21684 research outputs found
Sort by
Biomimicking human tactile sensation using flexible tactile sensors
Get PDFThe sense of touch is essential for human interaction with the environment, allowing us to perform tasks ranging from simple actions to intricate procedures. Human tactile perception, through mechanoreceptors in the skin, enables the detection of various stimuli, including pressure, texture, temperature, and vibration. This sensory capability is fundamental to tasks such as grasping objects, maintaining balance, and communicating emotions. Mimicking the complexity of human touch in artificial systems has been a longstanding challenge in robotics, prosthetics, and wearable technology. Achieving humanlike sensitivity in machines could transform industries, allowing robots to handle delicate objects, prosthetics to offer better tactile feedback for amputees, and wearables to enhance human-computer interaction. Advances in materials science and flexible electronics have opened new possibilities for developing tactile sensors that replicate the mechanical and sensory functions of human skin. Flexible sensors, which can conform to various surfaces and endure mechanical stress, are key to creating electronic skin (e-skin) capable of sensing pressure, strain, temperature, and other physical stimuli. This thesis explores the development of flexible tactile sensors that bio-mimic human tactile systems for use in robotics and prosthetics, while also leveraging computational simulations to improve design of these sensors.
Towards this, in response to the need for reliable and sensitive strain sensors, this work initially presents stretchable strain sensors that operate over a wide range and exhibit excellent gauge factors. The sensors, composed of elastomer, conductive filler, and graphene–carbon paste (GCP), demonstrated high stretchability and sensitivity. Molecular dynamics simulations showed that adding GCP to the composite material enhances sensor response. A strain sensor with a high gauge factor of 1,834,140 was also developed. Additionally, a neuromorphic strain sensor system was developed using this piezoresistive sensor and a simulated neuron to mimic the behavior of mechanoreceptors. Applications of this neuromorphic strain sensing system in stretch and angle bending have been successfully demonstrated.
Next, a hybrid sensor system was developed combining capacitive pressure sensors and triboelectric nanogenerators, mimicking both slowly and rapidly adapting mechanoreceptors. The hybrid system successfully detected static and dynamic stimuli and showed good applications for impact detection and slip monitoring. Later, a simulation-based study investigated PDMS based soft capacitive pressure sensors enhanced with ZnO nanowires. Results indicated a significant improvement in sensor sensitivity by ~3 times, especially with vertically aligned ZnO nanowires. This work demonstrates the potential of using simulation methods to optimize sensor design and enhance tactile sensor performance for future applications in robotics, prosthetics, and human-machine interfaces.
Finally, atomistic simulations explored the behavior of water molecules trapped between a graphene and gold interface, which is a common issue in graphene-based devices due to the wet-transfer process. The effect of contaminants like water on the electronic properties of such devices is highly unexplored. The molecular dynamics simulations revealed that when the water film thickness is below 5 Å, it forms an ice-like structure, potentially causing strain in the graphene layer and affecting sensor performance. As the water thickness increases, the water transitions to a liquid state, reducing strain. This study provides critical insights into the challenges posed by trapped water at the grapheneAu interface, which can impact the performance of devices such as graphene-based fieldeffect transistors (GFETs). The research work presented in this thesis could provide a foundation for future work in the field of flexible and hybrid sensor systems, with potential applications in robotics, prosthetics, and human-machine interfaces
Multi-Plane Light Converter based on metasurface and Machine Learning to understand the mode sorter’s applications
Get PDFThis scientific study delves into the realm of metasurfaces, offering an exhaustive investigation into their underlying principles, practical applications, and the fabrication methods imperative for their realisation. A focal point of this exploration is the detailed exposition of the fabrication process for the Multi-Plane Light Convertor (MPLC) device, supported by captured images validating the precision of each critical step. Initial results indicate the satisfactory functioning of the MPLC, yet further analyses and optimisations are deemed essential to unlock its full potential.
