143174 research outputs found
Sort by
Identification of feasible regions using R-functions
No full textThe primary objective of feasibility analysis is to identify and define the feasibility region, which represents the range of operational conditions (e.g., variations in process parameters) that ensure safe, reliable, and feasible process performance. This work introduces a novel feasibility analysis method that requires only that model constraints (e.g., defining product Critical Quality Attributes or process Key Performance Indicators) be explicitly provided or approximated by a closed-form function, such as a multivariate polynomial model. The method is based on V.L. Rvachev's R-functions, enabling an explicit analytical representation of the feasibility region without relying on complex optimization-based approaches. R-functions offer a framework for describing intricate geometric shapes and performing operations on them using implicit functions and inequality constraints. The theory of R-functions facilitates the identification of feasibility regions through algebraic manipulation, making it a more practical alternative to traditional optimization-based methods. The effectiveness of the proposed approach is demonstrated using a suite of well-known test cases from the literature
Direct and indirect impacts of the COVID-19 pandemic on life expectancy and person-years of life lost with and without disability: a systematic analysis for 18 European countries, 2020-2022
No full textBackground
The direct and indirect impacts of the COVID-19 pandemic on life expectancy (LE) and years of life lost with and without disability remain unclear. Accounting for pre-pandemic trends in morbidity and mortality, we assessed these impacts in 18 European countries, for the years 2020–2022.
Methods and Findings
We used multi-state Markov modeling based on several data sources to track transitions of the population aged 35 or older between eight health states from disease-free, combinations of cardiovascular disease, cognitive impairment, dementia, and disability, through to death. We quantified separately numbers and rates of deaths attributable to COVID-19 from those related to mortality from other causes during 2020–2022, and estimated the proportion of loss of life expectancy and years of life with and without disability that could have been avoided if the pandemic had not occurred. Estimates were disaggregated by COVID-19 versus non-COVID causes of deaths, calendar year, age, sex, disability status, and country. We generated the 95% uncertainty intervals (UIs) using Monte Carlo simulations with 500 iterations. Among the 289 million adult population in the 18 countries, person-years of life lost (PYLL) in millions were 4.7 (95% UI 3.4–6.0) in 2020, 7.1 (95% UI 6.6–7.9) in 2021, and 5.0 (95% UI 4.1–6.2) in 2022, totaling 16.8 (95% UI 12.0–21.8) million. PYLL per capita varied considerably between the 18 countries ranging between 20 and 109 per 1,000 population. About 60% of the total PYLL occurred among persons aged over 80, and 30% in those aged 65–80. If the pandemic were avoided, over half (9.8 million (95% UI 4.7–15.1)) of the 16.8 million PYLL were estimated to have been lived without disability. Of the total PYLL, 11.6–13.2 million were due to registered COVID-19 deaths and 3.6–5.3 million due to non-COVID mortality. Despite a decrease in PYLL attributable to COVID-19 after 2021, PYLL associated with other causes of death continued to increase from 2020 to 2022 in most countries. Lower income countries had higher PYLL per capita as well as a greater proportion of disability-free PYLL during 2020–2022. Similar patterns were observed for life expectancy. In 2021, LE at age 35 (LE-35) declined by up to 2.8 (95% UI 2.3–3.3) years, with over two-thirds being disability-free. With the exception of Sweden, LE-35 in the studied countries did not recover to 2019 levels by 2022.
Conclusions
The considerable loss of life without disability and the rise in premature mortality not directly linked to COVID-19 deaths during 2020–2022 suggest a potential broader, longer-term and partially indirect impact of the pandemic, possibly resulting from disruptions in healthcare delivery and services for non-COVID conditions and unintended consequences of COVID-19 containment measures. These findings highlight a need for better pandemic preparedness in Europe, ideally, as part of a more comprehensive global public health agenda
Spatially diffuse cAMP signalling with oppositely biased GLP-1 receptor agonists in β-cells despite differences in receptor localisation
No full textInternalisation of G protein-coupled receptors (GPCRs) can contribute to altered cellular responses by directing signalling from non-canonical locations, such as endosomes. If signalling processes are locally constrained, active receptors in different subcellular locations could produce different downstream effects. This phenomenon may be relevant to the optimal targeting of the glucagon-like peptide-1 receptor (GLP-1R), a type 2 diabetes and obesity target GPCR for which several ligands with varying internalisation tendency have been discovered. To investigate, we compared the signalling localisation effects of two prototypical GLP-1RAs with opposite signal bias and effects on GLP-1R trafficking: exendin-asp3 (ExD3), a full agonist that drives rapid internalisation, and exendin-phe1 (ExF1), which shows much slower internalisation. After using bioorthogonal labelling and fluorescent agonist conjugates to verify the divergent trafficking patterns of ExF1 and ExD3 in β-cell lines and primary pancreatic islets, we used live cell biosensors to monitor signalling at different subcellular locations. This revealed that cAMP/PKA/ERK signalling in β-cells is in fact distributed widely across the cell over short- (10 pM) concentrations, with no major differences in signal localisation that could be linked to internalised versus cell surface-bound GLP-1R. Moreover, washout experiments highlighted that, whilst fast-internalising ExD3 shows much greater accumulation and binding to GLP-1R in endosomes than slow-internalising ExF1, it is a rather inefficient driver of both cAMP production in β-cells and insulin secretion from perfused rat pancreata. These data provide a greater understanding of the cellular effects of biased GLP-1R agonism
Reconstructing Eocene Antarctic river drainage from provenance analysis of Amundsen Sea embayment sediments
