78146 research outputs found
Sort by
Polynomial matrix SVD via generalized sequential matrix diagonalization
Get PDFThe singular value decomposition (SVD) of polynomial matrices serves as a cornerstone in the analysis and optimization of broadband multi-input multi-output (MIMO) systems. This paper introduces novel algorithms for performing the SVD of polynomial matrices, leveraging a sequential matrix diagonalization (SMD) framework. The proposed methodology begins by identifying the column or row with the highest off-diagonal energy using a maximum search procedure. Subsequently, this energy is transferred to the zero-lag coefficient matrix through a delay operation, which is then diagonalized using a conventional SVD. This iterative process continues until the maximum off-diagonal element falls below a predefined threshold. The proposed framework encompasses multiple algorithmic variants, each designed to offer distinct convergence speeds, thereby addressing diverse computational and accuracy requirements. Rigorous proofs of convergence are provided, alongside a thorough comparative analysis of the computational efficiency and diagonalization accuracy of the algorithms. Extensive simulations, conducted on ensembles of randomly generated polynomial matrices, demonstrate that the proposed algorithms consistently outperform state-of-the-art polynomial SVD (PSVD) methods across all evaluated performance metrics. Furthermore, the application of the proposed algorithm to decouple broadband or convolutive MIMO channels validates its accuracy and effectiveness in practical scenarios
Cross-cultural adaptation and psychometric evaluation of the DoFEL-R for early detection of Alzheimer's related functional decline in Croatia
Get PDFBackground and Aims: Early-stage Alzheimer's disease (AD) is often associated with subtle functional changes that may go undetected by conventional assessment tools. The Details of Functions of Everyday Life – Revised (DoFEL-R) was developed to address this gap by embedding updated theoretical understandings from memory binding into the assessment of everyday tasks. This study aimed to translate, cross-culturally adapt, and conduct a preliminary validation of the DoFEL-R for use with Croatian-speaking older adults. Methods: Following the guidelines of the International Test Commission (ITC), the adaptation process included forward and back translation, a Delphi expert panel, cognitive interviews, and pilot testing. A total of 263 community-dwelling older adults completed the Croatian DoFEL-R, with fifty-six participating in a 2-week follow-up to assess test-retest reliability. Results: Exploratory factor analysis supported a two-factor structure aligned with relational and conjunctive binding, consistent with the original theoretical framework. The scale showed strong internal consistency (Cronbach's α = 0.89) and excellent test-retest reliability (ICC = 0.91). Cognitive interviews indicated good face validity and cultural appropriateness of the adapted items. Conclusion: These findings support the Croatian DoFEL-R as a valid and reliable tool for identifying early functional decline indicative of AD-related dementia in older adults. Its use may enhance early detection of preclinical AD and inform clinical assessment practices in Croatia and similar cultural contexts. However, the sample's relatively high educational and digital literacy levels may limit generalisability, and further research is needed to confirm applicability across more diverse populations
Assessment of spaceborne SAR micro-motion measurement for vibration-based SHM
No full textRemote sensing methods for vibration-based structural health monitoring (VBSHM) are gaining interest given the benefits of non-contact sensing for this application: broader spatial coverage, easier measurement access to structures such as bridges, and cost savings from reducing the need for embedded sensor networks. Vibration measurement using micro-Doppler synthetic aperture radar (MDSAR) is an emerging technique enabling the measurement of structural vibrations from synthetic aperture radar (SAR) satellite data, exploiting recent advancements in high-resolution X-band radar imaging. This paper presents a detailed metrological assessment of MDSAR–using processing based on the pre-existing subpixel offset tracking (SPOT) concept as a representative method–to evaluate its applicability to VBSHM. Measurements are made from real SAR images of vibrating targets with synchronous ground truth data. The results verify that the technique can be used to measure target motion time-histories with dynamics relevant to SHM, namely different modulations with frequencies from 1 to 4 Hz, and radial RMS displacement amplitudes from 10.43 to 0.10 mm. These results pave the way for future measurements of infrastructure
Association of apolipoprotein E variants on Alzheimer's disease in Latin America : A systematic review and meta‐analysis
No full textThe apolipoprotein E (APOE) ε4 allele represents the strongest genetic risk factor for Alzheimer's disease (AD), but its role in genetically diverse Latin American and Caribbean (LAC) populations is underexplored. We conducted a meta‐analysis of 35 studies from 11 LAC countries, encompassing 3206 patients with AD and 5515 controls. The ε4 allele demonstrated significant association with increased AD risk (odds ratio [OR] = 3.25, 95% confidence interval [2.82–3.76]), while ε3 showed lower odds (0.42, [0.37–0.48]). Homozygous ε4/ε4 carriers had elevated risk (6.84, [5.09–9.19]), and heterozygous ε3/ε4 carriers showed moderate risk (2.59, [2.31–2.91]). Country‐level analyses revealed variability, with Ecuador showing the highest OR for ε4/ε4 (13.29, [1.56–113.4]). These results confirm APOE ε4 as a major AD risk factor in LAC populations and highlight regional differences relevant to precision medicine
Intelligent operator-independent flexibility in robotic non-destructive evaluation
