21684 research outputs found
Sort by
The effect of prolonged glucose and insulin stimulation on mitochondrial function in endothelial cells
No full textAbstract not currently available
Transforming museum exhibits using interactive mixed-reality
Get PDFMuseums play a crucial role in preserving and communicating cultural, historical, and scientific knowledge. However, traditional exhibition methods often limit visitor engagement and comprehension. This research investigates how mixed-reality (XR) technologies (integrated with Artificial Intelligence (AI) and Computer Graphics (CG)) can transform visitor experiences within museum environments, with a focus on the Hunterian Museum’s Science Collection.
Building on these insights, a prototype mobile and mixed-reality application was designed and developed to enable immersive exploration of artefacts in 3D, interactive storytelling, and adaptive learning experiences. The system architecture integrated wearable interfaces, edge computing, cloud services, and AI-driven modules for speech recognition, natural language processing, and personalized recommendations.
After finishing the implementation of my App, functional and performance evaluations are operated. It identified optimal solutions for enabling natural user interaction, while user studies demonstrated significant improvements in engagement, comprehension, and satisfaction compared to traditional exhibition methods. The findings indicate that AI-enhanced XR applications can provide inclusive, dynamic, and educational museum experiences, bridging the gap between physical artefacts and visitor understanding. This research contributes a practical framework for integrating immersive technologies into museum practice and offers insights for future developments in adaptive, user-centered digital heritage experiences
Hydrological responses to wetland changes in the Yangtze River Basin
Get PDFAs the most productive ecosystems in the world, wetlands play a vital role in carbon cycling, climate change mitigation, socio-economic development, and natural disaster protection. The Yangtze River Basin (YRB) contains 40% of the national wetlands and the most frequent floods in China. In recent decades, the abundant wetland resources in the YRB have experienced substantial changes due to the climate change and human activities, significantly affecting the flood risk. Due to the lack of a long-term time series wetland dataset with comprehensive categories for the YRB, the effects of wetland variations on flood risk, as well as improved assessments of past and future flood risk incorporating wetland dynamics, remain underexplored. Moreover, wetland-related flood risk mitigation efforts in the YRB are less widely adopted. It is crucial to address these gaps for enhancing the sustainable wetland management and flood risk mitigation in the YRB.
This thesis aims to achieve three primary objectives: 1) to establish a long-term time series wetland classification dataset with the comprehensive categories in the YRB and analyze driving forces of their variations; 2) to investigate the long-term wetland effects on the flood risk in the YRB based on an improved GIS-based multi-index flood risk assessment model incorporating wetlands input; and 3) to predict the future flood risk with wetland effects in the middle-lower YRB under climate change and socio-economic scenarios.
The Long-term Wetland Classification Dataset for YRB (LTWCD_YRB) with nine wetland categories from 1984 to 2021 was created by using the Random Forest machine learning classifier on the Google Earth Engine platform with 30m resolution Landsat images. The LTWCD_YRB revealed that: 1) the total wetland area of the YRB in 2021 was larger than that in 1984, with a constant increase in human-made wetlands and fluctuating natural wetlands; 2) aquaculture ponds expanded the most by 4,987 km2, while inland marshes in the source region exhibited the most fluctuations; and 3) seasonal changes in wetlands were prominent in the Poyang Lake Basin, Dongting and Honghu Lake Basin, and YRB source region; and 4) human activities were found to be more dominant than natural driving forces in affecting wetlands. The LTWCD_YRB offers a consistent agreement of wetland area variations with the other satellite-based wetland datasets in the YRB.
