15234 research outputs found
Sort by
A Biophysics and Fluorescence Guided Surgery Inspired Artificial Intelligence Classifier for the Identification and Classification of Colorectal Cancer
No full textCancer detection within pathology remains one of the biggest challenges facing clinicians with its presence, or absence, significantly impacting the subsequent patient journey and experience. This dichotomy is demonstrated in the management of significant colorectal lesions with the presence of malignancy, in general, mandating major surgery in the form of a formal bowel resection and lymph node harvesting (at times with the formation of a stoma) whereas benign tumours can be adequately managed via a more limited local excision or even in situ ablation. This thesis introduction begins by exploring the current state of the art with respect to cancer detection and treatment in significant colorectal lesions, including a review of newly emerging Artificial Intelligence (AI) technologies to help clinicians in such endeavours. The second component comprises two parts. The first assesses clinician ability to accurately identify the presence of malignancy within significant rectal lesions in a manner that resembles current diagnostic pathways. As this thesis ultimately revolves around the development of an AI and fluorescence augmented tissue classifier, the second component analyses physician ability to assess and interpret ICG fluorescence signals using the naked eye. The development and validation of a fluorescence augmented AI classifier for the identification and classification of cancer represents the majority of this work and was carried out in the form of a clinical trial (NCT04220242). This thesis details the development and application of a novel, biophysics inspired method of ICG fluorescence angiography assessment to the characterisation of lesions of the colorectum. The expansion of the biophysics inspired fluorescence classifier developed to other areas of the body, namely the liver is also explored. As the principles exploited are based on differentials between tissues of differing nature (cancer vs benign) and are agnostic of tissue location such methodology can be deployed for the characterization and delineation of liver metastases with high levels of accuracy also. Study at a microscopic level was also performed to provide insights in the distribution and trapping of ICG within human tissues which has not previously been done in human colorectal cancer tissue. This thesis concludes with a discussion of the implications of this work on the field of fluorescent guided surgery and details the next steps to ensure continued development and application of the established methodologies.2025-11-13 JG: Author's signature removed from PD
Molecular profiling of AMA oocytes: from basic biology to improving fertility in aged women
No full textIn a world where more and more women are delaying motherhood, age-related fertility decline poses a significant challenge. Age-related infertility is primarily attributed to oocytes, as they diminish in both quantity and quality as women age. The deteriorating quality of ageing oocytes is closely associated with aneuploidy and alterations in cytoplasmic content. Nevertheless, the precise mechanisms driving this decline in oocyte quality remain unclear. The objective of the thesis studies was employed cutting-edge omic technologies to unravel the molecular intricacies of the oocyte maturation process and to determine the extent to which they are impacted by age and reflected in oocyte quality. Comprehensive molecular profiling was conducted on a substantial number of human oocytes sourced from women spanning nearly two decades of reproductive life and two oocyte maturation stages. This effort resulted in the creation of an innovative dataset. This research uncovered crucial insights into the transcriptomic, DNA methylation, and proteomic landscape of human oocytes, establishing a significant connection with maternal age. The findings highlight distinctive changes in both transcript and protein abundance during oocyte meiotic maturation, while CpG methylation levels exhibit remarkable consistency. Notably, most of these changes are evident in both young and advanced maternal age oocytes. Additionally, age-related modifications in oocytes are predominantly observed at the protein level. Specifically, several proteins crucial for meiosis control and proteostasis exhibited a decline with age, particularly within immature oocytes. This includes noteworthy changes in the proteasome complex which has also been validated to have a pivotal role in human oocyte maturation. The observed alterations in proteasome quantity, combined with other identified changes, likely contribute to the reduction in oocyte quality. The resulting dataset holds immense potential for informing the scientific community about oocyte ageing, offering valuable insights that can aid ongoing and future age-related studies in humans and other species. In conclusion, proteasome complex is proposed as a promising target for future interventions and treatments aimed at enhancing oocyte quality in reproductively aged women
