (ST/S003320/1). XSR was supported through the Brunel University internal Research England Quality-related Global Challenges Research Fund. CR and BT were funded through City, University of London,
Higher Education Innovation Funding (HEIF) “An environmental impact database for Recipe Analysis” (Project ID 512402). CR was also funded by 1) Healthy Soil, Healthy Food, Healthy People (H3) project (Project Reference: BB/V004719/1); 2) UKRI’s International Science Partnerships Fund (ISPF), part of the Co Centre
for Sustainable Food Systems, Grant number BB/YO12909/1; 3) funded by UKRI and NIHR as part of the Building A Green Future strategy the THRIVING Food Futures research hub (MR/Z506485/1). JTS, VPQ, AB, and BAH were funded by the Bill and Melinda Gates Foundation, grant number OPP1189436/INV-010508,
and the Food and Agriculture Organization of the United Nations
Environmental and economic life cycle sustainability assessment of reusable versus single-use anaesthetic face masks
No full textData availability:
All data used have been displayed and explained in the manuscript and in the suplementary information.In the United Kingdom, healthcare products and services contribute 62 % of the National Health Service's greenhouse gas emissions. One proposal to reduce this impact is by replacing single-use devices (SUDs) with reusable devices. This study employs life cycle assessment (LCA) and life cycle costing (LCC) methodologies to assess the environmental and economic sustainability of a reusable anaesthetic mask made primarily of Polychloroprene and Polyisoprene; and two single-use masks one made of polyvinyl chloride (PVC) and one of thermoplastic elastomer and polypropylene (TPE + PP). The reusable mask is shown to be cheaper and have lower environmental impact compared to the PVC single-use mask for nine of the 11 impact categories, including GWP, but has lower environmental impact than the TPE + PP single-use mask for only three categories (HTP, MAETP, and FAETP). The major contributor of the reusable mask's impact is the reprocessing stage, which represents over 70 % of all impact categories. The LCC showed PVC single mask to have the greatest cost (£5.89) compared to TPE + PP mask (£4.99) and the reusable mask (£4.44). Sensitivity and scenario analyses showed that the number of reprocessing cycles greatly influences the sustainability of the reusable mask when the number of reuses was less than 14 and that the energy consumption of the reprocessing machinery had a noticeable influence on the reusable mask's overall environmental impact. In conclusion, to make reusable masks a favourable option, manufacturers and health providers need to optimise the energy and packaging used in the reprocessing stage, together with ensuring that reusing practices i.e. minimum of cycles, are identified and communicated.PhD studentship from Brunel University of London, sponsored by a medical device manufacturer
Deep learning algorithms for complex traffic behaviour recognition in mono-camera intelligent vehicles
Get PDFThis thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University LondonIntelligent vehicles have the potential to revolutionise transport by enhancing safety,
reducing congestion, and improving efficiency. A critical component of IVs is perception—
the ability to accurately interpret the environment for informed driving
decisions. While multi-sensor systems are widely used, camera-only solutions offer a
cost-effective alternative but face challenges in recognising the intentions of diverse
traffic agents in complex urban environments. This thesis addresses limitations in
current mono-camera approaches, which often overlook explicit behavioural cues
and lack integrated pipelines for discrete intention behaviour recognition.
A literature review identified key gaps, including the absence of methods for predicting
discrete intention behaviours across various traffic agents, insufficient public
datasets capturing complex urban scenes for intention prediction, and underutilisation
of explicit cues such as vehicle light signals and object orientation. Additionally,
existing studies rarely consider integrated pipelines that combine detection, tracking,
and behaviour recognition while accounting for error propagation across stages.
To address these gaps, a monocular traffic hazard dataset was developed, capturing
diverse traffic agents and explicit behavioural cues relevant for hazard recognition.
Deep learning models, including Vision Transformers and Convolutional Networks,
were designed to leverage these features, demonstrating improved accuracy
in recognising complex traffic behaviours from single images. Experiments exploring
different input features, observation horizons, and class granularities revealed that
combining explicit and implicit cues enhances recognition performance.
A complete hazard recognition pipeline was implemented, integrating detection,
tracking, and behaviour recognition to assess system-level performance. Results
highlighted the challenges of error propagation across modules while demonstrating
the feasibility of end-to-end monocular pipelines for complex traffic behaviour
recognition.
The key contributions of this thesis include the development of a targeted monocular
dataset for behaviour recognition, the creation of models utilising underexplored
visual cues, and the integration of these models into a unified hazard recognition
pipeline for camera-only IV systems. This work demonstrates the potential
of monocular approaches for traffic behaviour recognition and provides a foundation
for cost-effective, scalable intelligent vehicle solutions. Future research should focus
on expanding dataset diversity, improving model robustness, and incorporating
multi-agent interactions to enhance real-world applicability.UK Research and Innovation and the EPSRC DTP Scholarshi
AMSTAR-PF: a critical appraisal tool for systematic reviews of prognostic factor studies
No full textSummary Points:
• Research into prognostic factors is vital for many areas of healthcare.
