36270 research outputs found
Sort by
A Critical Examination of a Leading Paradigm of Innovation Management: A Study of Frugal Innovation in the Agri-food Sector in Sub-Saharan Africa and its Role in Furthering Sustainable Rural Development
No full textThis thesis provides a critical examination of a leading concept in the field of innovation management, frugal innovation, and it further explores its significance in the context of the agri-food sector in sub-Saharan Africa and specifically regarding the smallholder farmer. In developing an original contribution, this research extends the analysis of the frugal innovation literature from the more prevalent focus upon cases and examples from the Indian sub-continent and expands the coverage to sub-Saharan Africa. The research also explores the agri-food sector as a suitable setting for research, and the cases used provide more nuance to help develop a coherent definition of frugal innovation and in helping to refine a concept that has been described as incomplete. The findings of the research suggest that frugal innovation is prevalent within the agri-food sector and across sub-Saharan Africa and that the cases identified also reflect the idea of frugal innovation in terms of products, services, systems, and business models that have as their essential elements, cost reduction, core functionalities, and optimized performance appropriate to the setting. Further, the topic of sustainability is a key area of focus for government and policymakers, and so this thesis also suggests that frugal innovation provides an appropriate approach for sustainable development in rural areas and aligns with the achievement of the United Nations Sustainable Development Goals (SDG). This represents a hitherto underexplored aspect of the existing literature and particularly regarding the smallholder farmer. The resulting cases help support the evolving canon of literature within the field of frugal innovation and serve to widen the context within which that research is conducted to a different context in terms of the agri-food sector of sub-Saharan Africa while also developing a further original contribution in terms of the role of frugal innovation in sustainable rural development.</p
Advancing Deep Learning in Medical Imaging Diagnostics under Small-Scale Dataset Constraints
No full textDeep learning has brought clear gains to medical image analysis, but most clinical datasets remain small and heterogeneous, which often leads to overfitting, unstable training, and unclear design choices for the model architecture and training strategy. This thesis focuses on small-scale dataset constraints in medical imaging diagnostics and proposes both a theoretical view and three concrete methods for multiple sclerosis (MS) and brain tumour classification. First, we introduce the Modelling Small-Scale Dataset Challenges Framework (MSSDCF), which links three core factors—data scale, model capacity, and optimisation behaviour—and explains why performance degrades when these factors are not matched. MSSDCF also offers three practical perspectives (sample, network, and search strategy) that guide the design of later methods. From the sample perspective, we propose Texture Embedded Multi-Stage Transfer Learning (TE-MSTL) for MS MRI classification. TE-MSTL embeds Gray-Level Co-occurrence Matrix (GLCM) texture images into a three-stage transfer pipeline. On a public MS MRI dataset, TE-MSTL improves the 10-fold cross-validated accuracy of a 6-layer ConvNet from 95.82% to 97.46% and raises the Matthews correlation coefficient from 91.65% to 93.02%. From the network perspective, we design RanMerFormer, a Vision Transformer with token merging and a random vector functional link head for brain tumour MRI. On a four-class public brain tumour dataset, RanMerFormer increases accuracy from 98.90% (ResNet152V2 / MobileNet + M-SVM) to 99.80%, while reducing FLOPs to about 70% of ResNet152V2 (8.2 G vs. 11.6 G) and remaining feasible on a single GPU. From the search strategy perspective, we propose PDAC-BTNet, a plane-decoupled adaptive curriculum network that combines MobileNetV2 with plane-specific curricula. On the same brain tumour dataset, PDAC-BTNet reaches 99.39% accuracy and 99.80% specificity with only 3.5 M parameters and 0.59 G FLOPs, and improves accuracy over RanMerFormer by 0.52 and 1.32 percentage points in two simulated imbalance scenarios. Together, these results show that the MSSDCF-guided designs can give consistent and quantifiable gains on small-scale brain MRI benchmarks while keeping model and computational costs under control.</p
Phages and their ecological strategies to advance phage therapy
No full textThe rise of antimicrobial resistance is a major global threat, driving the need for alternative approaches such as using bacteriophages (bacterial viruses). Bacteriophages are naturally occurring, diverse, and abundant, and have a plethora of unique strategies for hijacking bacterial transcription. However, our understanding of the traits beneficial to effective phage takeover is limited. Here, I aim to address this by applying an ecological framework (Grime’s CSR triangle) to phages using transcriptional profiling as a proxy for ecological strategy. I combined these data with phenotypic and genomic data to improve our understanding of what functional traits contribute to effective phage takeover. I focus on a successful urinary tract phage cocktail targeting Escherichia coli and Shigella sonnei, both major