21684 research outputs found
Sort by
A Bayesian approach to hyperelastic model selection and calibration for computational durability testing of stents
No full textAbstract not currently available
The molecular mechanism of tumour budding and its relationship with tumour microenvironment in colorectal cancer
Get PDFSummary
Colorectal cancer (CRC) is the third most diagnosed cancer and the second lethal disease worldwide (1). CRC development has recently been well-documented, and the screening program has been shown to improve patient outcomes and survival due to the early detection of the disease (2, 3). However, some patients still experienced disease metastasis with a 5 year survival of only 12% (4). Recent studies have focused on identifying prognostic biomarkers that can predict the adverse outcomes in CRC patients (5).
One promising factor that has recently been reported is tumour budding (TB). TB, the single or up to four tumour cells found at the invasive tumour area, is now a well-known prognostic independent biomarker in many solid cancers including CRC (6). Patients with high TB phenotype experienced a poor outcome with an incidence of disease recurrence and metastasis (7). In 2016, the international tumour budding consensus conference (ITBCC) was held and agreed to set up the criteria for TB assessment and suggested to include TB status in a routine clinical report (8). Since then, multiple studies have investigated the prognostic role of TB not only in CRC but also in other solid cancers such as pancreatic (9), breast (10, 11), head and neck (12) and lung (13) cancer.
Although TB has a strong prognostic value, few studies investigated its underlying mechanism and how it may relate to adverse features and disease metastasis in CRC. It has been hypothesised that TB could undergo epithelial-mesenchymal transition (EMT), thereby, allowing cells to escape from the main tumour and promote metastasis (14, 15). However, some studies argued that TB may only undergo partial EMT and there is another tumour-related signalling involved in its formation and induction of the metastasis (16, 17). Moreover, some studies have reported an inverse correlation between TB and cytotoxic T cells which could suggested an immunosuppressive role of TB leading to disease metastasis in CRC (18-20).
Until now, there has been little understanding of the underlying mechanism of TB and its relationship with the tumour microenvironment in CRC (21). This thesis aims to unravel the molecular mechanism of TB to identify the potential tumour signalling that drives TB formation and how TB is associated with the immune profile at the invasive edge of the tumour. To investigate this, TB status in CRC patients has been identified according to the TB assessment criteria from ITBCC. After that, bulk transcriptomic RNA (n=787) was used to identify tumour-related signalling expressed in tumours with high TB phenotype compared to low TB group. In addition, regional bulk spatial transcriptomic (GeoMx) (n=12) was performed to identify gene expression within the region of interested (ROI), the classification of tumour and stromal areas using specific protein mask (PanCK+/-) was done. This allows the identification of the potential genes related to both budding tumour cells and the surrounded tumour microenvironment between tumours with low and high TB profile as well as the different area of interest (AOI); tumour core, invasive edge, distant stromal area, within the same tumours. The results from GeoMx were later validated in a TMA of the full CRC cohort (n=787), using immunohistochemistry, to verify the translation from RNA to protein. Of these, cyclinD1 expression within TB was identified as a promising prognostic value in CRC patients. Additionally, multiplex immunofluorescence (mIF) using immune panels (lymphocytes and myeloid cells) were also performed to investigate the immune profile within the invasive budding area. Results showed a high density of regulatory and low cytotoxic T cells within the invasive compared to further stromal area of tumours with high TB. Nearest neighbour analysis also showed that TB tend to have a closer distant to regulatory cells as well as pan-macrophages. This finding suggested that TB may have a possible interaction with the surrounding immune cells leading to an alteration of the microenvironment to help it thrive and invade other parts of the body.
To investigate if TB formation can be observed within an in vitro setting, CRC spheroids were cultured. The induction of TNF-α and TGF-β were shown to stimulate more TB formation in CRC spheroids and that cyclinD1 expression within the TB was higher in treated spheroids compared to control groups. Moreover, mouse AKPT organoids showed an increased in roundness, which indicates disruption in the formation of TB, in treated compared to control groups. These results suggested that cyclinD1 expressed within TB could have a potential role as a prognostic marker and may be used as a biomarker for TB formation.