No full textThe non-destructive evaluation (NDE) of high-value components is commonly regarded as a bottleneck in the manufacturing industry. Even though robotic automation is leveraged these days to accelerate the process, productivity is still hindered by the limitations of robotic and automated systems requiring intensive calibrations and precise fixturing of inspection components. This restricts the systems’ flexibility and ability to adapt to uncertainties in part geometry, inspection process or changing environments. This study presents a system that is both flexible and adaptable, integrating 3D vision sensing and force-position control with the human-collaborative KUKA LBR robot. The system is designed for the geometry-agnostic, precise delivery of ultrasonic NDE sensors, enabling the inspection of materials with varying geometries and surface profiles. The developed system demonstrated its efficacy by providing the user with the ability to inspect a given workpiece at a random pose without having prior knowledge of the component. Pose independence was proven with multiple scans producing consistent results, achieving high inspection sensitivity via detecting 100% of the fabricated defects in the scan region. This study has succeeded in investigating the suitability of flexible human-collaborative manipulators for NDE and showcases the successful integration of an adaptable scanning system capable of identifying and scanning varying geometries regardless of pose while producing high-quality scans consistently. The study achieved a significant time savings of ≈97%, reducing setup and processing time from 35 to just 1 min
A qualitative exploration of stakeholders' views on vaccines for the elderly in South Africa
Get PDFBackground and objectives Advancements in healthcare and living standards have led to increased life expectancy globally, including in South Africa. Despite considerable evidence from high-income countries that vaccination is a cost-effective strategy for healthy ageing, South Africa currently does not have a formal policy for vaccination of the elderly. This study explored the opinions, experiences and recommendations of South African stakeholders (experts and key opinion leaders) in the field of vaccines and healthy ageing. Methodology Online qualitative in-depth interviews with 16 stakeholders (vaccinologists, infectious disease specialists, policymakers, geriatricians and epidemiologists) were conducted. Interviews were recorded, transcribed, coded and analysed thematically using NVivo 12.0™ software. Results The overarching unanimous theme was that a national immunisation programme for the elderly (NIPE) is warranted. Aligned to this were the themes: (i) Immunisation schedule for the elderly; (ii) Health system NIPE readiness; and (iii) Strategies ensuring adequate vaccine uptake by the elderly. Participants recommended a coordinated NIPE, which could build on the successes of the Expanded Programme on Immunisation of South Africa (EPI-SA) and the recent COVID-19 vaccination programme. The NIPE should ideally be integrated into the existing primary healthcare (PHC) system programmes for the elderly. Concurrently, interventions to increase access to vaccines and reduce vaccine hesitancy among the elderly and their healthcare providers must be implemented. Conclusion A NIPE needs to be prioritised, understood, communicated, and implemented within EPI-SA and integrated into the PHC system. This should be accompanied by the requisite logistical and financial support for such a programme. The NIPE should build on the success and achievements of EPI-SA and the COVID-19 vaccination programme. Interventions that increase vaccine access and address vaccine hesitancy should also be implemented to enhance uptake by the elderly
No language left behind? : Towards an integrated framework for linguistic rights, human rights and technology regulation
Get PDFRecent technological advances have resulted in a rapid increase in the use of generative artificial intelligence (AI). These tools and applications create new content using statistical models trained by (online) databases (Aydın & Karaarslan, 2023). This content is increasingly difficult to distinguish from that created by humans: blurring the lines between the digital and real worlds (Ferrara, 2024). These developments have been recognised as creating opportunities to improve human rights, ‘including access to information, health, education, and public services’ whilst, at the same time, there have been significant concerns raised that AI has the potential to also ‘dramatically intensify online harms’ (United Nations, 2023). This has resulted in the United Nations AI Advisory Body (2023) calling for AI to be used for the benefit of all. Initiatives to regulate AI at supra-national level, for example, the EU AI Act, have focussed on the risk they pose to either individual or State values (Fink, 2021), or focussed on governance (United Nations AI Advisory Body, 2023) but have not explicitly considered the way these impact on individuals use technology to communicate, with each other, the world around them, and also with the tools and technologies, in particular how these impact on linguistic and cultural diversity These regulatory or advisory frameworks have, both implicitly and explicitly, acknowledged that access and provision to AI technologies is not universal, and that the largest growth has been in a high- or middle-income economy countries and among populations with a high level of English-proficiency (Liu & Wang, 2024). This is, perhaps, not surprising: current digital tools and online spaces are significantly biased towards users of English – with this being the main language used on 63% of all websites and a further 9 languages making up 75% of the internet (The Centre for Internet and Society et al., 2022). This dominance of a small number of big languages in online spaces, with significant bias towards large and mostly