healthcare-associated pathogens. I first screened a large collection of clinical bloodstream isolates to assess sensitivity and distinguish ecological specialists from generalists. I then extensively characterised the phenotypic characters that define the lifecycles and stability of urinary tract phages, identifying key parameters that I could correlate with transcriptomic data and performance in urinary tract-like environments. I present the first transcriptionally profiled Phapecoctaivrus phage, which revealed that most genes were expressed mid-infection. I also profiled three Tequatrovirus phages and showed that they have a more generalist approach, with similar amounts of genes maximally expressed throughout their infection cycle. The core genes observed revealed strong temporal conservation, with 92% being expressed in the same phase. The transcriptomic response of the infected bacterial cells revealed metabolic shifts and prophage gene induction during infection. These data highlight takeover strategies and challenges for therapeutic application. Together, these results support the idea of a transcription-based ecological framework that links phage traits and bacterial response to therapeutic performance. This ecological framework provides a foundation for selecting ecologically informed cocktails, advancing optimal phage trait selection, and contributing to the broader development of phage therapy against antimicrobial resistance.</p
Evaluation and implementation planning of an online self-management resource for people with chronic kidney disease: a mixed methods study
No full textPeople with chronic kidney disease (CKD) face increased health risks, including cardiovascular complications and progression to end-stage kidney disease requiring kidney replacement therapy. Enhancing self-management behaviours is vital to delaying disease progression and improving outcomes. Previously, CKD care in the UK has lacked resources for non-dialysis patients; however, recent efforts have focused on developing supportive tools to address this gap. One such resource, ‘My Kidneys & Me’ (MK&M), offers potential support, but little is known about its adoption and integration into routine healthcare.Aims: This study explored healthcare professionals’ and providers (HCPs) perspectives on CKD self-management and how MK&M could be embedded into clinical workflows. It also identified priority outcomes for CKD self-management and examined MK&M’s use in primary care.Methods:• Qualitative interviews: Semi-structured interviews with HCPs examined barriers, facilitators, and requirements for adopting self-management tools. Thematic analysis was applied.• Delphi survey: A three-round online survey involving patients, caregivers, and professionals identified key CKD self-management outcomes using content and descriptive analysis.• Pilot implementation: MK&M was piloted in a Primary Care Network with people living with CKD stages 3-4. Sociodemographic data, staff surveys, and interviews assessed feasibility and perceptions.Results: Interviews highlighted the need for CKD education and support. Barriers included organisational constraints; enablers involved stakeholder awareness and guidance. The Delphi survey prioritised outcomes in ‘clinical’ and ‘knowledge, skills and confidence’ domains. The pilot showed MK&M could align with existing CKD pathways, though contextual factors must be addressed for broader implementation.Discussion: MK&M shows potential as a self-management support resource for people living with CKD. However, refinements in design and delivery are needed to enhance uptake. This research informs future implementation strategies by highlighting HCP perspectives and key outcome priorities.The original contribution within this thesis includes: understanding kidney professionals' views on self-management in kidney disease; this study was also the first to systematically identify the key outcomes to focus on for self-management; the data showed how an implementation trial could potentially be conducted when the online self-management resource was adopted in clinical practice.</p
Investigating the Acute Effects of Histone Deacetylase 1 Degradation in Mouse Embryonic Stem Cells
No full textHistone deacetylase (HDACs) enzymes remove acetyl-moieties from post-translationally modified lysine residues on histones, resulting in chromatin compaction and gene silencing. The Class I HDACs, HDAC1 and HDAC2, are well-characterised and are largely redundant. Genetic ablation experiments have, however, shown that HDAC1 is uniquely required in early mouse embryonic development since deletion of Hdac1 but not Hdac2 is lethal by embryonic day 10.5. Although current genetic manipulation techniques have provided some answers, they are limited by slow depletion rates and downstream effects diluting the primary impacts of HDAC1-degradation. Herein, the degradation tag (dTAG) system has been leveraged to achieve rapid (within 2 hours) degradation of HDAC1 enabling the immediate effects of HDAC1 removal to be investigated.As anticipated, HDAC1 degradation resulted predominantly in the upregulation of transcription within 2 hours, with progressive increases in downregulated genes over time. In particular, pluripotency-associated factors and ribosome biogenesis genes were downregulated, while circadian rhythm and nervous system development genes were upregulated within 2 and 6 hours of HDAC1 depletion. ATAC-seq demonstrated that dysregulated gene expression was accompanied by widespread alterations in chromatin accessibility, with an unexpected gradual loss of accessibility at super-enhancers (SEs), correlating with a reduction in the expression of SE-dependent genes. Using the dTAG system, this study has both reinforced HDAC1’s role as a chromatin repressor and has also provided novel insights into the immediate regulatory dynamics of HDAC1, uncovering complexities that were previously obscured in genetic knockout and inhibitor-based studies.</p
