143174 research outputs found
Sort by
Stochastic 1D search-and-capture as a G/M/c queueing model
No full textWe study the accumulation of resources within a target due to the interplay between continual delivery, driven by 1D stochastic search processes, and sequential consumption. The assumption of sequential consumption is key because it changes the commonly used G/M/∞ queue to a G/M/c queue. Combining the theory of G/M/c queues with the theory of first-passage times, we derive general conditions for the search process to ensure that the number of resources within the queue converges to a steady state and compute explicit expressions for the mean and variance of the number of resources within the
queue at steady state. We then compare the performance of the G/M/c queue with that of the G/M/∞ queue for an increasing number of servers. We extend the model to consider two competing targets and show that, under specific scenarios, an additional target is beneficial to the original target. Finally, we study the effects of multiple searchers. Using renewal theory, we numerically compute the inter-arrival time density for M searchers in the Laplace space, which allows us to exploit the explicit expressions for the steady-state statistics of the number of resources within G/M/1 and G/M/∞ queues, and compare their behaviour with different numbers of searchers. Overall, the G/M/c queue shows a tighter dependence on the configuration of the search process than the G/M/∞ queue does
A retrospective on the inception and development of structural power composites
No full textThis paper sets out the motivation for structural power composites: structural materials imbued with the capability to store and deliver electrical energy. The conception and development of structural supercapacitors at Imperial College London is described, with current devices now starting to approach the performance of conventional ‘monofunctional’ composite laminates and supercapacitors. Although these materials could offer tremendous lightweighting and energy storage benefits, there are considerable research challenges yet to be addressed. Melding of composite mechanics and electrochemistry disciplines leads to a daunting research landscape, so the effort has been partitioned into four themes: Constituent Development, Device Assembly and Characterisation, Multifunctional Modelling and Design and Scale-Up and Demonstration. This paper culminates by setting out, for each theme, where future research should focus to advance this exciting technology
Unravelling the transcriptome of the human tuberculosis lesion and its clinical implications
No full textThe tuberculosis (TB) lesion is a complex structure, contributing to the overall spectrum of TB. We characterise, using RNA sequencing, 44 fresh human pulmonary TB lesion samples from 13 TB individuals (drug-sensitive and multidrug-resistant TB) undergoing therapeutic surgery. We confirm clear separation between the TB lesion and adjacent non-lesional tissue, with the lesion samples consistently displaying increased inflammatory profile despite heterogeneity. Using weighted correlation network analysis, we identify 17 transcriptional modules associated with TB lesion and demonstrate a gradient of immune-related transcript abundance according to spatial organization of the lesion. Furthermore, we associate the modular transcriptional signature of the TB lesion with clinical surrogates of treatment efficacy and TB severity. We show that patients with worse disease present an overabundance of immune/inflammation-related modules and downregulated tissue repair and metabolism modules. Our findings provide evidence of a relationship between clinical parameters, treatment response and immune signatures at the infection site
Upadacitinib therapy during hospitalisation for moderate to severe ulcerative colitis in patients previously treated with infliximab: a UK tertiary care centre experience and literature review
No full textObjective: A plethora of patients diagnosed with ulcerative colitis (UC) will develop an episode of moderate to severe UC flare during the disease course, which may require hospitalisation. Rescue therapy after intravenous corticosteroid failure traditionally includes
infliximab (IFX) or ciclosporin; however, many patients admitted with a moderate to severe UC flare have prior IFX exposure, limiting its utility. Janus kinase inhibitors (JAKi), particularly upadacitinib, have shown promise in observational studies, but data remain limited.
Design/Methods: We conducted a retrospective, single-centre observational study of adults admitted with moderate to severe ulcerative colitis flare and previously exposed to IFX who received upadacitinib 45 mg daily as inpatients between January 2023 and June 2025. The primary outcomes were colectomy-free survival at week 12 and clinical remission at discharge, defined by a simple clinical colitis activity index (SCCAI) score ≤2. Secondary outcomes included clinical response, biochemical remission evaluated by faecal calprotectin (FCP) and C-reactive protein (CRP), endoscopic response, treatment persistence, adverse events, and colectomy rates up to 12 months.