• Confidence in the findings of systematic reviews of prognostic factor research can be compromised in a variety of ways.
• AMSTAR-PF (A MeaSurement Tool to Assess systematic Reviews of Prognostic Factor studies), based on AMSTAR 2, has been developed, refined, and tested AMSTAR-PF uses signalling points and 19 questions over 14 domains to assist in coming to an overall judgment of confidence in the results of the review.
• Providing a standardised and reliable tool to appraise review quality will assist users to ascertain confidence in the review’s findings.Data availability statement:
Data related to agreement testing can be found in the referenced paper, or on reasonable request from the corresponding author at [email protected]. The draft versions of the tool and guidance notes generated during the development process are available on reasonable request, from the corresponding author.A preprint version of the article is available at medRxiv, https://www.medrxiv.org/content/10.1101/2025.04.10.25325555v1 . It has not been certified by peer review. It reports new medical research that has yet to be evaluated and so should not be used to guide clinical practice. The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license.The ability to predict the onset or natural history of an illness, or how people may respond to a treatment, guides clinical decision making. These predictions are commonly based on prognostic factors: clinical, patient, or societal variables that are identified as being predictive of a certain future outcome. Prognostic factor research has increased across fields, with a subsequent increase in the number of systematic reviews of prognostic factors studies. Understanding the quality of such prognostic factor reviews is essential for confidence in their findings, but there is no quality appraisal instrument to specifically assess systematic reviews of prognostic factor studies. A MeaSurement Tool to Assess systematic Reviews of Prognostic Factor studies, AMSTAR-PF, has been developed to fill this gap.This work received no specific funding. MLH was supported by an Australian Government Research Training Scholarship. RDR was supported by funding from the MRC Better Methods Better Research panel (grant MR/V038168/1) and by the National Institute for Health and Care Research (NIHR) Birmingham Biomedical Research Centre at the University Hospitals Birmingham NHS Foundation Trust and the University of Birmingham. GLM was supported by a leadership investigator grant from the National Health and Medical Research Council of Australia (ID 1178444). The funders had no role in considering the study design or in the collection, analysis, interpretation of data, writing of the report, or decision to submit the article for publication
Valuation of EQ-5D Health States for Adults in Low-, Lower-Middle, and Upper-Middle-Income Countries: A Systematic Review
Get PDFData Availability: All data are publicly available and summarized within the article and supplemental files.Supplemental Material is available online at: https://www.valuehealthregionalissues.com/article/S2212-1099(25)00431-5/fulltext#supplementary-material .Objectives:
Preference-based measurement of health-related quality of life is crucial for informing resource allocation decisions, with the EQ-5D instrument widely used as a measure of health-related quality of life. Although country-specific value sets are well established in many high-income countries, current summarized evidence from valuation studies in low- and middle-income countries (LMICs) remains limited. This review systematically identified EQ-5D valuation studies in LMICs, summarized methodologies and scoring algorithms by country type, and highlighted key challenges.
Methods:
A systematic search was undertaken across 7 academic databases and the EuroQol website. Two independent reviewers screened titles and abstracts and performed full-text reviews and data extraction. Reporting followed Checklist Reporting Valuation Studies of Multi-Attribute Utility-Based Instruments for quality assessment. The synthesis included study characteristics, methodologies, and summarized scoring algorithms from the best-performing models, highlighting variations across countries.
Results:
Through screening 9378 studies, 35 studies from 22 LMICs were included. Of these, 20 (58%) were from upper-middle-income countries, whereas low-middle and low-income countries accounted for 13 (37%) and 2 (6%) studies, respectively. Eighteen (51%) studies reported EQ-5D-5L valuations. Sample sizes ranged from 148 to 5503, with the time trade-off method being predominant. Scoring algorithms showed no significant variation between upper-middle- and low-middle-income countries, except for the pain/discomfort dimension in EQ-5D-5L. Mobility was the most reported utility decrement among studies.