To investigate the long-term effects of wetland variations on flood risks in the YRB, this thesis developed an improved GIS-based multi-index flood risk assessment model incorporating the wetland input obtained from the LTWCD_YRB. The findings indicated that: 1) wetlands in the Taihu Lake Basin, Wanjiang Plain, Poyang Lake Basin, and Dongting and Honghu Lake Basin could mitigate flood risks, while wetlands in the Sichuan Basin have aggravated but limited impacts on flood risks; and 2) Precipitation in the Taihu Lake Basin and Poyang Lake Basin, runoff and vegetation cover in the Wanjiang Plain, GDP in the Taihu Lake Basin, population density in the Taihu lake Basin, Dongting and Honghu Lake Basin, and the Sichuan Basin are dominant flood risk indicators under wetland effects. The wetland-related suggestions to mitigate flood risks including maximizing stormwater storage capacity of wetlands and increasing vegetation coverage in urbanized and precipitated regions.
The flood risk prediction of the middle-lower YRB applied the improved flood risk model to assess the flood risk from 2021 to 2100 under the Shared Socioeconomic Pathways (SSP) 2 - Representative Concentration Pathways (RCP) 4.5 and SSP5-RCP8.5 scenarios. The results indicated that: 1) the high and very high flood risk areas will totally cover 38% and 40% of the total study area under the SSP2-4.5 and SSP5-8.5 scenarios by 2100, respectively; 2) the overall flood risk of the MLYRB was predicted to become severer by 2100 under both scenarios; and 3) there would be a prominent northward expansion of the high and very high flood risk areas in Jiangxi, Hunan and the southern part of the Taihu Lake Basin in Jiangsu.
In summary, this thesis provides the data support for the long-term wetland variations in the YRB, develops an improved flood risk model to investigate the long-term wetland effects on the flood risk and predicts the future flood risk incorporating wetland dynamics in the YRB. The efforts of thesis contribute to the sustainable wetland conservation and flood risk mitigation in the YRB, aligning with the United Nations Sustainable Development Goals
Investigating the potential importance of Hedgehog signalling in conferring Venetoclax resistance in AML
Get PDFHedgehog (Hh) signalling pathway is a highly conserved pathway, which plays a crucial role in the development, maintenance and regeneration of human tissues. Its role in the primitive and definitive haematopoiesis has been extensively studied. Its activity has been proven to be a critical mechanism in the pathogenesis of tumours, either solid or haematological. It has a vital role in promoting the proliferation and survival of acute myeloid leukaemia (AML) cells, including leukaemia stem cells (LSCs). Moreover, it contributes to therapy resistance. Several Hh signalling pathway inhibitors have been generated and are involved in AML clinical trials. The outcomes of these trials are promising. Glasdegib is a Smoothened (SMO) inhibitor, which has been approved in combination with low-dose Cytarabine (LDAC) for newly diagnosed cases of AML in patients who are not fit for extensive chemotherapy. This follows the promising results of the BRIGHT AML 1003 trial.
The introduction of Venetoclax, a BCL2 inhibitor, into clinical practice marked a revolution in the treatment of leukaemia, particularly since the anti-apoptotic B-cell lymphoma 2 (BCL2) proteins were observed to be overexpressed in different types of leukaemia and are associated with chemoresistance. Testing Venetoclax for the newly diagnosed AML patients in the VIALE-A and VIALE-C trials demonstrated promising outcomes and superiority over low-dose chemotherapy alone regimens. Thus, in AML, Venetoclax has been approved for use in combination with Hypomethylating agents (HMA) (such as Azacitidine or Decitabine) or LDAC in newly diagnosed AML patients who are unfit for intensive chemotherapy.
Before that time, Glasdegib regimens were more involved in clinical trials and clinical settings; however, the promising efficacy achieved by Venetoclax resulted in the widespread adoption of the Venetoclax-based regimens. Venetoclax resistance has been a critical challenge hindering the efficacy of several leukaemia-treatment protocols.
This thesis aimed for the repurposing of the clinically approved drugs, Glasdegib and Venetoclax, to combat therapy resistance. This work investigated the potential involvement of the Hh signalling pathway in Venetoclax resistance in AML and whether its role in Venetoclax resistance is dependent on anti-apoptotic BCL2 proteins or not.