The MPLC device demonstrates versatile applications across telecommunications, energy-related fiber sensing, medical imaging, and biological tomoholography. However, at present, no physical devices based on metasurfaces are available that can fully implement these functions.
In parallel, a novel and robust fibre bend sensor has been developed, showcasing the capability to precisely locate bends through inter-modal coupling. Modal decomposition reduces sensitivity to relative phase, revealing features providing accurate information about the shape or position of bends within the fiber. The simplicity and cost-effectiveness of this approach offer potential applications in wearable technology, motion sensors and aircraft wing shape sensing.
Both experiments revolve around the concept of a mode sorter. The first experiment focuses on creating a novel device not yet available on the market, specifically the Multi-Plane Light Converter (MPLC). The second set of experiments, on the other hand, is centered around the practical application of the mode sorter as an instrumental component. The combination of mode de-multiplexing with machine learning holds promise for powerful applications, particularly in scenarios where constant variations in relative phase can be treated as noise, such as monitoring atmospheric conditions or extracting information from environments with dense scattering.
Practical deployment considerations include the need for retraining in cases of significant system or fiber type changes. Once fully trained, retraining intervals are typically weeks to months under normal temperature fluctuations, necessitating further research into extreme temperature variations encountered in applications like aviation. The use of multi-core fibers is recommended to enhance sensitivity to multiple directions.
In summary, the study demonstrates the feasibility of utilizing machine learning for accurate millimetric-scale curvature detection by incorporating a mode sorter into the optical setup. While exhibiting robust performance, limitations exist in detecting bends or movements not introducing changes in inter-modal coupling and relative phase shifts. The consistent alignment and outcomes observed in experiments underscore the stability and reliability of the experimental setup, instilling confidence in the algorithm’s performance
Development and utilisation of predictive modelling tools for optimising the operations of future cellular networks
Get PDFWith the advent of ultra-densification and the proliferation of diverse underlying technologies, 5G and beyond networks promise unprecedented capabilities but also introduce significant challenges in network management due to the manifold increase in complexity. Two such key challenges are network resources optimization and network maintenance. It is challenging to perform these tasks manually, with 5G and beyond networks it does not remain viable at all. Self-aware solutions like Self-Organizing Networks (SON) have, therefore, been proposed and well explored in research to address these challenges.
However, the legacy SON, still relies on the predefined instruction sets making them inelastic to network changes. On the other hand, they use the results from field tests or customers complaints for the identification of network issues which cause unnecessary delays and make existing SON reactive. But, to cater for the exponential increase in network complexity and diverse use cases in 5G and beyond networks, SON needs to be more intelligent, adaptable and autonomous. They are required to be proactive rather than reactive. Artificial intelligence can play a crucial role here and machine learning can equip SON with this intelligence by exploiting the hidden patterns in the real network data. Another advantage of machine learning is that it also brings prediction capacity important for making SON proactive.
Artificial intelligence, particularly machine learning, offers the potential to transform SON into proactive systems by extracting actionable insights from network data and enabling predictive capabilities. This research focuses on leveraging machine learning to enhance two key SON functions: self-optimization and self-healing.
This study begins by exploring and classifying various data types generated within the network, highlighting their potential roles in wireless cellular networks (WCN). A review of existing and potential use cases for SON functions and machine learningbased approaches provides the foundation for this work. For self-optimization, a Support Vector Machine (SVM) model is developed to predict internet traffic loads using Call Detail Records (CDR), achieving up to 91% prediction accuracy. Additionally, a novel machine learning model leveraging Geohash global indexing predicts users’ next locations with approximately 95% accuracy, marking a significant contribution to mobility prediction.
To enable self-healing, a hybrid machine learning scheme is proposed. Using CDR data, network cells are grouped based on performance via K-means clustering. Subsequently, an SVM classifier is employed to categorize cell performance with 98% accuracy. The traffic prediction model for self-optimization is further refined through a Support Vector Regression (SVR) approach, achieving 97% prediction accuracy. The predictive capabilities of the model contribute to energy savings of up to eightfold, underscoring its practical impact.
