21684 research outputs found
Sort by
Stratification of respiratory disease through early diagnostic sampling
Get PDFBoth lung cancer and Coronavirus Disease 2019 (COVID-19) present with a wide range of prognoses and eventual outcomes. In both conditions early diagnosis and commencement of treatment (where appropriate) can improve survival. Given the spectrums of disease seen in lung cancer and COVID-19, stratification of patients also aids in the selection of the most appropriate individual management options to further optimise patient outcome. The overall aim of this thesis is to examine the stratification of respiratory disease through early diagnostic sampling.
Chapter 2 describes the design and delivery of the multi-centre, prospective STRATIFY study (Staging by Thoracoscopy in potentially radically treatable Lung Cancer associated with Minimal Pleural Effusion). It has been established from retrospective data that those presenting with early stage, otherwise potentially radically treatable non-small cell lung cancer (NSCLC) and minimal pleural effusion do significantly worse in terms of survival than those without such effusions. Based on this previous retrospective data it has been hypothesised that this difference in survival is due to the presence of occult pleural metastases (OPM) not otherwise detected in the routine diagnostic work up of those with suspected NSCLC and minimal pleural effusion. STRATIFY was therefore designed as the first prospective, multicentre, observational study with the primary aim of determining the true prevalence of OPM in this cohort of patients through the addition of thoracoscopy to the diagnostic pathway.
Unfortunately, due to a multitude of issues many of which stemmed from the COVID-19 pandemic, recruitment to STRATIFY was slower than expected and therefore the decision to close the trial to recruitment was taken in May 2024. Primary outcome data including the prevalence of OPM and important safety data for the 27 recruited patients are however included here.
Patients with lung cancer (and many other common malignancies) often present with large pleural effusions which go on to be proven malignant through simple aspiration. The diagnostic yield of pleural fluid cytology is well documented at 60% on average but varies considerably by tumour type. The yield of predictive markers from effusion cytology, which are now mandated in the diagnostic work up of lung and breast adenocarcinoma, is less well established. Chapter 3 of this thesis therefore aimed to assess the utility of pleural fluid cytology for the detection of predictive markers in lung and breast adenocarcinoma in a real-world setting. This multicentre, retrospective cohort study found that the full panel of predictive marker (PM) testing required by contemporaneous international cancer treatment guidelines was returned in only 20% of cases where these were requested on pleural fluid cytology. Performance differed by individual marker with yields for many markers improving over the time period of the study. No clinico-radiological factors were significantly associated with PM testing yield to guide pre-aspiration likelihood of success.
Perhaps even to a greater extent than lung cancer, outcomes from COVID-19 are markedly heterogeneous with some patients complaining of little to no symptoms while others go on to develop potentially fatal pneumonitis. Although risk factors for poor prognosis are well documented, less is understood about the individual immune response and how these immunological events affect disease outcome. In chapter 4 individual immune response at the time of COVID-19 diagnosis and how this relates to disease severity was examined. In keeping with previous work, severe COVID-19 (as defined by the need for supplemental oxygen) was associated with obesity and hypertension (p= 0.0456, p= 0.0071 respectively) in this cohort. Flow cytometry of baseline serum samples revealed lower activation (phosphorylation) of STAT 5 in response to stimulation across almost all immune cell subpopulations in those with severe disease. Interrogation of potential mechanisms linking metabolic syndrome and the altered STAT5 signalling observed are ongoing by collaborators at the time of writing
From Siena to Syon: a study on the transmission and translation of Catherine of Siena’s Middle English texts