Venetoclax-resistant (VE-RE) cell line models have been established to investigate the potential mechanisms of resistance with particular focus on the role of Hh pathway. Inducing resistance was accomplished by exposing the Venetoclax-sensitive cells of MOLM-13 cells to increasing concentrations of Venetoclax in an intermittent (group 1) or continuous (group 2) manner. Venetoclax resistance was validated in VE-RE MOLM-13 groups 1 and 2 by comparing their half-maximal inhibitory concentration (IC50) to Venetoclax with that of the wild-type MOLM-13. There was an approximately 209-fold and 418-fold increase in the Venetoclax IC50s in VE-RE MOLM-13 group 1 and 2, respectively.
VE-RE MOLM-13 groups 1 and 2 cells were profiled at the transcriptomic and genomic levels to identify potential mechanisms of Venetoclax resistance. Additionally, Venetoclax-treated (VE-TR) MOLM-13 cells at 2nM, 4nM, and 6nM were profiled, as well, to determine the trend of change and the onset of the mutations, as it is believed that Venetoclax resistance develops gradually over time, rather than immediately upon exposure to Venetoclax. The expression of BCL2 family members and Hh signalling pathway was found to be deregulated. BCL2, Myeloid cell leukaemia-1 (MCL1), BCL2 Like 1 (BCLxL), Suppressor of Fused (SUFU) and Patched 1 (PTCH1) were significantly changed from the sensitive cells to the resistant cells.
Screening VE-RE cells for the potential causes underlying Venetoclax resistance revealed interesting, promising contributors. Based on the observations reported in clinical studies, these contributors provide valuable insights into Venetoclax resistance. Thereby, further investigations are warranted to explore their therapeutic implication in VE-RE AML. Among these potential contributors are Mitogen-Activated Protein Kinase Kinase Kinase Kinase 1 (MAP4K1), KIT Proto-Oncogene (KIT), Mesothelin (MSLN), S100 Calcium Binding Protein A4 (S1004A), N-Myristoyltransferase 2 (NMT2), Mouse Double Minute 4 (MDM4) and Transmembrane BAX Inhibitor Motif Containing 4 (TMBIM4). Several deregulated genes were found to have a significant association with adverse patient survival based on multiple AML datasets.
VE-RE MOLM-13 groups 1 and 2 cells were tested against Glasdegib alone and in combination with Venetoclax. Based on our preclinical results, the combination experiments between Glasdegib and Venetoclax have improved the sensitivity of the wild-type (Bliss synergy score of 23.40) and VE-RE MOLM-13 cells of both groups to Venetoclax (Bliss synergy score of 8.59 and -5.33, respectively). Studying the cell cycle kinetics changes upon adding Glasdegib to Venetoclax revealed a significant increase in the percentage of cells in the sub G0 phase in wild-type MOLM-13 and VE-RE MOLM-13 groups 1 and 2.
This combined effect can be explained by either inducing the quiescent leukaemic cells from G0 phase by Glasdegib to be targets for the action of Venetoclax, Glasdegib-mediated apoptosis, or Glasdegib-induced downregulation of anti-apoptotic BCL2 family proteins.
The Hh signalling pathway was found to depend, to some extent, on anti-apoptotic BCL2 family proteins in inducing Venetoclax resistance. This was shown by the role of Glasdegib in the downregulation of BCL2, MCL1, and BCLxL. This was detected following the treatment of wild-type MOLM-13, VE-RE MOLM-13 groups 1 and 2 by two concentrations of Glasdegib.
Venetoclax resistance is a collaboration of multiple mechanisms that carry variable intensities upon continuous or intermittent exposure to Venetoclax. Continuous exposure was concluded to cause a higher degree of Venetoclax resistance.