By integrating prognostics and self-aware systems into SON, this research demonstrates a pathway to achieve self-optimization and self-healing in 5G and beyond networks, laying the groundwork for sustainable, intelligent network management
Mechanistic analysis of collagen IV mutation in stroke and cerebral small vessel disease
No full textAbstract not currently available
Itaconate mediated NLRP3 inflammasome tolerance in the context of human monocyte Tenascin-C activation
Get PDFDespite the extensive research conducted on innate immune cell memory, there are still many unknowns when it comes to the ability of endogenous danger molecules, also known as damage-associated molecular patterns (DAMPs), to influence the tolerisation of myeloid cells to repeated stimulation. This is partly due to the majority of tolerisation studies using exogenous pathogen-associated molecular patterns (PAMPs) such as Lipopolysaccharide (LPS). While understanding virulence factor-induced LPS tolerance is important for treating infectious diseases and sepsis patients, it is essential to examine cellular stimulations using disease-relevant DAMPs to fully comprehend the development of chronic inflammatory conditions such as rheumatoid arthritis. In my work, I have utilized the Toll-like receptor 4 (TLR4) activating fibrinogen-like globe domain (FBG-C) of the extracellular matrix protein tenascin-C to advance our scientific understanding of the ability of DAMPs to trigger and tolerise a critical aspect of TLR-mediated inflammation known as the NLRP3 inflammasome.
I discovered significant differences in the capacity of primary human monocytes to produce and secrete the non-inflammasome regulated cytokine tumour necrosis factor (TNF) compared to the inflammasome regulated cytokine interleukin-1 beta (IL-1β), as well as in their capacity to secrete IL-1β via classical NLRP3 mediated pyroptotic means versus alternative non-pyroptotic means. Additionally, I found that although the DAMP FBG-C and the PAMP LPS induce a similar cytokine response in primary human monocytes in the first 24 hours of activation, subsequent restimulation revealed fundamental differences in their capacity to induce tolerisation. Notably, this effect is specific to IL-1β, and hence the NLRP3 inflammasome, and not due to differences in the requirement for the TLR4 co-receptor cluster of differentiation 14 (CD14).
Furthermore, I found that the inability of monocytes to tolerise the NLRP3 inflammasome following FBG-C activation is due to a delayed upregulation response of the itaconate-producing enzyme aconitate decarboxylase 1 (ACOD1). I then uncovered that ACOD1 upregulation is crucial for inhibiting the processing of the pore-forming protein Gasdermin D and thus for inhibiting pyroptosis in tolerised monocytes. Importantly, I also showed that monocytes isolated from RA patient blood have a delayed ACOD1 upregulation response following both LPS and FBG-C TLR4 activation, which prevented the tolerisation of RA monocytes in a Gasdermin D and pyroptosis-dependent manner. Finally, I uncovered that a 2% oxygen (hypoxic) culture environment renders human monocytes unable to tolerise the NLRP3 inflammasome in an ACOD1-independent manner.