Results: We recruited 24 patients. The mean age was 44.3 years (±14.5) with a median disease duration of 5 years (IQR 2.6–14.8). Colectomy-free survival at week 12 was 91.7%, with colectomies performed on days 8 and 16 post-initiation. Among the 12 patients with 12-month follow-up, no further colectomies occurred. SCCAI improved from a median value of 9 on admission to 2 at week 12. 9 out of 22 patients (41%) on upadacitinib achieved biochemical remission at week 12 with FCP below 150 ug/g and CRP below 5 mg/L. The ulcerative colitis endoscopic index of severity (UCEIS) was also evaluated with 4 patients achieving endoscopic response (66.7%) at week 12. No major adverse events occurred, including thromboembolic events and malignancies; minor adverse events included acne,
flu-like symptoms, and transient abnormal liver function tests.
Conclusion: Upadacitinib shows promise as a potential rescue option in moderate to severe UC among patients previously exposed to IFX. Randomised controlled trials are needed to confirm its efficacy and safety in both IFX-naïve and previously exposed populations
Power to hydrogen: a geospatial and economic analysis of green hydrogen for UK high-heat industry
No full textThis work analyses the economic feasibility of the core parts of a Power-to-Hydrogen system and provides a rigorous rational and methodology for sizing the facility by minimizing the Levelized cost of Hydrogen (LCOH) as a primary aim and reducing the carbon intensity of the hydrogen produced as a secondary aim. The study started with an in-depth study into the literature in the surrounding topic areas. The scaled hydrogen demand profile for high heat industry is synthetically produced allowing for a reasonably sized facility across regions of the UK. Monthly resolution of wind and solar data from each chosen location is fed into the optimization model, yielding a LCOH between 3.76 £/kg to 4.87 £/kg. The lowest LCOH is in regions with high quality wind and solar availability, and storage in the range of 40–96 h of average demand. All locations achieved the EU standard for green hydrogen and are tolerant to increases in grid Electricity prices. Further research would benefit from higher resolution renewable resource data, carbon pricing
Estimating future wildfire burnt area over Greece using the JULES-INFERNO model
No full textOur previous studies have shown that fire weather conditions in the Mediterranean and specifically over Greece are expected to become more severe with climate change, impling potential increases in burnt area. Here, we employ the Joint UK Land Environment Simulator (JULES) coupled with the INFERNO fire model driven by future climate projections from the UKESM1 model to investigate the repercussions of climate change and future vegetation changes on burnt area over Greece. We validate modelled burnt area against the satellite-derived GFED5 dataset, and find the model's performance to be good, especially for the more fire-prone parts of the country in the south Greece. For future simulations, we use future climate data following three Shared Socioeconomic Pathways (SSPs), consisting of an optimistic climate change scenario where fossil fuel emissions peak and decline beyond 2020 (SSP126), a middle-of-the-road scenario (SSP370), and a pessimistic scenario where emissions continue to rise throughout the century (SSP8.5). Our results show increased burnt area in the future compared to the present-day period in response to overall hotter and drier climatological conditions. We use an additional JULES-INFERNO simulation in which dynamic vegetation was activated, and find that it features smaller future burned area increases compared to our simulation with static present-day vegetation. For this dynamically changing vegetation simulation the greatest burnt area increases are found for southern Greece, due to higher future availability of flammable and heat-resistant needleleaf trees and the smallest decreases in agricultural areas of northern Greece due to a reduction in the aforementioned tree category
Large-scale plasma proteomics reveals bidirectional associations between sleep patterns and inflammatory bowel disease: a prospective cohort study
No full textBackground
Sleep disturbances are common in individuals with inflammatory bowel disease (IBD) and may worsen its progression. This study investigates the bidirectional association between unhealthy sleep and IBD and explores proteomic mechanisms underlying this relationship.
Methods
Data from 381,228 UK Biobank participants were analyzed to calculate adjusted odds ratios (ORs) for prevalent IBD and hazard ratios (HRs) for IBD incidence in relation to sleep patterns. A subset of 40,392 participants underwent plasma proteomic profiling, where differential expression analysis and weighted gene co-expression network analysis (WGCNA) identified key protein modules. A prognostic risk model for IBD was developed using least absolute shrinkage and selection operator (LASSO)-Cox regression.