Invasive plants detection and distribution patterns analysis through Self-Attention enhanced semantic segmentation in UAV imagery and Moran's Index
No full textThe development of sustainable agriculture necessitates the rapid identification and efficient removal of invasive plants. Traditionally, the investigation of invasive plants relies on manual surveys, which are prone to subjective errors and require significant human labor. This study introduces an innovative approach for accurately identifying and efficiently eliminating invasive plants. High-resolution aerial images were captured using drones, and a novel semantic segmentation network, based on DeepLabV3+ and the Self-Attention mechanism, was developed to reliably identify two globally distributed invasive species: Erigeron annuus (L.) Pers. and Erigeron canadensis L. at the pixel level. Experimental results from orchard imagery revealed that the proposed approach demonstrated remarkable performance, achieving a mean Precision (mPrecision) of 0.972, a mean Intersection over Union (mIoU) of 0.947, and a mean Pixel Accuracy (mPA) of 0.973 for its overall effectiveness. A high-resolution species distribution map at pixel-level was generated using the results of this model. The study further explored ecological analysis methods of this distribution map, and successfully calculated the coverage area, coverage rate, global Moran's index, and local Moran's index of the two invasive plants. The findings revealed that the intraspecific distribution patterns of both species are characterized by clustering, with Global Moran's Index values of 0.03 for Erigeron annuus (L.) Pers. and 0.23 for Erigeron canadensis L. The clustering map facilitates the rapid identification of invasive plant cluster centers, enabling more targeted weed control measures. The efficient pixel-level invasive plant identification model and species distribution pattern analysis proposed in this study holds significant implications for agricultural production and ecological surveys. They support precise and rapid invasive plant control and to reduce pesticide use. The proposed method can also be implemented on other platforms to provide fast, flexible, and accurate invasive plant mapping and precision agriculture applications
An environmental policy-aligned information landscape for the built environment
No full textIn the context of the global climate emergency, the built environment stands as a critical leverage point for decarbonisation due to its scale and resource intensity. Environmental policies increasingly push for reduced resource consumption through efficient procurement, operation, and disposal of buildings and infrastructure. Life cycle analysis (LCA) has emerged as a key methodology to support these policies, offering a comprehensive view of resource use. However, the built environment's slow adoption of digital practices poses challenges in consistent LCA implementation. This thesis identifies a gap between environmental policies which require life cycle insight, and the sector's limited capabilities, risking unworkable policies. The thesis hypothesises that “the environmental policy trajectory in the built environment necessitates a fundamental reorganisation of the information landscape based on a harmonisation of the broad socio-techno-economic requirements of governance and practice.” To test this, the research presents three interrelated contributions. The first contribution expands domain knowledge in asset management through a qualitative analysis of practitioner interviews, highlighting systemic issues like organisational complexity and resource scarcity. It positions asset managers as key agents of change with their life cycle perspective, providing an evidence base for Life Cycle Asset Information Management (LCAIM) priorities. The second contribution develops and tests a novel methodology for scaling financial life cycle cost analysis using a reference architecture-based approach. Validated through case studies and stakeholder focus groups, this approach demonstrates effectiveness over traditional IT development strategies. It calls for reforming the information landscape to better align with policy ambitions, offering practical blueprints for integrated information management. The third contribution presents an environmental policy-aligned decision-support system prototype that integrates governance and industry needs into a novel LCAIM ontology. By using a knowledge graph to connect distributed asset data, the system supports multi-stakeholder information needs throughout the asset life cycle. Validated through a real case study, the system demonstrates its ability to address real-world challenges, culminating in a holistic framework describing an information landscape compatible with environmental policy goals. This research concludes that existing technology can address key technical challenges in LCAIM, suggesting the need for systemic reforms to tackle broader barriers to sustainability in the built environment. Overall, the work engages diverse stakeholders and presents a decision-support prototype that aligns with environmental policy, advocating for a shift in focus toward systemic issues crucial for achieving environmental sustainability objectives