Building Soft Power: Nation Branding Through Digital Diplomacy At The Saudi Ministry Of Culture In The Context Of Vision 2030
No full textPrior to the launch of the Vision 2030 national development framework in 2016, the Kingdom of Saudi Arabia had limited cultural engagement with the world beyond. This in turn contributed to the development of negative and even Islamophobic stereotypes regarding the Kingdom, which undermined its global brand and soft power. Against this backdrop, this thesis examined the communication strategies used by the Saudi Ministry of Culture (MOC) to rebrand Saudi Arabia’s global image and enhance its soft power through digital diplomacy on the X/Twitter platform, as part of the broader Vision 2030 framework. Moreover, the study also investigates the perspectives of Saudi decision-makers regarding the implementation of Vision 2030 in the cultural sector.The thesis applied a case study methodology, wherein two main methods were deployed to collect and analyse data. The first was a multimodal analysis of the MOC’s posts on the X/Twitter platform from June 2021 to December 2023. The second was semi-structured interviews with elites and experts working in the cultural sector. The analysis found that the MOC’s X/Twitter account (@MOCSaudi_en) deployed a variety of communication strategies aimed at expanding its global reach by posting in multiple languages, particularly English, which is the language of media in globally influential countries. The findings also showed that the MOC deployed hashtags such as #UNESCOHeritage linked to #SaudiCulture as a digital strategy to connect Saudi cultural initiatives, partnerships, and events to the global arena. Furthermore, the findings indicated that the MOC’s digital content relies on visual elements such as videos, photography, and infographics to promote Saudi cultural heritage, along with live coverage of international events such as the Cannes Film Festival, with digital tools such as AI, VR, and AR also utilised to reinforce its cultural messaging.The experts and elites interviewed expressed support for the MOC’s digital strategies, which in reflecting the broader social, cultural and economic changes underway in the Kingdom, mean that the MOC’s efforts to rebrand the country occur on a foundation of authenticity. That said, they also noted the need for greater communication efforts to reach a global audience, and in particular, cautioned against relying on digital diplomacy alone.</p
Authentic expertise: Influencers, misinformation, and authenticity in the online cancer community
No full textIn the online cancer community, influencers have become powerful sources of information,advice, and community for their followers, many of whom are struggling with cancerthemselves. Unfortunately, these actors are also in a strong position to propagatedangerous misinformation that jeopardizes their followers’ health unwittingly. In thisarticle, we draw from 12 semistructured, in-depth interviews with influencers in the onlinecancer community about their experiences navigating cancer misinformation on socialmedia. Findings show that influencers attempt to avoid spreading misinformation aboutthe disease in two overlapping ways: (1) by couching their content production in thediscourse of personal experience and (2) through periodic performances of transparencyand fallibility. Taken together, these strategies represent the ascent of what we call“authentic expertise,” a hyper-individualized form of expertise cultivated and performedthrough the enactment of personal truths and imperfection. We argue that authenticitymay operate as an important narrative device that shields influencers from theconsequences of spreading inaccurate information. Implications for theory and practicalefforts to address misinformation are discussed.</p
From organizational resilience capability to firm performance: A time-lagged investigation
No full textAn organizational resilience capability is essential for firms to adapt in an ever-changing world characterized by diverse crises and adversities. However, the mechanisms and contingencies needed to harness and channel organizational resilience capability remain uncertain. This article argues that this resilience capability, proactively developed in calmer times, only becomes apparent in a crisis, but its effectiveness relies on additional organizational capabilities around innovation and human resources. Using data from a sample of large Taiwanese firms collected in two waves during and after a crisis, we show how incremental and radical innovative capabilities mediate the relationship between a proactively developed organizational resilience capability and firm performance. Furthermore, we demonstrate the role of strategic human resource management practices in fostering and regulating this relationship and the functioning of innovative capabilities. Based on these findings, we provide implications for theory and practice. Managerial Summary Organizational resilience is not something a firm