Finally, this work established a valuable fact, in which Hh signalling pathway is involved in the emerging Venetoclax resistance. Furthermore, activation of Hh signalling in VE-RE cells can occur in a canonical or non-canonical manner. Glasdegib-mediated blocking of the canonical activation of the Hh signalling improves the apoptotic effect of Venetoclax on the VE-RE MOLM-13 cells.
SUFU, which encodes for a GLI Family Zinc Finger (GLI) sequestering protein, is a valuable biomarker that warrants further investigation, particularly as it is significantly downregulated in VE-RE MOLM13 of both groups. Moreover, it has been observed from studying its effect on patient survival that its low expression has a significant association with adverse patient survival in the early follow-up period in cohorts of either AML or other malignancies
From measurement to management of kidney function: population-based insights into biomarker discordance, risk factors, and outcomes
No full textChronic kidney disease is a major risk factor for cardiovascular diseases and mortality, yet both the measurement of kidney function and the management of kidney-related risk in the general population require further exploration. This thesis aims to move from measurement to management of kidney function by examining modifiable behaviours, identifying patterns of kidney function and their associations with adverse outcomes, and evaluating treatment strategies and their potential cardiorenal benefits.
A total of five published manuscripts were included in this thesis to achieve the above aims. These studies utilized data from the UK Biobank and several Swedish national registers.
The first study in Chapter 4 presents a systematic review and meta-analysis of intervention studies in generally healthy populations, examining how changes in physical activity influence kidney-related biomarkers. Higher physical activity was consistently associated with higher serum creatinine, while other biomarkers showed little or no change. The second study in Chapter 5 uses UK Biobank data to examine how changes in physical activity relate to changes in kidney function in more than 11,000 participants free of chronic kidney disease at baseline. The findings differed by biomarker. When rapid decline was defined using creatinine-based estimated glomerular filtration rate (eGFR), lower physical activity appeared protective, whereas cystatin C-based eGFR showed the opposite pattern, with higher activity associated with lower risk. The creatinine-cystatin C based eGFR showed no clear association.
The following two studies in Chapters 6 and 7 focus on discordance between creatinine-based and cystatin C-based eGFR. Chapter 6 synthesises evidence from 18 studies and shows that individuals with lower cystatin C-based eGFR relative to creatinine-based eGFR have substantially higher risks of mortality and cardiorenal events, while those with higher cystatin C-based eGFR tend to have lower risks. Chapter 7 extends this work in a large UK Biobank cohort, showing that 15.5% of the study population has discordantly lower cystatin C-based eGFR and that this discordance is strongly associated with all-cause mortality, particularly in older (65 years and above) and obese (BMI≥30 kg/m2) people, independent of established risk factors.
The last study in Chapter 8 shifts to pharmacological management in type 2 diabetes, comparing oral fixed-dose combination therapy with loose-dose combinations using Swedish national registers. In a propensity score matched cohort of 27,766 people, fixed-dose combinations were associated with a lower risk of heart failure, especially in older adults (65 years and above), with 47% of this benefit being mediated by improved medication adherence. No clear differences were observed for other cardiovascular or kidney outcomes.
In conclusion, this thesis provides population-based evidence that selection of kidney function biomarkers may play an important role in evaluating health outcomes, while intraindividual discrepancies in kidney function may reveal distinct physiological or pathological processes. In addition, pharmacological strategies may affect cardiorenal risk. Key limitations include reliance on estimated rather than measured GFR, potential residual confounding, the inability to establish causality inherent to observational analyses, and limited generalisability to more diverse populations
Integrated photonic circuits on SOI and SiNOI platforms for high-sensitivity sensing and reconfigurable multi-passband filtering
No full textIntegrated photonics offers compact and energy-efficient platforms for label-free biochemical sensing and agile spectrum control. This thesis establishes a unified design framework on silicon on insulator (SOI) and silicon nitride on insulator (SiNOI) that links two capabilities: enhancing light-matter interaction for refractometric sensing while constraining excess loss, and realising reconfigurable multi-passband photonic filters with predictable, programmable spectra. The work combines electromagnetic and coupled-mode modelling, reconstruction equivalent chirp design, heater-matrix control, foundry-compatible fabrication, and system-level characterisation using optical frequency comb sources and thermo-optic tuning. It demonstrates that sensitivity enhancement and spectral programmability can be obtained through spatially distributed design and algorithmically coordinated actuation, rather than by accepting large insertion-loss penalties.
