143174 research outputs found
Sort by
Spectroscopy motivated probabilistic latent variable models
No full textLatent variable models are a powerful approach for capturing underlying structure in high dimensional data. One such setting for this high dimensional data is spectroscopy, which has an inherent structure due to the physical process generating the data. Spectroscopy has a wide range of applications, each of which results in specific sets of challenges, this thesis is motivated by the challenge presented by monitoring the manufacture of Pharmaceuticals.
This thesis is interested in how to use a probabilistic approach to latent variable models to the challenge through two main contributions focused on uncertainty quantification and encoding flexible, physically relevant priors. We first tackle uncertainty estimation in Partial Least Squares (PLS) regression, a widely used technique in spectroscopy. While existing methods rely on problematic linear approximations, we develop a bootstrap-based approach that naturally captures non-linear parameter interactions. We demonstrate its effectiveness across multiple pharmaceutical case studies, showing particular strength in Design Space identification where accurate uncertainty estimates are crucial.
The second part of the thesis introduces a novel probabilistic framework, the Weighted-Sum Gaussian Process Latent Variable Model (WS-GPLVM), which combines physical understanding from Beer-Lambert's law with flexible Gaussian Process models.
This model allows variations in conditions which cause changes in the pure component spectra to be found via including additional latent variable. We develop this model over the course of two chapters of this thesis, resulting in a novel, flexible method with demonstrable real world effectiveness.Open Acces
Biodiversity conservation requires integration of species-centric and process-based strategies
No full textConservation science and policy are geared primarily toward the preservation of species and habitats, with priority often given to the rarest, most vulnerable or most charismatic forms. This pattern-based approach has broad appeal and offers a pragmatic short-cut for targeting conservation action. However, the long-term efficacy of species and landscape conservation programs remains highly uncertain, amid growing evidence that sustainable conservation action requires an increased emphasis on preserving ecological and evolutionary processes. This reframing of conservation goals was first proposed 50 y ago, but the concept has struggled to gain traction, particularly in terms of translation into policy. Nonetheless, recent events have shifted the narrative, with multiple interlinked global challenges—including biological invasions, food security, disease, and climate change—putting ecological processes firmly back on the agenda. Concurrently, conservation finance is changing rapidly, driven in part by the 2022 Kunming-Montreal Global Biodiversity Framework, which prioritized actions to enhance and restore ecosystem stability, connectivity, and resilience. These ecosystem properties are fundamentally process-driven and appear to create an operational gulf between current conservation practice and the targets of international agreements. We describe how new approaches can be used to close this gap by redirecting conservation attention toward processes at the heart of ecosystem function, including adaptation, gene flow, dispersal, and trophic interactions. Wider adoption of these approaches is urgently needed to forge a deeper connection between conservation practice and policy targets, thereby ensuring that ongoing investment in biodiversity conservation goes beyond damage limitation and instead leaves a lasting legacy of resilient ecosystems
Behaviour, numerical modelling and design of fixed-ended unequal-leg angle section steel columns
No full textThe mechanical behaviour and design of fixed-ended unequal-leg angle section steel members subjected to axial compression are studied herein. The mechanical response of unequal-leg angle section columns is first described, with a particular emphasis on the relationship between torsional and local buckling, alongside the important transition from equal-leg to unequal-leg angle behaviour. Existing experimental data on unequal-leg angle section steel columns are then employed to validate numerical models developed within the commercial finite element package ABAQUS. A comprehensive parametric study is subsequently conducted that encompasses a broad spectrum of geometric configurations and global slenderness values. The mechanical behaviour and ultimate resistance of fixed-ended unequal-leg angle section columns are shown to be dependent on not only the global slenderness, but also on the