Advancing Fluorescence Angiography
No full textIntroduction: Intravenously injected Indocyanine green (ICG) can be visualised using surgical near infrared (NIR) cameras as it permeates tissues and vasculature. ICG fluorescence angiography (ICGFA) is used intraoperatively to address malperfusion related complications in colorectal and reconstructive surgery. However, ICGFA technique, interpretation, and equipment is unstandardised and is impacted by patient physiology. This research seeks to analyse current ICGFA practices and equipment. Subsequently, this work aims to investigate and develop quantitative ICGFA (QICGFA) methodologies, with a view to advancing contemporary applications and also developing new uses. Methods: Commercially available NIR systems were assessed for fluorescence variations relating to target positioning using stereotactic sensors. Regarding operator ICGFA interpretation, recordings were integrated into a digital interactive platform to evaluate decision-making and test the impact of proposed protocols on interpretation consistency. Post hoc quantification from diagnostic and therapeutic colorectal ICGFA recordings as well as from reconstructive procedures were analysed for associations to tissue status and postoperative outcomes. Subsequently, novel computational and clinical workflows were developed that could deliver surgical guidance and complication prediction based on comparing the target tissue perfusion to similar regions or the same tissue at earlier time points. Results: The presented NIR signal varied by system, setup (e.g. camera-scope angle configuration), and target positioning. While expert users unconsciously appreciated fluorescence fluctuations, proposed protocols which controlled for spatial variables did not improve user interpretation. The developed methodologies demonstrated feasible QICGFA perfusion pattern assessment and presentation for autologous and implant-based breast reconstructions, with these profiles being used to train complication-predicting artificial intelligence methods for the latter. Colorectal mucosa QICGFA curve patterns were associated with polyp pathology. Regarding post resection anastomotic guidance, the proposed clinical and algorithmic workflow comparing bowel ICGFA to an earlier time point demonstrated recommendations which were concordant with expert level judgements. Conclusions: Supported by an understanding of the underlying science, QICGFA perfusion patterns can be determined in a manner that potentially supports surgical decision-making for optimal healing outcomes and pathological diagnosis. Technical and clinical considerations associated with current practice need careful consideration for best use of these capabilities.2025-11-13 JG: Author's signature removed from PD
The use of synthetic hand pose data in training sign language fingerspelling models
No full textThis thesis presents a deep learning model that was developed to recognise sign language fingerspelling sequences and that was trained entirely on computer generated training data. This important contribution to sign language recognition (SLR) confirms that the distributional shift between the synthetic and real world domains can be mitigated through the use of suitable data augmentation. Furthermore, the model architecture is significantly more efficient than those proposed in previous SLR works; it enhances its suitability as a real-time assistive application for a Deaf user
Characterisation of the Immune Landscape in Multiple Myeloma
No full textMultiple myeloma (MM), a haematological malignancy, is characterised by a clonal proliferation of plasma cells in the bone marrow (BM). Incidence rates have been increasing since 1994. Whilst 5-year survival rates have also increased, MM remains an incurable frequently relapsing disease. The development, progression and resistance to therapy in MM are intricately connected to the heterogeneous tumour microenvironment (TME), the BM. Immune cells in the MM BM are highly altered. Hypothesising that dysfunction arises in the major cytotoxic cells present in MM BM, alongside an alteration of suppressive subsets, this thesis performed a deep characterisation of the major lymphocytes of the TME. Using a combination of flow cytometry