For sensing, three-slot hybrid plasmonic waveguides are engineered to localise the evanescent field at the analyte interface while routing most optical power through low-loss analyte sections. This architecture supports stable C-band operation and provides a route to array-scalable sensors intended for monitoring biomolecular binding and concentration changes in microfluidic lab-on-chip diagnostics.
For filtering, phase-shifted sampled Bragg grating devices are synthesised using the reconstruction equivalent chirp (REC) technique to spatially decouple reflection points, enabling independent control of multiple passbands. Integrated micro-heater matrices provide reconfiguration and algorithm-assisted phase synthesis, combining self-adaptive differential evolution with local optimisation, coordinates multi-parameter tuning to obtain low-ripple spectra. Three architectures validate the framework: a dual-passband filter based on coupled phase-shifted cavities, a programmable four-phase shift sampled Bragg grating (4PS-SBG) on SiNOI enabling selectable multi-passband and stopband responses, and a cascaded superstructure grating with a microring resonator enabling optical-domain frequency multiplication and hopping.
Overall, the thesis provides design rules for co-optimising waveguide confinement, photon distribution, and heater matrix control in integrated photonics, underpinning scalable programmable photonic circuits and future heterogeneous integration with on-chip light sources
An investigation into performance monitoring in women’s elite rugby sevens, with consideration towards the Countermovement Jump (CMJ) Test and the GPS-Derived Metric High Metabolic Load Distance (HMLD)
Get PDFThis study examined the effectiveness of performance monitoring strategies in elite women’s rugby sevens, by evaluating the Countermovement Jump (CMJ) test and the GPS-derived metric, High Metabolic Load Distance (HMLD). While GPS tracking is widely used to quantify athlete workload, HMLD remains underexplored in rugby sevens, despite its potential to reflect training and match intensity.
Seventeen elite female rugby sevens players (age: 26.4 ± 2.9 years; height: 167.6 ± 5.2 cm; body mass: 70.5 ± 8.9 kg) were monitored across one competitive season. CMJ performance was assessed using the Output Sports V2 Sensor, and GPS data were collected from STATSports Apex Pro technology.
Backs outperformed forwards in the CMJ (40.78 cm vs. 38.02 cm) and accumulated greater HMLD in both training (560.68 m vs. 470.17 m) and competition (263.85 m vs. 216.64 m). Significant positional differences were observed across GPS metrics. HMLD correlated most strongly with total distance (r = 0.82), suggesting that in rugby sevens, HMLD behaves more as a volume metric, rather than a measure of intensity. Current speed thresholds may not adequately capture the sport’s intensity, highlighting the need for sport-specific interpretations. CMJ performance improved over the season, though not linearly, suggesting varying levels of fatigue and physiological adaptation. HMLD volumes reflect the fluctuations and variability in training and competitive demand across a rugby sevens season.
This study highlights the value of integrating CMJ testing with monitoring of HMLD across a competitive season in elite women’s rugby sevens. The CMJ provides insight into neuromuscular function and lower-body power, whilst HMLD reflects the physical demands of training and competition. Together, they offer a comprehensive view of athlete readiness.