In summary, data from this thesis have demonstrated potential biomarkers of TB and the relationship with tumour microenvironment in CRC. This will help understand the underlying mechanism of TB, and how they might interact with the surrounding microenvironment and could also pave the way for a future target therapeutic approach in CR
Improving hurricane-related flood risk assessments in urban areas: Integrating a Bayesian network modelling approach with an indicator-based approach
No full textGlobally, hurricane-induced flooding poses a significant threat, particularly for urban areas with typically high populations and infrastructure densities. Comprehensive flood risk assessment is becoming increasingly important for understanding the potential impacts of hurricane-induced flooding on an urban area. In this context, this research contributes to developing urban flood risk assessments using an indicator-based approach and Bayesian network modelling approaches, comprising three incremental steps.
The first step is based on an indicator-based flood risk assessment. Through a literature review, this research found that most assessments often overlook urban ecosystem elements (accounting for less than 15% of total indicators), focusing more on social and economic aspects. To address this gap, this research proposed a social-ecological systems (SES) urban flood risk assessment framework with 117 indicators. Flood risks were then analysed in Houston during Hurricane Harvey, applying the improved analytic hierarchy process (IAHP) and equal weighting methods. The results showed that the equal weighting method identified a wider range of high flood risk areas, with both methods indicating that Houston’s western parts were at the highest flood risks during Hurricane Harvey.
The second step aims to enhance the proposed framework with a Bayesian network (BN) model. This innovative approach can overcome shortcomings of the indicator-based method, such as its inability to incorporate new information and quantify uncertainties. The results showed that Houston’s western and north-eastern regions faced the highest flood risks during Hurricane Harvey. Overall, the BN model’s performance was largely in line with an indicator-based approach, with over 80% similarity in outcomes.
To further enhance the BN model, the third step was to develop a dynamic BN (DBN) model to obtain a more detailed understanding of urban flood risks. In this model, two hazard nodes were specifically designated as dynamic, while exposure and vulnerability indicators were treated as static. The comparison of results from the BN model and the DBN model shows that the latter efficiently captures the dynamic nature of flood risks.
Overall, the research offers a perspective on the enhancement of the indicator-based approach through the BN modelling approach. This research can help in identifying more effective disaster risk management strategies, offering insights into the evolving nature of flood risks in urban settings
An ethnographic exploration of the everyday lives of 51 people with heroin use experience in rural Dumfries and Galloway, Scotland
Get PDFScotland has a profound public health crisis related to drug use with the unenviable position of holding the highest drug related deaths rate in Europe. Approximately two to three people die from drug use every day in Scotland. The gravity of this situation warrants immediate, comprehensive and efficacious strategies to address the underlying factors that are contributing to this national crisis. One-fifth of people who use heroin in Scotland live in rural areas. Prevailing research exploring heroin use, however, focuses mainly on urban contexts. This thesis explores the “everyday lives” of rural people who use heroin within context and framework of the then Scottish Drugs Policy, “The Road to Recovery” (2008-2018) (Scottish Government, 2008). It investigates and scrutinises how people who use heroin in a rural setting acquire, administer and finance their heroin use, illuminating the significance of “recovery” from their perspectives. Examining the experiences of people who use heroin within a rural setting, provides an avenue to comprehend how aspects of rural living, such as transportation challenges, employment dynamics and lack of anonymity inherent within rurality, impact, shape and influence their daily realities. People who engage with heroin use in rural settings are notoriously difficult to locate and involve, posing challenges not only for research but policy development and practical interventions specifically targeted at their requirements given the background of the public health crisis.
This thesis presents the findings of a year-long ethnographic research study conducted throughout 2018 in rural Dumfries and Galloway in South-West Scotland. The study employed nonparticipant observation, participant-observation, field notes, and interviews involving 51 people who used heroin and 20 recovery service staff. Braun and Clark’s (2006, 2019) reflexive thematic analysis framework was adopted and adapted to analyse my data, the research aim was to determine how this group perceived their everyday role and place within their rural communities. It explored the unique challenges and opportunities inherent in a rural setting and investigated what recovery means to people who actively use heroin through their lived realities. The analysis uncovered the distinctive and unique experiences of people who use heroin in rural Dumfries and Galloway, highlighting the challenges and opportunities that presented in relation to substance use and recovery. These experiences frequently contradicted themes identified in research undertaken in urban settings.