ratio of the elastic torsional-flexural to minor-axis flexural buckling loads. The existing experimental data alongside the numerical parametric study results are employed to evaluate the resistance predictions given by the current Eurocode 3 design provisions, revealing an excessive level of conservatism. Finally, a new design approach for fixed-ended unequal-leg angle section steel columns, suitable for incorporation into future revisions of Eurocode 3, is proposed that significantly improves the accuracy and consistency of the resistance predictions. A reliability analysis of the proposed design approach is conducted in accordance with the procedure within EN 1990, resulting in a recommended partial safety factor γ_M₁=1.0
Myeloid cell networks govern re-establishment of original immune landscapes in recurrent ovarian cancer
No full textImmunotherapy has shown limited success in recurrent ovarian cancer (OC), with prognostic insights largely derived from treatment-naive tumors. We analyzed 697 tumor samples (566 primary and 131 recurrent) from 595 OC patients across five independent cohorts, capturing tumor-infiltrating lymphocytes (TILs) heterogeneity and identifying four immune phenotypes linked to prognosis and TIL:myeloid networks driving malignant progression. We found that in preclinical mouse models, mirroring inflamed human OCs, the recurrent Brca1mut tumors maintained activated TILs:dendritic cells (DCs) niches but evaded immune control through upregulation of COX/PGE2 signaling. Conversely, recurrent Brca1wt tumors displayed loss of TILs:DCs niches and accumulated immunosuppressive tumor microenvironment (TME) networks featuring Trem2/ApoEhigh tumor associated macrophages (TAMs) and Nduf4l2high/Galectin3high malignant states. Recurrent tumors recapitulate the immunogenic landscapes of original cancers. Our findings reveal BRCA-dependent TIL:myeloid crosstalk as key to persistent immunogenicity in recurrent OC and propose new targets to enhance chemotherapy efficacy
Experimental investigation of pile-supported Oscillating Water Column devices
No full textThe present study investigates the impact of key geometric parameters in the design of an Oscillating Water Column (OWC) integrated into a pile-supported breakwater. This is achieved through an extensive experimental campaign and a systematic investigation of key device parameters. Specifically, the present study examines the pneumatic efficiency of the OWC, the geometric characteristics of both the OWC and the breakwater, as well as the position of the OWC within the breakwater. The effect of these device characteristics on the performance of the OWC is assessed by considering monochromatic waves of varying steepness and effective water depths. The performance of the OWC is evaluated in terms of its wave transmission and reflection coefficients, as well as its energy generation efficiency. In turn, these are quantified using arrays of collocated sensors and high-speed imaging. Taken together, the parametric study provides physical insights into the effect of key device parameters on the efficiency of the OWC. Once optimal configurations are employed, the power output of the device is shown to increase by up to 164%, while wave transmission is reduced by 55%, compared to the initial design configuration. These results offer a valuable perspective for the development of more efficient wave energy converters
The pre-operative management of fracture blisters: a systematic review
No full textPurpose:
The pre-operative management of fracture blisters is an area of uncertainty within trauma and orthopaedic surgeries. Management strategies vary significantly between and within orthopaedic departments across the United Kingdom. The purpose of this systematic review was to comprehensively appraise and synthesize the existing literature pertaining to this topic, highlighting current practices and areas for ongoing research.
Methods:
Extensive electronic literature searches were performed on PubMed/MEDLINE (January 1946–May 2024), Embase (January 1974–May 2024) and Cochrane library (January 1933–May 2024) databases. The search terms were as follows: (fracture blister OR bone blister*) AND (dress* OR drain* OR aspirat* OR deroof* OR manage*). These keywords were searched in the subject headings, in title and in abstract.
Results:
The results of the search methodology revealed five articles, which represented the best evidence to the clinical question. These papers reported on rates of wound healing and post-operative infection, time to surgical readiness and treatment costs, following varying treatment modalities in 1162 patients. The authors, publication dates, countries, patient groups, study outcomes and results of these papers are tabulated in Supplementary Table 1.