This research addresses the gap in the literature surrounding the application and interpretation of HMLD in women’s rugby sevens, and the disparity in research in the women’s game. These results contribute new insight into the use of HMLD in an elite rugby sevens environment, and emphasise that alongside CMJ testing, HMLD monitoring can bring significant value to an elite training programme. The metric has the potential to be individualised to athletes, for the most effective approaches in performance optimisation, injury prevention and return-to-play strategies
Visualising infections across scales
Get PDFThe ability to visualise biological structures and processes at the microscale has been instrumental in the development of modern medicine and the control of disease. In recent years, technological innovations have enabled imaging of large, mesoscale tissues, from organs to whole mammalian organisms. However, the development of these methods demonstrates the inherent weaknesses of imaging modalities designed to function at specific length scales. While modern microscale methods enable imaging at resolutions beyond the diffraction limit of light, resolving objects tens of nanometers apart, the field of view and working distance of these systems prevents analysis of mesoscale samples. Conversely, cutting-edge mesoscale imaging methods are capable of imaging whole mice, but lack the resolution of modern confocal microscopes. Holisic analyses are crucial in understanding biological processes and only by designing imaging pipelines capable of coupling high resolution microscopy with large sample mesoscopy can we develop a complete understanding of processes occurring within organisms. Influenza A virus (IAV) exemplifies the need for multiscale imaging approaches. Viral replication is governed by microscale, intracellular processes, while pathology and spread on the scale of the whole lung dictates virus transmission and patient outcomes. Here, I develop methods for imaging tissues across length scales, primarily using IAV as a model pathogen. On the mesoscale, I develop means of visualising infections in whole murine lungs, as well as skin, nerve, and bone tissue without the need for sectioning. On the microscale, I develop computational tools and laboratory assays capable of visualising interactions between viruses, elucidating the processes governing the emergence of viral strains with pandemic potential. By designing holistic imaging pipelines capable of analysing IAV infections across scales, these insights develop a more complete understanding of the spatial context of IAV infection, the interactions between viral populations, and the relationship between microscale processes and meso- and macroscale pathology
The search for strangeonium resonances: partial wave analysis of polarised photoproduction reaction γp → ηφp at the GlueX experiment
Get PDFThe strong force gives rise to a large hadron spectrum, many areas of which are still not well understood. Progress in mapping out these regions is vital for building better models of the strong force and verifying the predictions of its quantum field theory, Quantum Chromodynamics. This thesis focuses on the study of the so-called strangeonium states, which are mesons with s¯s quark content, and in particular on members of the φ meson series φ(1680) and φ3(1850). The former is a radially excited vector that has been seen by both e⁺e⁻ collider and photoproduction experiments, but whose reported mass does not match between the two, with observations from photoproduction being systematically higher; the latter is a tensor that has not been measured since the 1980s and never in experiments with high statistical power.
The data analysed in this thesis was collected at the GlueX experiment, which has produced the world’s largest photoproduction dataset to date since starting to run in 2016. A 9 GeV linearly polarised photon beam incident on a liquid hydrogen target was used to produce a variety of mesonic and baryonic resonances. The GlueX spectrometer allows both neutral and charged final states to be exclusively reconstructed with good resolution and angular acceptance. Results from the partial wave analysis of decay channel γp→ηφ p→K⁺K⁻γγ p are presented.
After signal was isolated from background using a combination of event selection criteria and statistical subtraction via the sPlot method, an extended maximum likelihood fit to angular decay distributions allowed the extraction of partial wave amplitudes. A partial wave consistent with the φ(1680) was observed, but no clear evidence of a φ3(1850) was found. The obtained amplitudes indicated a preference for natural parity exchange. A fit to the intensity of the wave identified as the φ(1680) resulted in a Breit-Wigner mass more compatible with the e⁺e⁻ measurements. A cross-section was also calculated, allowing an order-of-magnitude estimate that matched the existing predictions.
The results provide evidence against the additional vector meson states that have been previously hypothesised to explain the higher observed mass of the φ(1680) in photoproduction. They should also aid the development of models of the strangeonium spectrum
Preclinical investigation of the BCL-2 family protein MCL-1 as a potential therapeutic target in breast cancer
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! 👇