Conclusions:
There is a growing trend in developing country-specific value sets in LMICs. Contextually relevant designs and adequate pilot studies could enhance the accuracy of value sets in culturally diverse settings, particularly where severe health states are commonly reported.Marufa Sultana’s time is supported by the Alfred Deakin Postdoctoral Research Fellowship by Deakin University. No additional funding was received to conduct this research
Search for a heavy pseudoscalar Higgs boson decaying to a 125 GeV Higgs boson and a Z boson in final states with two tau and two light leptons in proton-proton collisions at Search for a heavy pseudoscalar Higgs boson decaying to a 125 GeV Higgs boson and a Z boson in final states with two tau and two light leptons in proton-proton collisions at √ = 13 TeV
No full textA version of the article is available at arXiv:2501.14825v2 [hep-ex] (https://arxiv.org/abs/2501.14825v2). Comments: Replaced with the published version. Added the journal reference and the DOI. All the figures and tables can be found at https://cms-results.web.cern.ch/cms-results/public-results/publications/HIG-22-004 (CMS Public Pages). Report number: CMS-HIG-22-004, CERN-EP-2024-313. Journal reference: JHEP 10 (2025) 074. Submission history: From: The CMS Collaboration: [v1] Tue, 21 Jan 2025 22:46:14 UTC (486 KB); [v2] Wed, 15 Oct 2025 16:32:06 UTC (489 KB).A search for a heavy pseudoscalar Higgs boson, A, decaying to a 125 GeV Higgs boson h and a Z boson is presented. The h boson is identified via its decay to a pair of tau leptons, while the Z boson is identified via its decay to a pair of electrons or muons. The search targets the production of the A boson via the gluon-gluon fusion process, gg → A, and in association with bottom quarks, bb̅A. The analysis uses a data sample corresponding to an integrated luminosity of 138 fb⁻¹ collected with the CMS detector at the CERN LHC in proton-proton collisions at a centre-of-mass energy of √ = 13 TeV. Constraints are set on the product of the cross sections of the A production mechanisms and the A → Zh decay branching fraction. The observed (expected) upper limit at 95% confidence level ranges from 0.049 (0.060) pb to 1.02 (0.79) pb for the gg → A process and from 0.053 (0.059) pb to 0.79 (0.61) pb for the bb̅A process in the probed range of the A boson mass, mA, from 225 GeV to 1 TeV. The results of the search are used to constrain parameters within the Mh,EFT125 benchmark scenario of the minimal supersymmetric extension of the standard model. Values of tan β below 2.2 are excluded in this scenario at 95% confidence level for all mA values in the range from 225 to 350 GeV.SCOAP³
Digital transformation or digital divide? SMEs' use of AI during global crisis
Get PDFHighlights:
• Explores how AI-driven digital shifts affect SMEs amid global crises.
• Highlights risks of widening digital divide despite leading edge technological adoption.
• Identifies how technological, organisational, and external factors shape SME digital transformation.
• Offers insights to foster inclusive digital transformation among SMEs.Data availability:
Data will be made available on request.There is a notable deficiency in empirical research examining how digital transformation, particularly through the adoption of technologies such as Artificial Intelligence (AI), can empower SMEs to navigate the challenges posed by global crises. While new technologies are frequently regarded as solutions, they can coexist with existing systems and potentially exacerbate the digital divide. This paper aims to investigate the factors that either facilitate or hinder digital transformation among SMEs and their impact on either bridging or widening the digital divide. To explore this issue, we employed a maximum variation sampling method and conducted 15 in-depth interviews with SMEs in the restaurant sector – encompassing small, medium, and large establishments – in London. Our findings indicate that technological factors (such as digital inertia), organisational aspects (including perceptions and collaboration), and external environmental influences (like lockdown measures) are critical in determining the extent of digital transformation or the continuation of the digital divide among SMEs during the pandemic. This paper contributes to understanding the interrelationships among innovative technologies, including AI, SMEs, and their external environments in the context of rapid digital transformation. It underscores the adverse effects of technology adoption on the equitable distribution of innovative practices and provides insights into promoting inclusive digital transformation to mitigate the digital divide among SMEs.Funding Statement: Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2025R727), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
A UNIFIED THEORY FOR ARMA MODELS WITH VARYING COEFFICIENTS: ONE SOLUTION FITS ALL
Get PDFA preprint version of the article is available at: arXiv:2110.06168v1 [math.ST], https://arxiv.org/abs/2110.06168 under a CC BY licence. It has not been certified by peer review.A new explicit solution representation is provided for ARMA recursions with drift and either deterministically or stochastically varying coefficients. It is expressed in terms of the determinants of banded Hessenberg matrices and, as such, is an explicit function of the coefficients. In addition to computational efficiency, the proposed solution provides a more explicit analysis of the fundamental properties of such processes, including their Wold–Cramér decomposition, their covariance structure, and their asymptotic stability and efficiency. Explicit formulae for optimal linear forecasts based either on finite or infinite sequences of past observations are provided. The practical significance of the theoretical results in this work is illustrated with an application to U.S. inflation data. The main finding is that inflation persistence increased after 1976, whereas from 1986 onward, the persistence declines and stabilizes to even lower levels than the pre-1976 period.Magdalinos gratefully acknowledges financial support by the British Academy: grant SRG2324\241667. Alessandra Canepa acknowledges financial support under the National Recovery and Resilience Plan (NRRP), Mission 4, Component 2, Investment 1.1, Call for tender No. 104 published on February 2, 2022 by the Italian Ministry of University and Research (MUR), funded by the European Union—NextGenerationEU—Project Title 20223725WE—Methodological and computational issues in large-scale time series models for economics and finance – CUP J53D23003960006—Grant Assignment Decree No. 967 adopted on June 30, 2023 by the Italian Ministry of Ministry of University and Research (MUR)
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