Get PDFThis thesis studies the medieval English reception of texts by and about Catherine of Siena (1347–1380) and, in particular, their transmission across medieval Europe and their translation into Middle English. After surveying the corpus of Catherinian texts, the thesis examines case by case the five texts for which circulation in medieval England can be established conclusively: Raymond of Capua’s Legenda maior (chapter 1); two related hagiographical letters, Stefano Maconi’s Epistola de gestis et virtutibus sanctae Catharinae and Bartolomeo da Ravenna’s Epistola 7omae Antonii de Senis (chapter 2); William Flete’s Documento spirituale (chapter 3); and Catherine’s Dialogo della divina provvidenza, on which some preliminary observations are made (coda to Part I: Transmission). Building on previous studies on the transmission of the Legenda maior with original archival and philological research, this thesis situates manuscript and early-print evidence from medieval England—both in Latin and Middle English—within a larger corpus of European archival material. Ke chapters in Part I of the thesis outline the textual histories of each work under consideration, paying attention to the text’s genesis, its later recensions, and the historical context in which these recensions were produced and circulated. By way of textual and historical evidence, Part I of the thesis argues that English copies of the Legenda maior and of Maconi’s and Bartolomeo’s Epistolae show signs of a Carthusian transmission, while Flete’s Documento spirituale bears possible traces of a Dominican transmission. Part II turns to translation and analyzes the Orcherd of Syon, a Middle English version of Catherine’s Dialogo prepared for the Birgittine nuns at Syon Abbey (chapter 4). Kis adaptation is read alongside other Middle English translations of texts by European visionary women. Ke comparison shows that the Orcherd presupposes a sophisticated reading process and a non-interventionist approach to its source that elevate both its women readers and woman author, in contrast to the typical treatment of European contemplative texts, especially if translated for women readers. All in all, medieval English reception of Catherine of Siena shows a deep engagement with European mystical literature as well as a cosmopolitanism not often recognized in the religious and literary environment of fifteenth-century England
A new measurement of the neutron electric form factor with the Super Bigbite Spectrometer apparatus
Get PDFProtons and neutrons, collectively known as nucleons, make up the nuclei at the core of atoms which form our world. The nucleon has been under intensive study for over 100 years, and yet we still do not fully understand the internal dynamics which govern properties like its spin or its mass - which contributes to almost all of the visible mass in the universe. These dynamics are governed by quantum chromodynamics (QCD), the predictions of which are experimentally tested at high energy accelerator facilities such as Jefferson Lab. The GEN-II experiment (E12-09 016) is one such experiment.
GEN-II is part of the Super Bigbite Spectrometer (SBS) experimental form factor programme taking place in Hall A at Jefferson Lab, which aims to make precision measurements of the nucleon electromagnetic form factors (EMFFs) at record high values of squared four-momentum transfer Q2 . EMFFs describe the electric and magnetic moment distributions within the nucleon. They can be measured through elastic electron scattering off the nucleon, and describe the recoil response of the target nucleon at a given energy scale.
GEN-II is a double polarised semi-exclusive beam target asymmetry (BTA) experiment, seeking to measure the electric form factor of the neutron, GnE, at three new values of squared four-momentum transfer Q2 = 2.92,6.74 and 9.82 GeV². The latter two points being at record high Q2 . The form factor is determined through measuring the BTA of quasielastic scattering of a neutron from a polarised nuclear target. The experiment utilised the CEBAF accelerator to produce longitudinally polarised electrons up to ∼85% polarisation, which were scattered off neutrons within a novel polarised helium-3 (³He) target. This new polarised ³He target was employed by building on the technology of its precursors which existed in similar preceding experiments. This target was designed to operate at the high luminosities typical of Hall A, and reached a record breaking combination of polarisation and beam intensity known as figure of merit, three times larger than those predecessors.
The SBS collaboration designed and constructed two brand new high acceptance spectrometers for these experiments, an electron arm named Bigbite (BB) and a hadron arm named Super Bigbite. Both spectrometers featured a large acceptance EM dipole magnet, and complementary detector systems. The electron arm contained gaseous electron multipliers (GEMs) which were used for high precision tracking of the scattered electrons, a heavy gas cherenkov (GRINCH) which was used for PID between electrons and pions, a plastic scintillator timing hodoscope to provide high resolution timing of the start of events, and a pair of EM calorimeters (BBCal) which provided energy measurements of detected particles, and provided the experimental trigger. The hadron arm also contained a system of GEMs which will be utilised for future SBS experiments, and a hadron calorimeter designed to provide position, timing and energy measurements of the recoiling nucleon.