Results
At baseline, 26.4% of participants exhibited unhealthy sleep, which was significantly associated with higher odds of prevalent IBD (OR = 1.250, 95% CI 1.165–1.340, p < 0.001) and an increased risk of developing IBD (HR = 1.237, 95% CI 1.136–1.348, p < 0.001). Proteomic profiling revealed 182 differentially expressed proteins common to both unhealthy sleep and IBD, with WGCNA identifying modules enriched in pathways related to cell activation, chemotaxis, and amino acid and organic acid metabolism. The proteomic risk model achieved an AUC of 0.81 for predicting 2-year IBD onset. Participants with both unhealthy sleep and high proteomic risk scores had a markedly increased risk of incident IBD (HR = 3.370, 95% CI 2.300–4.938, p < 0.001).
Conclusions
Unhealthy sleep and IBD are bidirectionally linked, with inflammatory and metabolic processes mediating this association. These findings highlight potential biomarkers and therapeutic targets, underscoring the importance of integrating sleep assessments into IBD management strategies
Fire and ice: investigating links between mantle dynamics and ice sheet stability
No full textOceanic and atmospheric warming threaten melting and collapse of Earth's ice sheets. Global mean sea level rise disrupts coastal ecosystems and communities by increasing destructive potential of coastal inundation events, and disturbing ocean circulation patterns. Therefore, robust projections of spatiotemporal patterns of sea level change are critically important. To construct them, improved understanding of solid Earth structure is required, due to physical couplings between mantle dynamics and ice sheet stability. Poor constraint on Earth's interior structure has obfuscated reliable estimation of future sea level change. Here, a Bayesian inverse framework for self-consistent conversion of seismic velocity into estimates of mantle thermomechanical structure is applied to Antarctica. Low viscosity anomalies are inferred in West Antarctica, such as western Marie Byrd Land, where viscosity is 10^(19.5 +/- 0.3) Pa s at 150 km depth. Thick lithosphere, high viscosity, and low geothermal heat flow is inferred in East Antarctica, consistent with cratonic lithosphere. By consideration of time-dependent viscosity perturbations (1 order of magnitude), seemingly disparate inferences of West Antarctic mantle viscosity derived from GPS data are reconciled. Variations in Antarctic geothermal heat flow from 20-130 mW/m^2 are inferred, based on a novel method incorporating crustal composition. Modifying the framework developed for estimating mantle structure, to incorporate use of xenolith-derived palaeogeotherm constraints, Australian lithospheric structure is mapped. It is demonstrated that 97% of mass mined from Australian base metal deposits is located within 200 km of the 195 km LAB depth contour. Finally, the impact of transient rheology on ice sheet stability is explored, and applied to a simple model of Antarctic glacial isostatic adjustment. For short melting timescales, significantly more near-field deformation is caused by a novel rheological model (exhibiting transient behaviour), as compared to a Maxwell model. When melting occurs over 25 years, a 52% increase in Earth surface displacement is observed
A deep learning-based approach to enhance accuracy and feasibility of long-term high-resolution manometry examinations
No full textBackground
High-resolution manometry (HRM) is the gold standard for diagnosing esophageal motility disorders. However, its short-term laboratory setting often fails to capture intermittent abnormalities. Long-term HRM (LTHRM, up to 24h) provides richer insights into swallowing behavior, but the resulting data volume is immense. Manual analysis by medical experts is laborious, time-consuming, and prone to errors, limiting its clinical feasibility.
Methods
We propose a deep learning-based approach for automatic analysis of LTHRM data. Our method detects both swallow events and secondary non-deglutitive motility disorders with high accuracy. Detected swallows are then clustered into distinct classes of similar events, creating a structured overview of motility patterns and their frequency. This reduces the analytical burden by allowing clinicians to focus on a small number of representative swallows rather than manually reviewing thousands of individual events. We evaluate our pipeline on 25 LTHRMs that were meticulously annotated, resulting in a dataset of more than 23,000 expert-labeled events.
Results
Our approach is able to detect more than 94% of all relevant events in LTHRM sequences, while the subsequent clustering is able to capture and group all relevant events into distinct swallow groups. To evaluate the overall approach, we conduct a user study with medical experts, demonstrating its effectiveness and positive clinical impact.
Conclusions
Our findings demonstrate that deep learning-based approaches to analyze LTHRM examinations are capable of providing a more reliable and efficient diagnostic process, ultimately making LTHRM assessments more feasible in clinical care