Western-European, languages (Toupin et al., 2021), is contributing to wider notions of digital imperialism and colonisation (Jin, 2013) as individuals are only being able to access or use these tools and technologies in a language that is supported(Pimienta, 2022). The digitally under-represented languages are often already considered ‘endangered’ or at risk of disappearing as a community language, but digital imperialism also affects small(er) state languages which, measured by conventional language vitality criteria, are currently considered to be ‘safe’ (Rehm & Uszkoreit, 2013). This article will introduce a new framework for the assessment of digital language vitality in the AI era, which will be used to identify high-level policy recommendations to support users of languages that are currently digitally disadvantaged in ensuring equity of access which is culturally and linguistically appropriate, situated within human rights legislation and in particular the frameworks to protect minorities within the European context
Neural networks for faster laser ultrasound tomography in tissue phantoms
No full textSpeed of sound (SoS) mapping provides quantitative and localised information about a material’s acoustic properties, allowing identification of spatial compositional changes. In biomedical applications, SoS variations can inform tissue characterisation or improve image reconstruction algorithms that typically assume a constant SoS. However, conventional time-of-flight (ToF) tomography methods remain computationally intensive. This study presents experimentally derived tomographic reconstructions of SoS maps of heterogeneous structures from all-optically acquired data using a convolutional neural network (CNN). The CNN, trained on simulated data, enables near real-time, high-quality tomographic reconstructions. The novelty of this work lies in the integration of a laser ultrasound (LU) data acquisition setup with a CNN-based reconstruction approach, demonstrating its potential for remote and flexible inspection of biomedically relevant samples. The CNN was trained using simulated data based on ultrasonic wave propagation models and achieved tomographic reconstructions of a 77 mm × 77 mm area in less than 6 ms. Data were acquired from four tissue-mimicking phantoms (30 mm diameter) with inclusions of varying size (minimum 6 mm diameter) and SoS (minimum variation 25 m/s). When compared with published, in vivo studies using mammography (MM), conventional ultrasound, and magnetic resonance imaging (MRI), the proposed method yielded 5.73% mean sizing error for phantoms and inclusions relative to the ground truth, highlighting the clinical potential of the LU-CNN framework and the need for further in vivo testing. These findings underscore the method’s potential as a precise, faster alternative to conventional imaging and reconstruction methods used in clinical practice
Reflections on a reimagined future for consumer research
Get PDFThis paper is born of the joy and pain of institutional life. It responds to recent work by Craig Thompson and is a response to an invitation to participate from the Editor of the JMM (Mark Tadajewski). It seeks to take seriously the role of consumer research and consumption within society. It seeks to tell ‘better stories’ about the past, present and what it sees as a reimagined future for consumer research. To do so, it revisits some of the canonical works of consumer research, such exemplary work it suggests are worthy of revisiting and critiquing if we are to chart a rejuvenated form of consumer research that takes seriously our contemporary predicament and is more critical and incisive in its intent. With renewed purpose it takes on board the insights around Bauman’s notion of ‘liquid times’, offers a range of opportunities, possibilities and dangers around the conflicted consumer self and puts forward an ABC of critical consumer research, where notions of affect, breakdown, contradiction are taken seriously for their theoretical and practical insights
Ambulatory tasks and journeys : a framework for free-living behaviour
Get PDFBackground: Standard accelerometer summaries obscure meaningful differences in how people move while upright. We introduce an operational two-class Ambulatory Behaviour Framework that separates Ambulatory Tasks—periods of standing and short continuous stepping bouts (<1 min) that are indicative of activity in a single locus—from Ambulatory Journeys—long continuous stepping bouts (≥1 min) that are indicative of movement between locations. Methods: We analysed thigh-worn activPAL3 data from 3545 participants in the age-46 sweep of the 1970 British Cohort Study (24,815 valid monitor-days). Event-based algorithms grouped upright events and classified them as an Ambulatory Task or Journey; linear models examined associations with sitting time and differences by sex and BMI. Results: Mean upright time averaged 6.50 h day−1; Ambulatory Tasks dominated (5.91 h; 90.6% of upright exposure), whereas Ambulatory Journeys contributed 0.61 h (9.4%). Each additional hour of Ambulatory Tasks corresponded to 0.61 h less sitting (β = −0.61 h; 95% CI: −0.63 to −0.61), while an extra hour of Ambulatory Journeys displaced only 0.04 h of sitting (β = −0.04 h; 95% CI: −0.044 to −0.039). Women accumulated significantly more time in Ambulatory Tasks and less sitting time than men. Both upright behaviours declined with increasing BMI. Conclusions: Ambulatory Tasks substantially replace sitting time, whereas Ambulatory Journeys leave sitting essentially unchanged. Interventions to displace sitting should concentrate on fostering frequent, brief, context-embedded tasks throughout the day. This novel framework yields interpretable, sensor-agnostic metrics to target behaviour change and standardise reporting of free-living mobility
Access Repository Dashboard
Do you manage University of Strathclyde Institutional Repository? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!