discovers in a crisis. Instead, organizational resilience is something managers must build long before a crisis arrives. The article shows that organizations perform better in turbulent times when they have proactively developed resilience and paired it with strong innovative capabilities. Both incremental improvements and more radical innovations play a crucial role, giving firms a wider set of response options when crises differ in nature and intensity. The findings also highlight that resilience alone is not enough. Its impact depends on strategic human resource practices that build agility, learning capacity, and rapid decision making. Firms with employees who can quickly learn, unlearn, and apply new knowledge are better positioned to turn resilience into real performance gains. Managers should invest early in resilience routines—communication, readiness planning, and post crisis learning—and maintain a balanced innovation portfolio. Managers should strengthen HR systems that support flexibility, skill development, and coordinated action. Finally, managers should conduct regular resilience health checks to assess preparedness and ensure that existing measures genuinely work. Together, these actions help organizations not only withstand crises but also emerge from them stronger and more competitive.</p
Bacteriophages and nanoMIPs within the context of alternative treatments to antibiotics
No full textAntimicrobial resistant bacteria infections are increasing in prevalence and are predicted to be the major contributor to global mortality by 2050. Furthermore, an increasing number of studies show associations between antimicrobial use with chronic diseases. To combat both antimicrobial resistance and the negative effects of antimicrobial treatments, alternative strategies need to be investigated, two of which are investigated in this thesis. Nano-sized molecularly imprinted nanopolymers (nanoMIPs) are nanopolymers synthesised to recognise a specific target molecule and can be used as targeted drug delivery vectors. Here, I synthesised nanoMIPs to recognise Clostridioides difficile and Helicobacter pylori to specifically bind to these bacteria. Despite binding of the nanoMIPs to their targeted bacteria, non-specific binding of nanoMIPs was also observed. The use of bacteriophages therapeutically, is another alternative treatment capable of treating antimicrobial resistant bacteria with minimal side effects. However, in H. pylori and Streptococcus pneumoniae, few bacteriophages have been characterised and their interaction with bacteria is poorly understood. To address this issue, the identification and characterisation of prophages, and putative bacteriophage defence mechanisms are investigated in H. pylori and S. pneumoniae. In total, 37 H. pylori prophages were identified and characterised from H. pylori genomes. Phylogenetic analysis of the prophages showed independent evolution with regards to their bacterial hosts and showed phylogeographic clustering. Excision of the prophages were investigated, with a correlation observed between att site conservation and excision rate, however no bacteriophages were isolated. In S. pneumoniae, bacteriophages were used to study the phase-variable type I RM system showing that while the SpSL1 phage is restricted by the system, another bacteriophage SpGS1 and the lytic bacteriophages CP-1 and DP-1 were not. The SpSL1’s ability to activate platelets was investigated due the identification of a PblB-like protein as its tail fibre and platelet activation by SpSL1 was observed in ex vivo experiments.</p
Vision Mamba in Remote Sensing: A Comprehensive Survey of Techniques, Applications and Outlook
No full textDeep learning has profoundly transformed remote sensing, yet prevailing architectures like Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) remain constrained by critical trade-offs: CNNs suffer from limited receptive fields, while ViTs grapple with quadratic computational complexity, hindering their scalability for high-resolution remote sensing data. State Space Models (SSMs), particularly the recently proposed Mamba architecture, have emerged as a paradigm-shifting solution, combining linear computational scaling with global context modeling. This survey presents a comprehensive review of Mamba-based methodologies in remote sensing, systematically analyzing about 120 Mamba-based remote sensing studies to construct a holistic taxonomy of innovations and applications. Our contributions are structured across five dimensions: (i) foundational principles of Vision Mamba architectures, (ii) micro-architectural advancements such as adaptive scan strategies and hybrid SSM formulations, (iii) macro-architectural integrations, including CNN–Transformer–Mamba hybrids and frequency-domain adaptations, (iv) rigorous benchmarking against state-of-the-art methods in multiple application tasks, such as object detection, semantic segmentation, change detection, etc. and (v) critical analysis of unresolved challenges with actionable future directions. By bridging the gap between SSM theory and remote sensing practice, this survey establishes Mamba as a transformative framework for remote sensing analysis. To our knowledge, this paper is the first systematic review of Mamba architectures in remote sensing. Our work provides a structured foundation for advancing research in remote sensing systems through SSM-based methods. We curate an open-source GitHub repository to foster community-driven advancements.</p