In summary, my research revealed significant differences in the way primary human monocytes interpret pathogen and damage signals, in addition to uncovering substantial changes in their ability to produce the pleiotropic proinflammatory cytokine IL-1β over a 48-hour period. This work demonstrated that environmental stressors, endogenous triggers and chronic pathology could all trigger loss of tolerance and a prolonged NLRP3 inflammasome priming phenotype
Investigating the effects of augmented reality cues during non-driving related tasks on the situational awareness of drivers
Get PDFAs automated vehicles become responsible for more of the driving task, the way in which drivers need to process the road will change. This will free up capacity to engage with non-driving related tasks, activities which are unrelated to operation of the vehicle or attending to the road. However, the current state of automation still requires a driver to be ready to resume control of the vehicle at all times. This thesis evaluates the effects of using augmented reality for presenting these non-driving tasks to drivers of automated vehicles. In particular, the ability of drivers to react to a hazard and predict what happens next in a road scene, a key component of situational awareness, were measured while they were performing a non-driving related task presented in augmented reality. Six experiments, using a mix of validated empirical tests of situational awareness, expert focus groups and eye tracking measures, were designed and conducted to assess the impact of engaging with a distracting non-driving task while attempting to maintain attention on the road. Results showed that, contrary to prior recommendations, a heads-up display presentation of a non-driving task at eye level does not convey the same benefits found when displaying non-driving related information. Evidence of intentional blindness was found when evaluating the use of attentional cues within a dynamic augmented reality display. This demonstrated that using eyes-on-road as a measure of attention is not wholly appropriate when investigating how an augmented reality interface overlaid onto the real-world impacts driver attention. Further exploration into how to design efficacious attentional cues highlights how the inclusion of salient attention capturing elements in a positional cue can enhance driver awareness of the road when they are distracted by an NDRT. Additionally, it was shown that drivers were able to utilise social cues from a virtual agent highlighting the position of a hazard, indicating the potential application of this modality for enhancing situational awareness through in-vehicle virtual assistants. This thesis contributes to the field by providing evidence of the impact of presenting an NDRT via AR on driving performance measures and methods in which this can be overcome with attentional cues. Overall, the findings in this thesis have significant implications for the applied transport psychology and automotive user interfaces domains
Essays on knowledge and justice
Get PDFThis thesis can be seen as a modest contribution to a growing literature that aims to challenge some core assumptions of a widely shared image of epistemology. This is an image of epistemology as a discipline centred on the individual, whose core assumptions concern the way in which this individual relates to the world around them and to other people. This thesis contributes to a critique of this image in a non-direct and non-unitary manner. The critique is not direct because (with the exception of chapter one and, to some extent, chapter six) none of the works here collected offers an explicit challenge to this image. The critique is instead positive, as it furthers a competing image of epistemology as a deeply social discipline. Finally, this critique is non-unitary because the chapters put forward independent arguments, each attempting to capture a different angle of the multiple ways in which social and political relations structure how we think about core epistemic concepts. The result is a harlequin work describing the branching trajectory of an ongoing research into some fundamental philosophical questions on the nature of our epistemic lives
Development and clinical translation of virus-specific T cell therapies for treatment of life-threatening viral diseases
Get PDFAdoptive immunotherapy with virus-specific T cells (VST) has demonstrated clinical efficacy in restoring antiviral immunity in immunocompromised patients.
Epstein-Barr virus (EBV) is an extremely common herpesvirus (90% prevalence worldwide), that infects the majority of individuals during childhood or adolescence and thereafter establishes a lifelong latency in the host. In immunocompromised individuals, EBV can drive a malignant transformation of B cells due to the absence of circulating memory EBV-specific T cell surveillance. Transplant patients are particularly at risk of developing an aggressive EBV-induced lymphoma, and despite first line treatments the mortality rate remains high.
The Scottish National Blood Transfusion Service has been involved in adoptive EBV VST therapy of EBV-associated lymphomas for over 20 years. Initial EBV VST products were manufactured from healthy donor leukapheresis using repeated stimulations with EBV-infected lymphoblastoid cell lines to induce expansion of EBVspecific T cell clones. These therapies have treated over 200 patients with excellent safety and efficacy in tumour regression (>60% complete response rates). More recently, we developed a new manufacturing process by stimulating donor leukapheresis with EBV peptide pools, followed by isolation of responding memory EBV VST using cytokine capture selection, with further culture expansion. This study aimed to comprehensively compare EBV VST therapies manufactured from the two processes, with characterisation assays developed to assess quality, phenotype, functionality, migratory capacity and clonal repertoire of T cell products. Peptidederived VST demonstrated enhanced degranulation and cytokine production to a broad range of EBV latent antigens, as well as a dominant central memory phenotype, which may improve persistence and targeted tumour clearance in patients. Moreover, the peptide process had clear benefits in terms of process biosafety, reduced culture duration and massively higher yield of patient doses.