immunophenotyping, single-cell RNA sequencing (scRNA-seq) and functional analysis of MM patient samples, we elucidate the cellular factors associated with MM. We reported increased frequencies of immune subsets in MM including CD57+ T cells and CD56-CD16+ NK cells that may be relevant targets for novel therapeutic strategies. Functional studies demonstrated a reduced response to stimulus in both the T and NK cells. To understand the immune suppressive mechanisms, we characterised the frequency and phenotype of regulatory cells. Decreased frequencies of myeloid suppressor cell subsets were found, and deep immunophenotyping of T regulatory cells found a highly suppressive phenotype. High heterogeneity was observed in the MM plasma cells. We characterised the changing immune cell landscape in daratumumab treatment using a combination of transcriptional and functional studies, and we reported increased cell subsets and a shift towards T cell effector memory post-daratumumab treatment. We also demonstrated the use of intracellular epitope labelling in combination with surface AbSeq in a scRNA-seq. This research addressed the gap in the BD Rhapsody technology by demonstrating the successful use of this assay to identify key cell subsets. The work in this thesis improves our understanding of the immune landscape in MM and may inform the future development of cellular therapies
Cultivation of Microalgae Nannochloropsis for Bioremediation of Lactose-Enriched Dairy Waste Streams and Co-Production of High-Value Biomass
No full textThe EU is one of the world's largest milk producers (ca. 150 million metric tons). Dairy processing generates significant amounts of nutrient-rich wastewater and by-products, which pose significant environmental problems due to high organic matter contents (e.g., lactose, proteins and fats), BOD level and COD level. Microalgae biotechnology is an attractive solution for dairy waste treatment due to its potential to achieve simultaneous bioremediation and co-generation of useful biomass and high-value compounds in a circular economy model. The main objective of this thesis was to develop a sustainable and cost-effective microalgae-based strategy for lactose enriched dairy waste stream (specifically whey powder solution) using Nannochloropsis, an autotrophic lipid-rich microalgal genus with a typically high omega-3 polyunsaturated fatty acids (ω-3 PUFAs) content in the form of eicosapentaenoic acid (EPA or C20:5). The application of Nannochloropsis can alleviate dairy systems reliance on a linear “collect-treat-discharge” practice of handling waste and instead promote a more sustainable, cost effective and circular practice whereby valuable ‘waste’ resources are continuously recovered and reused. The study presents an integrated approach which addresses fundamental research gaps in Nannochloropsis cultivation on dairy waste streams (such as the mechanism of lactose assimilation) and transfer these insights into the development of an effective pre-treatment and growth strategy to attain optimal bioremediation. The specific objectives of the study were: (1) to evaluate the mechanism of lactose assimilation in Nannochloropsis under different trophic modes and their intrinsic capacity for metabolising dairy waste, (2) to assess the effect of different waste pretreatment regimes (e.g. salinity intervention and sterilisation techniques) on physicochemical and biological characteristics of the waste and their ability to support microalgal growth, (3) to understand the critical role that phycospheric bacteria and their interaction with microalgal cells play in driving growth and bioremediation performance, (4) to assess the effect of cultivation on waste on the proximate and lipid composition of resulting microalgal biomass to determine end-user applications, and (5) using critical insights from Objective 1-4, to develop an innovative multiple-stage growth strategy which harnesses the power of probiotic bacteria in the wastewater and symbiotic bacteria in microalgal co-culture in order to maximise Nannochloropsis growth, bioremediation efficiency, and lipid productivity on the waste. Overall, this thesis demonstrated the potential of Nannochloropsis-based strategies for the effective bioremediation of lactose-rich dairy waste streams and co-generation of valuable products for circular bioeconomy development, such as ω-3 PUFAs, and β-galactosidase enzymes. Future studies can combine the two-step strategies developed in the thesis with other nutrient-feeding (e.g. batch, fed-batch, and continuous) and biological strategies (e.g. adaptive laboratory evolution) to further optimise growth and bioremediation performance. Lab-scale discoveries made in the study should also be substantiated at pilot scale and supported with predictive process modelling and robust trials using diverse waste streams generated throughout the dairy processing chain (e.g. CIP cleaning) in order to validate performance consistency and commercial scalability.2025-10-29 JG: Author's signature removed from PD