The findings indicate that participants did not problematise heroin use. Instead, their heroin use was characterised by a high degree of regulation, evidenced from the findings. Regulated heroin use was shaped by accessible and free methadone maintenance treatment prescriptions and the presence of sympathetic and supportive staff within drug recovery services. Their familiarity within their wider communities shielded them in the main from feeling stigmatised and promoted inclusivity and even protection. Exploring this group’s everyday lives uncovered that they were not solely people who engaged with heroin use; they were also employees, partners, friends, and parents, who perceived their lives as “normal”, functional, content and fulfilling despite their long-term heroin use. Without exception they clarified that they were not seeking recovery in the form of abstinence. Instead, they engage with recovery services to access prescription methadone and other benefits provide by these services. Recovery services capitalised on these opportunities to engage in health promotion and education with their “clients.”
Similarities to urban settings were also observed, including similar numbers of people who used heroin per capita, increased poly-drug use, overdose incidents, and increasing drug-related deaths. Distinctive factors were also noted, such as negligible homelessness, widespread adoption of naloxone kits and training, reported low purity of heroin, a high proportion of older people who used heroin, rejection of public heroin use and communal environments during heroin use. Unexpected findings extended to the accessibility of heroin even within the remotest of areas, regulated and controlled use of heroin, reliance on state benefits rather than criminality for cash generation, a decline in heroin use and limited recovery services operating within traditional 9-5 hours five days per week. This thesis makes a meaningful contribution to the literature by exploring the everyday lived experiences of people who use heroin, identifying the unique challenges and opportunities associated with rural substance use and recovery highlighting a new concept termed “recovery inertia”. Above all, it offers a voice to an overlooked and unrepresented population
Targeting muscarinic receptors in treating and slowing the progression of neurodegeneration
Get PDFBackground
Alzheimer’s disease (AD) is a progressive neurodegenerative disease and the leading cause of dementia, which is projected to affect 1 million people by 2030. Despite the recent conditional approval of potential disease-modifying treatments by the Food and Drug Administration, there remains an urgent social and economic need for treatments which not only alleviate symptoms of AD but also slow its clinical progression. Loss of brain cholinergic innervation is a key hallmark of AD; inhibiting the breakdown of acetylcholine at synapses provides short-term symptomatic relief but produces dose-limiting side effects due to overactivation of peripheral receptors. The therapeutic effects of these drugs are largely attributed to increased activation of the M1 muscarinic acetylcholine receptor (mAChR) subtype, which has led to drug discovery programmes aimed at identifying positive allosteric modulators which directly enhance activation of the M1 mAChR to avoid adverse effects mediated by peripheral mAChRs.
M1-selective compounds have shown promise in preclinical studies, including delaying the onset of terminal disease in a prion model of neurodegeneration, but no compounds have yet been approved for clinical use. Translation of preclinical efficacy to effective treatments that slow the progression of AD requires robust biomarkers that can be longitudinally monitored to investigate the ability of drug candidates to modify the course of neurodegenerative disease. AD produces robust and progressive changes in brain oscillatory activity recorded using electroencephalography (EEG) or magnetoencephalography (MEG), which reflect the large-scale coordination of neuronal activity. Consequently, identifying whether animal models of aspects of AD produce similar alterations to neuronal oscillations can support the use of such models for preclinical testing of drug candidates.
Main aims
The first major aim of this thesis was to determine the impact of a model of progressive terminal neurodegeneration on cognition-relevant neuronal oscillations. The second aim was to identify electrophysiological signatures of selectively enhancing the activity of M1 mAChRs. The third aim was to robustly quantify the impact of muscarinic deficit, an additional feature of AD, on neuronal oscillations. The final aim was to assess the impact of potentiating M1 mAChRs on the electrophysiological signatures of muscarinic deficit.