Furthermore, the emergent severe acute respiratory syndrome coronavirus-2 (SARSCoV-2) outbreak at the beginning of this study allowed us to investigate SARS-CoV-2 immune responses in individuals following natural resolution of primary infection. Natural killer cell frequency within peripheral blood mononuclear cells was significantly increased in individuals within 1-3 months convalescence compared to unexposed individuals. Convalescent donors had detectable CD4 and CD8 memory T cells populations to SARS-CoV-2 spike, nucleocapsid and membrane peptides whereas unexposed individuals showed no lymphocyte responses to the SARS-CoV-2 antigens. Interestingly, the frequency of SARS-CoV-2-specific T cells within the total T cell compartment decreased over time from symptoms resolution, indicating memory T cell responses of unvaccinated individuals after primary infection decline without antigen re-exposure. With detectable memory T cell populations, we were able to isolate SARS-CoV-2-specific T cells from donor blood using cytokine capture or T cell activation-induced marker selection. We further developed a culture protocol to rapidly expand a purified SARS-CoV-2 VST population with desirable central memory phenotype and broad SARS-CoV-2 antigen effector functionality. Given that disease severity has been associated with a reduced or dysfunctional T cell response, adoptive transfer of healthy donor SARS-CoV-2 VST may provide a potential treatment strategy. To this end, the peptide-mediated process developed for EBV was rapidly translated to manufacture an allogeneic bank of SARS-CoV-2 VST, with clinical products tested in a first-in-human trial for high-risk hospitalised patients.
The development and clinical manufacture of two T cell therapies targeting very different viral diseases required comprehensive analytical testing to understand the potential functional mechanisms of these cell products. The extensive suite of characterisation assays developed was used to build a profile for antigen-specific T cell therapies, in order to evaluate the optimal characteristics for clinical efficacy
Customer engagement and bias in online reviews: comparative analysis between Airbnb and traditional hotels
Get PDFThis study advances the understanding of customer engagement and online review dynamics in the hospitality industry by comparing traditional hotels and Airbnb within the sharing economy framework. Although online reviews have become central to shaping consumer decision-making, prior research has underexplored the distinct engagement mechanisms and biases between standardised hotel services and peer-to-peer accommodation platforms. In particular, limited attention has been given to how cognitive, emotional, and behavioural engagement dimensions influence review content, or how service-dominant logic (SDL; Vargo and Lusch, 2008) and customer-dominant logic (CDL; Heinonen et al., 2010) shape these patterns. This study addresses these gaps by examining the role of reciprocity, emotional connection, and platform design in driving review biases.
A robust multi-qualitative design was employed, integrating large-scale text-mining of TripAdvisor and Airbnb reviews using Leximancer with purposive manual coding, 26 semi-structured interviews, and a follow-up quantitative survey (Ryan, 2019; Creswell, 2014). This methodological integration ensures scalability in detecting thematic patterns while retaining contextual validity, aligning with critical realism’s emphasis on combining objective structures with subjective interpretations and with recent recommendations for ensuring paradigm–method fit in qualitative international business research (Aguzzoli et al., 2024).
Findings reveal that SDL-driven hotel engagement more emphasises structured, standardised interactions centred on service consistency and reliability, with relational elements such as guest recognition contributing to loyalty. In contrast, CDL-driven Airbnb engagement prioritises personalised, emotionally resonant experiences, often resulting in positively biased reviews reinforced by reciprocal host–guest relationships and bilateral review systems. Platform-specific features were found to amplify positive sentiment while potentially obscuring dissatisfaction.
Academically, the research bridges SDL and CDL within a comparative platform context, offering a novel conceptual model that links engagement dimensions with review biases. Methodologically, it demonstrates the value of combining automated semantic analysis with qualitative triangulation and quantitative validation. Managerially, the findings provide actionable insights for hotels to incorporate personalised engagement strategies and for sharing economy platforms to strengthen review authenticity. By addressing a critical gap in understanding engagement and bias in online reviews, the study contributes to both theory development and practice in digital hospitality management
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! 👇