Conclusion:
Fracture blisters pose a significant challenge in clinical practice, leading to delays in surgery, suboptimal surgical approaches and complications in wound healing post-operatively. Currently, there is no consensus describing the optimal management of these blisters. This review challenges the conventional belief that fracture blisters are sterile, highlighting that the application of topical agents to the deroofed blister bed may expedite surgical readiness
A small serving of mash: (Quantum) algorithms for SPDH-Sign with small parameters
No full textWe find an efficient method to solve the semidirect discrete logarithm problem (SDLP) over finite nonabelian groups of order p3 and exponent p2 for certain exponentially large parameters. This implies an attack on SPDH-Sign, a signature scheme based on the SDLP, for such parameters. In particular, SDLP instances over such groups are parameterised by an n < (p − 1)p6: we develop a method to solve instances when n ≤ poly(log p) · p. Letting λ be the security parameter of SPDH-Sign, which is taken p = exp λ, we find we may solve instances of SDLP corresponding to SPDH-Sign instances with exponentially large p. However, for n ≈ p2 and larger, our method no longer completely solves the SDLP instances. We also study the linear hidden shift problem for a group action corresponding to SDLP, and take a step towards proving the quantum polynomial time equivalence of SDLP and the semidirect computational Diffie-Hellman problem
Beyond humanoid prosthetic hands: modular terminal devices that improve user performance
No full textDespite decades of research and development, myoelectric prosthetic hands lack functionality and are often rejected by users. This lack in functionality can be partially attributed to the widely accepted anthropomorphic design ideology in the field; attempting to replicate human hand form and function despite severe limitations in control and sensing technology. Instead, prosthetic hands can be tailored to perform specific tasks without increasing complexity by shedding the constraints of anthropomorphism. In this paper, we develop and evaluate four open-source modular non-humanoid devices to perform the motion required to replicate human flicking motion and to twist a screwdriver, and the functionality required to pick and place flat objects and to cut paper. Experimental results from these devices demonstrate that, versus a humanoid prosthesis, non-humanoid prosthesis design dramatically improves task performance, reduces user compensatory movement, and reduces task load. Case studies with two end users demonstrate the translational benefits of this research. We found that special attention should be paid to monitoring end-user task load to ensure positive rehabilitation outcomes
Inference of a three-gene network underpinning epidermal stem cell development in Caenorhabditis elegans
No full textGene regulatory networks are crucial in cellular decision-making, making the inference of their architecture essential for understanding organismal development. The gene network of Caenorhabditis elegans epidermal stem cells, known as seam cells, remains undefined. Here, we integrate experimental data, mathematical modeling, and statistical inference to investigate this network, focusing on three core transcription factors (TFs), namely ELT-1, EGL-18, and CEH-16. We use single-molecule FISH to quantify TF mRNA levels in single seam cells of wild-type and mutant backgrounds across four early larval stages. Using Modular Response Analysis, we predict TF interactions and uncover a repressive interaction between CEH-16 and egl-18 consistent across time points. We validate its significance at the L1 stage with ordinary differential equations and Bayesian modeling, making testable predictions for a double mutant. Our findings reveal TF regulatory relationships in seam cells and demonstrate a flexible mathematical framework for inferring gene regulatory networks from gene expression data
Seismic fault identification of deep fault-karst carbonate reservoir using transfer learning
No full textSeismic fault identification is a critical step in structural interpretation, reservoir characterization, and well-drilling planning. However, fault identification in deep fault-karst carbonate formations is particularly challenging due to their deep burial depth and the complex effects of dissolution. Traditional manual interpretation methods are often labor intensive and prone to high uncertainty due to their subjective nature. To address these limitations, this study proposes a transfer learningebased strategy for fault identification in deep fault-karst carbonate formations. The proposed methodology began with the generation of a large volume of synthetic seismic samples based on statistical fault distribution patterns observed in the study area. These synthetic samples were used to pretrain an improved U-Net network architecture, enhanced with an attention mechanism, to create a robust pretrained model. Subsequently, real-world fault labels were manually annotated based on verified fault interpretations and integrated into the training dataset. This combination of synthetic and real-world data was used to fine-tune the pretrained model, significantly improving its fault interpretation accuracy. The experimental results demonstrate that the integration of synthetic and realworld samples effectively enhances the quality of the training dataset. Furthermore, the proposed transfer learning strategy significantly improves fault recognition accuracy. By replacing the traditional weighted cross-entropy loss function with the Dice loss function, the model successfully addresses the issue of extreme class imbalance between positive and negative samples. Practical applications confirm that the proposed transfer learning strategy can accurately identify fault structures in deep fault-karst carbonate formations, providing a novel and effective technical approach for fault interpretation in such complex geological settings