Eroding the Substance of Constitutional Liberal Democracy? Crimmigration and Democratic Decay in Sweden: A Qualitative Study
No full textThis thesis examines the current status of Sweden’s constitutional democracy. Utilising an interdisciplinary, empirical approach, it focuses particularly on the potential relationship between right-wing populism and non-Western immigration to Sweden. An analysis of qualitative data from 23 focus groups and 10 interviews develops both the meta-concept of democratic decay and crimmigration theory. Using these frames, this thesis subjects Sweden’s constitutional framework to a stress test and finds that the state is well-placed to withstand an authoritarian attack on the structures of constitutional liberal democracy. However, it is not as safeguarded when it comes to the erosion of the substance of constitutional liberal democracy. This thesis contends that the Sweden Democrats are a right-wing populist party that are ideologically illiberal. As populist actors, they are capitalising on and conflating the salient political issues of immigration and violent crime to push for punitive and exclusionary policy in these areas that is emblematic of democratic decay in Sweden. Data from focus groups and interviews provide nuance and counternarratives to those of the Sweden Democrats. They demonstrate that the Swedish public view right-wing populism and the Sweden Democrats as harmful. The data suggest that people in Sweden are not inherently opposed to immigration but instead are frustrated with how the state has approached integration. The data further suggest that people feel that the state is responding inadequately to violent crime. This study unfolds against the backdrop of a growing interest in understanding the increase in support for anti-immigrant populist parties across Europe and the Western world, as forced migration continues to be one of the most significant global challenges of our time. In conclusion, the empirical work suggests that Sweden, which sits among the most well-established democracies in the world, is hitting certain key points on a trajectory of democratic decay
Methodology Development for the Asymmetric Synthesis of Organophosphorus Compounds
No full textThis thesis describes the development of two methodologies for the preparation of P-stereogenic compounds.
The work described in Chapter 2 explores the application of anion-binding organocatalysis in the selective formation of ⍺-aminophosphinates bearing contiguous P- and C-stereogenic centres. It was proposed that an organocatalyst could induce enantioselective P-C bond formation and could also initiate an unprecedented stereoselective Arbuzov collapse to access these P-, C-stereogenic ⍺-aminophosphinates. Thus, a range of ⍺-aminophosphinates were prepared through a three-component catalytic asymmetric dearomatisation reaction between a prochiral phosphonite nucleophile, isoquinoline and Troc chloride in the presence of a thiourea organocatalyst. 13 examples of P-, C-stereogenic ⍺-aminophosphinates were prepared with yields up to 98%, enantioselectivities up to 96% ee and diastereoselectivities up to 30% de (although an outlier was prepared in 74% de). The mechanism of the reaction was studied to elucidate why the reaction proceeded with high enantioselectivity but poor diastereoselectivity. It was elucidated that the P-C bond formation step is selective, which results in good enantioselectivity at carbon. However, the phosphonium salt intermediate underwent a non-selective Arbuzov collapse to generate the phosphorus stereocentre with no selectivity, resulting in poor diastereoselectivity overall. A number of other organocatalytic systems for the preparation of compounds bearing P-stereocentres were also investigated but these did not give promising results. The work described in Chapter 3 involves the development of an efficient one-pot chiral auxiliary-based method for the preparation of enantioenriched P-stereogenic phosphines without the need to purify intermediates. Furthermore, P,P-dichlorophenylphosphine was utilised as the starting material, a P(III) compound which is commercially available. Thus, this reaction is redox-economic and avoids the need to use harsh reduction conditions to access tertiary phosphines. The P-stereogenic phosphines synthesised could be further derivatised (in one-pot) to phosphine boranes, phosphine oxides or phosphine sulfides. The synthesis of phosphine boranes was explored in detail using this methodology since they are typically challenging to access using current literature methods. A range of diarylalkyl phosphines (18 examples, up to 97% ee) and aryldialkyl phosphines (4 examples, up to 99% ee) were prepared, most isolated as the corresponding phosphine borane