The calibration of all detector subsystems, beam and target data is discussed, with a focus on novel timing calibrations to the hodoscope and hadron calorimeter. An analysis of selecting quasielastic events and suppressing background contributions from a number of sources which contaminate the final event sample is given. The largest irreducible backgrounds are found to be from misidentified protons, timing accidentals and inelastic events. The physical asymmetry is measured and used to extract a value for the form factor ratio GnE/GnM. High precision Q2 data for GnM is used to then extract GnE. This work finds at Q2 = 2.92 GeV2 that GnE = 0.0129+0.0019 −0.0020. This result is in statistical agreement with existing fits to world data, and predictions from the constituent quark model and Dyson–Schwinger equations, in this region of Q2
The faecal phenotype: analysing faeces as a means of assessing nutrient utilisation and digestive health
No full textAbstract not currently available
Educators’ understandings of multimodality in relation to picturebooks and their pedagogic potential in China
No full textAbstract not currently available
Quantum many-body integrable systems and related algebraic structures
Get PDFThis thesis deals with various many-body quantum integrable Hamiltonian systems and algebraic structures related to them. More specifically, it discusses generalisations of Calogero–Moser–Sutherland (CMS) and Macdonald–Ruijsenaars (MR) type systems and their connections with the theory of double affine Hecke and related algebras.
Firstly, we consider the generalised CMS operators associated with the deformed root systems BC(l, 1) and a CMS type operator associated with a planar configuration of vectors called AG2, which is a union of the root systems A2 and G2. We construct suitably-defined (multidimensional) Baker–Akhiezer eigenfunctions for these operators, and we use this to prove a bispectral duality for each of these generalised CMS systems. In the case of AG2, we give two corresponding dual difference operators of rational MR type in an explicit form, which we generalise to the trigonometric case as well by using the theory of double affine Hecke algebras (DAHAs). In the case of BC(l, 1), the bispectral dual is a rational difference operator introduced by Sergeev and Veselov.
Secondly, we study systems with spin degrees of freedom. Quantum integrable spin CMS type systems with non-symmetric configurations of the singularities of the potential appeared in the rational case in the work of Chalykh, Goncharenko, and Veselov in 1999. In this thesis, we obtain various trigonometric spin CMS type systems by making use of the representation theory of degenerate DAHAs. Particular cases of our construction reproduce in the rational limit the examples discovered by Chalykh, Goncharenko, and Veselov.
Finally, inside the DAHA of type GLn, which depends on two parameters q and τ , we define a subalgebra Hgln that may be thought of as a q-analogue of the degree zero part of the corresponding rational Cherednik algebra. We prove that the algebra Hgln is a flat τ -deformation of the crossed product of the group algebra of the symmetric group with the image of the Drinfeld–Jimbo quantum group Uq(gln) under the q-oscillator (Jordan–Schwinger) representation. We find all the defining relations and an explicit PBW basis for the algebra Hgln . We describe its centre and establish a double centraliser property. As an application, we obtain new integrable generalisations of Van Diejen’s MR system in an external field
Memory and materiality at British-colonial detention camps in central Kenya
No full textAbstract not currently available
Exploring the role of lipid peroxidation and glutathione peroxidase 4 in childhood acute lymphoblastic leukaemia
No full textAbstract not currently available
Laser-induced graphene-based elastomer nanocomposites for soft sensing applications
Get PDFLaser-induced graphene (LIG) has gained significant attention as a multifunctional material for flexible and wearable sensor technologies due to its exceptional electrical conductivity, mechanical flexibility, and tunable properties. This thesis presents a comprehensive study of the fabrication, characterization, and application of LIG-polymer nanocomposites, emphasizing the optimization of their electromechanical and thermomechanical performance for soft strain sensing applications.
The fabrication process emphasizes the precise tailoring of LIG properties through controlled laser processing and strategic composite design. In addition to doping LIG with conductive nanoparticles to enhance electrical sensitivity and structural robustness, the integration of LIG with polymer matrices such as PDMS was systematically optimized to modulate the composite’s mechanical flexibility and thermal responsiveness. The effects of laser parameters, substrate interactions, and doping strategies on the microstructure, porosity, and conductivity of the LIG network were thoroughly investigated. This approach enabled the tuning of both electromechanical and thermomechanical behaviour to suit soft sensing applications. Environmental factors, including humidity and temperature, were also considered, as they significantly influence the stability and performance of the composite in real-world conditions.