Methods
These studies utilised a wireless electrophysiological recording system and electrodes implanted into the skull to record bulk neuronal activity from the brain surface of mice. This approach was employed to maximise the translatability of findings to human EEG recordings. Progressive terminal neurodegeneration was modelled using a prion intracerebral injection model. One cohort was used for behavioural characterisation and to determine the time to onset of early symptoms and terminal disease. Subsequent cohorts underwent surgical implantation of recording electrodes and weekly electrophysiological recording in the home cage until they reached the terminal disease end point. For pharmacology experiments, wild-type mice were implanted with surface electrodes and underwent recording at timepoints before and after dosing with M1-selective compounds and/or the non-selective muscarinic receptor antagonist, scopolamine.
Outcomes
The prion model exhibited a slowing of peak theta frequency, which is also a key electrophysiological signature of AD, suggesting that this is a marker of neurodegenerative disease progression which may be a useful biomarker for preclinical assessment of the disease modifying potential of drug candidates. The most robust effect of directly activating or enhancing the activation of M1 mAChRs was an increase in gamma power, which was reduced by muscarinic blockade, suggesting bidirectional modulation by changes in muscarinic activation. However, the effect of muscarinic blockade was not overcome by enhancing the activation of M1 mAChRs at a dose that restores memory deficits in prion mice
A framework for effective intermediate task selection in transfer learning
Get PDFThis thesis investigates strategies to improve the performance of natural language processing (NLP) models across diverse tasks, particularly in environments with limited training data. Central to this investigation is the concept of transfer learning, a method where a model developed for one task is repurposed as the starting point for a model on another task. Determining which model will yield improved performance on a specific task is a complex and non-trivial challenge. This complexity arises due to the varying natures of tasks, the intricacies of model architectures, and the unpredictability of their interactions. Accurately estimating which models will be most effective before committing to extensive training can provide substantial benefits, including significant reductions in runtime, environmental impact, and other associated costs. To address this challenge, we propose a framework designed to determine, from a pool of candidate models, which one will provide the greatest performance enhancement for a given task. This framework consists of five components, each addressing a particular concern in selecting tasks for transfer. Parallel to this, and running continuously throughout the process, is the Cost Estimation background process. This module evaluates the resource efficiency of all other components, ensuring that the model development and adaptation processes are both effective and sustainable. The Domain Adapter Generation component involves developing resource-efficient models using training documents from various text-based tasks. The Domain Transfer Analysis component involves evaluating the models created in the previous stage on documents other than those they were originally trained on, providing an understanding in how these models perform on different types of textual data. The Representation Construction component involves the development of profiles or “representations” of each task based on, for example, terms or linguistic characteristics. These representations are intended to be expressive of the features of the underlying data, which we use in subsequent stages of our analysis. The Divergence Estimation component systematically quantifies the degree of variation between different representations through the use of statistical methods. By assessing the divergence between task-specific representations, this component helps identify which intermediate task models exhibit the most promising alignment for a specific target task. Finally, in the Intermediate Task Selection component uses the divergence data to rank tasks by their potential to improve model performance on a given target task. This ranking provides guidance on which intermediate task models, when used to transfer to the target task, are most likely to yield the best performance. In addressing the challenge of identifying tasks that are conducive to effective transfer learning, this thesis places a significant emphasis on evaluating representations against the performance scores derived from the Domain Adapter Generation stage. The core of this evaluation lies in assessing the “effectiveness” of these representations. Effectiveness, in this context, is defined as the capacity of representations to accurately estimate the most beneficial ordering of task combinations. This estimation is based on comparing the outputs of the divergence measures with the inherent ordering of tasks according to their relative transfer gain. Here, transfer gain is measured by the performance ranking of models that have been adapted from intermediate tasks to target tasks, where intermediate tasks are typically tasks abundant in training data, which are then used to transfer knowledge to resource-scarce target tasks that we would like to improve performance on. The theoretical basis of this approach is rooted in a fundamental principle of transfer learning: tasks with higher similarity in their representations are expected to offer greater improvements in model performance when transferred to a target task. Consequently, the thesis investigates the relationship between task similarity, as quantified by our divergence measures, and the actual performance gains observed in transferred models. By analysing the correlation between divergence scores with the model performance rankings across tasks, we aim to validate the hypothesis that task similarity, in terms of representational divergence, is a key predictor of transfer success. This correlation not only provides a practical method to predict the projected effectiveness of task combinations but also offers insights into the nature of transfer learning itself, shedding light on which task characteristics most significantly impact model adaptability and performance enhancement. To evaluate the effectiveness of our framework, we simulate a scenario akin to a user employing the framework to “search” for the most suitable model to transfer to their specific target task. Traditionally, this process would involve exhaustive training and evaluation of all candidate tasks—often a time-consuming and resource-intensive process. We posit that, by predicting which tasks are likely to yield the largest performance gains ahead of time, through the analysis of task similarity, we can substantially improve the accuracy of task selection and also significantly reduce the time and resources required to find effective task pairs. Our framework allows users to bypass the labour-intensive cycle of trial and error, directly focusing on task combinations that are most likely to enhance their model’s performance on a target task. The central contributions of this thesis are the introduction of an effective and efficient intermediate task selection framework for transfer learning in natural language processing. This thesis draws from a diverse range of experiments, covering a broad range of NLP domains and experimental settings, to validate and refine the framework. The experiments presented in this thesis demonstrate the potential of task selection approaches to provide more efficient, sustainable, and impactful practices in the field of transfer learning
Sequential measurement in quantum learning
Get PDFIn an increasingly quantum world with more and more quantum technologies nearing practical use, the importance of interacting directly with quantum data is becoming clear. Although doing so often leads to advantages, it also presents us with some uniquely quantum challenges: for example, information about a quantum system cannot, in general, be extracted without disturbing the state of the system. In this thesis, we primarily focus on how performing a learning task on quantum data disturbs it, and affects one’s ability to learn about it again in the future. In particular, we focus on the learning task of unsupervised binary classification, and how it affects quantum data when it is performed on a subset of it. In such a binary classification task, we are given a dataset that is made up of qubits that are each in one of two unknown pure states, and our aim is to cluster, with optimal probability of success, the data points into two groups based on their state. To investigate how well we can perform this task sequentially, we first consider a base case of a three-qubit dataset, made up of qubits that are each in one of two unknown states, and investigate how an intermediate classification on a two-qubit subset affects our ability to subsequently classify the whole dataset. We analytically derive and plot the tradeoff between the success rates of the two classifications and find that, although the intermediate classification does indeed affect the subsequent one in a non-trivial way, there is a remarkably large region where the first classification does not force the second away from its optimal probability of success. We then describe this scenario as a quantum circuit and simulate the tradeoff using Qiskit’s AerSimulator. Following on from this, we go on to investigate whether an intermediate measurement can leave a subsequent one unaffected in the more general setting of an n-qubit dataset, again made up of qubits that are each in one of two unknown states. We see that numerics hint that nothing about the order of the qubits in a (n − 1)-qubit dataset can be learnt without affecting a subsequent classification on the full dataset. We make steps to prove that this is indeed the case and show that an immediate consequence of this is that, for some m > 1, a non-trivial intermediate classification on n−m qubits will always negatively affect a subsequent one on all n qubits. We conclude this line of work by deriving two bounds to how successful an intermediate classification of n − 1 qubits can be without affecting the following n-qubit one, hypothesising that one of these is optimal.
We then shift our focus to the field of indefinite causal order (ICO). Motivated by ICO’s connection to non-commutivity, we explore the idea of implementing quantum key distribution (QKD) in an indefinite causal regime. After showing that it is possible to share a key in an ICO, we find that, unlike other QKD protocols in the literature, eavesdroppers can be detected without publicly discussing a subset of the shared key. Indeed, we show that this is true for any individual attack in which the eavesdroppers abide by the causal structure chosen by the sharing parties. Further, we prove the security of this protocol for a subclass of these individual attacks. We then ask whether this “private detection” is a truly consequence of ICO and show that there is a definite causal ordered strategy that appears to yield the same phenomenon. Although we note that there are hints of some more subtle differences between the definite and indefinite causal cases, we conclude that carrying out QKD in an ICO is unlikely to offer any advantage, at least when considered in the form that we did. Finally, we close this thesis by summarising what we have found and noting some possible directions for future study
Additive manufacturing-enabled multifunctional cellular composites: Characterization, experiments and modelling
No full textAbstract not currently available
Manual vs automated quantified analysis of Distributive Fluvial Systems (DFSs)
No full textAbstract not currently available
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! 👇