In addition to experimental advancements, the development of a LabVIEW Python platform for electromechanical testing is highlighted. This platform provided an efficient and precise system for characterizing the strain-resistance behaviour of LIG-polymer composites, enabling the acquisition of high-resolution data critical to this work.
A theoretical model was developed to elucidate the relationship between strain distribution and resistance in LIG-polymer composites. The model incorporates Gaussian strain distributions and integrates experimental resistance-strain data with finite element simulations to quantify the effective resistance under thermal and mechanical loads. Simulations of strain distribution under thermal expansion revealed that temperature-induced narrowing of strain distributions significantly impacts the composite’s resistance response.
Applications of the LIG-polymer composites were explored across diverse fields, including healthcare and robotics, demonstrating the potential of these sensors for wearable health monitoring, soft robotics, and adaptive environmental sensing. Experimental results validated the proposed models and highlighted the influence of laser processing and doping on sensor performance, enabling a deeper understanding of the interplay between material properties and device functionality.
By bridging experimental characterization, theoretical modelling, and practical application, this thesis advances the development of LIG-based strain sensors, providing a foundation for future innovations in soft and wearable electronics
Collaborative Distributed Machine Learning: from knowledge reuse to sparsification in federated learning
Get PDFDistributed Machine Learning (DML) leverages distributed computing resources to train models and perform inference on decentralized datasets efficiently. A high-quality distributed system ensures optimal Quality of Service (QoS) by delivering low latency, high reliability, efficient resource utilization, and robust security. However, DML frameworks face significant challenges with the proliferation of devices generating vast volumes of data and the increasing complexity of tasks. For instance, heterogeneous feature spaces from data generated by different users or locations can undermine model reliability. Additionally, the growing size and complexity of models impose substantial burdens on resourceconstrained devices, particularly for inference and storage. Latency becomes a critical concern in distributed online systems such as intelligent transport systems for autonomous vehicles.
This work explores leveraging knowledge reuse, a key meta-learning technique, combined with sparsification methods to build efficient and effective distributed learning systems. To enhance efficiency, we aim to reduce redundant computation and communication, assuming that distributed data exhibit similarities despite not being identical. Specifically, we propose identifying reusable models by examining statistical patterns and meta-features derived from trained models. These reusable models are selected and adapted to local environments without requiring full retraining. Multi-task learning further improves the effectiveness of these adaptations, ensuring comparable performance to locally trained models while significantly reducing the number of models that need to be trained. By clustering models with shared characteristics, the system reduces the computational and communication overhead in a network of M devices, where only K ≪ M models need to be trained, maintaining strong overall performance.
To tackle real-world challenges where data is often non-independent and identically distributed (non-i.i.d.), we incorporate pruning techniques to enhance both system efficiency and effectiveness. This approach reduces communication and computation costs while simultaneously improving model performance and accuracy. Centralized federated learning (CFL) relies on a central server for model aggregation, while decentralized federated learning (DFL) operates without central server coordination, enabling direct communication between clients. In CFL, dynamic pruning strategies with error feedback and adaptive regularization achieve extremely sparse models, reducing computation and communication costs while accelerating inference. These models retain high sparsity with minimal accuracy loss. In DFL, the absence of a central node enhances robustness against adversarial attacks. Efficiency is further improved through dynamic pruning, allowing progressively sparser training, and a hybrid approach combining sequential and parallel training to reuse updates within the same round. Personalized pruning masks address data heterogeneity across clients, promoting both system efficiency and local model performance.
The proposed framework is validated experimentally across diverse datasets and models, including air pollution data from weather stations, temperature data from unmanned surface vehicles, and image classification tasks. The tested models span traditional approaches such as regression and support vector machines to modern deep learning architectures like convolutional neural networks. Theoretically, we conduct hypothesis testing and complexity analysis, including the development of convergence theorems for federated learning scenarios.
Overall, this thesis presents comprehensive frameworks and algorithms backed by robust experimental and theoretical results. It addresses key challenges in federated learning and edge computing through knowledge reuse and sparsification, enhancing the efficiency, effectiveness, and robustness of modern AI applications
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! 👇