No full textSedimentary records can illuminate relationships between the climate, topography and glaciation of West Antarctica by revealing its Cenozoic topographic and paleoenvironmental history. Eocene fluvial drainage patterns have previously been inferred using geochemical provenance data from a ~44-34 Ma deltaic sandstone recovered from the Amundsen Sea Embayment. One interpretation holds that a low relief, low-lying West Antarctic landscape supported a >1500 km transcontinental river system. Alternatively, higher relief topography in central West Antarctica formed a drainage divide between the Ross and Amundsen seas. Here, zircon U-Pb data from Amundsen Sea Embayment sediments are examined alongside known regional bedrock provenance signatures, suggesting that all provenance indicators in the Eocene sandstone derive from West Antarctic rocks. This implies that a local river system flowed off a West Antarctic drainage divide, helping constrain the mid-late Eocene evolution of West Antarctic topography with implications for rifting history and the characteristics of sediments infilling interior basins
Single cell profiling framework reveals metabolic subpopulations as drivers of bioproduction heterogeneity
No full textHeterogeneity within clonal cell populations remains a critical bottleneck within bioprocess engineering, notably by undermining bioproduction yields. Efforts to mitigate its impact have, however, been hampered by technological difficulties quantifying metabolism at the single-cell level. Here, we propose a framework based on single-cell biosensor analysis that enables robust characterisation of cell's metabolic states, leveraging it to detect and isolate isogeneic heterogeneity in response to environmental perturbations and within microbial cell factories. We identify acute and gradual glucose depletion to induce differentiation of metabolically distinct subpopulations and reveal these subpopulations to exhibit differential production capabilities, with lower intracellular pH subpopulations exhibiting enhanced product accumulation within violacein-producing strains but reduced yields within lycopene-producing strains. Lastly, we highlight galactose cultivation as a method to modulate subpopulation dynamics towards higher-producing lycopene phenotypes. Altogether, our research provides insights into subpopulation differentiation and establishes promising avenues for the engineering of more robust and higher-producing strains
Exploring the predictors and barriers to accepting smoking cessation support within a targeted lung health check setting
No full textBackground: The Quit Smoking Lung Health Intervention Trials (QuLIT-1 and -2) and other studies show that providing immediate smoking cessation within lung cancer screening services substantially improves quit rates. However, in the QuLIT2 trial only around half of
those offered smoking cessation support actually accepted it. Understanding what underpins
this and how to facilitate higher smoking cessation rates would enhance the health impact of the Targeted Lung Health Check programme.
Method: We compared characteristics of participants in the intervention arm of the QuLIT-2 study who accepted or declined the offer of smoking cessation support and conducted thematic analysis of interviews with 15 smokers who had declined it.
Results: Of 152 randomised to smoking cessation support (61.3±4.8 years, 42% female), 80 declined the offer and 15 dropped out after the initial session leaving 57 “accepters”. Accepters were more likely to be female [53% vs 40% AOR: 3.30, 95%CI 1.47-7.48], younger
[AOR: 0.90(0.80-0.98)] and were more likely to live in areas of medium or low deprivation, [AOR: 5.30(1.86-22.85)]. Thematic analysis of the interviews revealed four main barriers to acceptance: concerns about mental health, beliefs about quitting smoking and about the
effectiveness of interventions, and negative past experiences of smoking cessation support.
Discussion: Cessation services embedded in lung screening clinics need to anticipate barriers such as mental health concerns, past experiences and personal beliefs. Efforts should be made to design and offer equitable services that meet the needs of this population.
Trial registration: This study is registered online: ISRCTN12455871.
What is already known on this topic- Intensive smoking cessation support embedded within lung cancer screening services significantly increases 3 and 12 month quit rates among this
high-risk population.
What this study adds- Despite the success of cessation embedded into screening, a high proportion of smokers declined or drop out of smoking cessation support. This study highlights the demographic predictors of people who are more likely to accept the support, alongside shedding light on the barriers and reasons why these high-risk smokers refused support.
How this study might affect research, practice or policy – This study provides suggestions and implications for lung cancer screening services that provide smoking cessation support, particularly how these services should anticipate barriers and provide holistic behavioural
theory with multiple options in terms of location and pharmacotherapy
Deep learning predicts real-world electric vehicle direct current charging profiles and durations
No full textAccurate prediction of electric vehicle charging profiles and durations is critical for adoption and optimising infrastructure. Direct current fast charging presents complex behaviours shaped by many factors. This work introduces a deep learning framework trained on 909,135 real-world sessions, capable of predicting charging profiles and durations from minimal input with uncertainty estimates. The model initiates predictions from a single point on the power and state-of-charge profile and incrementally refines them as new observations arrive, enabling real-time updates. The model generalises across vehicle types and charging scenarios. It achieves 90% accuracy in predicting charging duration from a single point, and 95% accuracy with an absolute error under one minute using six points within five minutes. This work shows that using readily available input data at charge time enables accurate prediction of charging behaviour and offers a practical, scalable solution for deployment, energy planning, and infrastructure reliability
A machine learning‐driven pore‐scale network model coupling reaction kinetics and interparticle transport for catalytic process design
No full textA pore-scale dual-network model is presented with kinetics (DNMK), enhanced by machine learning (ML), for efficient multiscale modeling of reaction-transport coupled catalytic processes in porous systems. In such systems, apparent catalytic performance arises from the intricate interplay between intrinsic microkinetics and inter-particle transport phenomena. By explicitly resolving these coupled effects, DNMK provides mechanistic insight into how spatial particle arrangements and transport limitations govern overall reactor performance. A key innovation of this work is the integration of ML-based surrogates to accelerate the microkinetic module, effectively bridging the large spatial and temporal scale mismatches between transport and catalytic reactions. This hybrid approach achieves up to a 750-fold computational speed-up while preserving full physical and chemical fidelity. The framework is demonstrated for sorption-enhanced CO2 hydrogenation to methanol, where DNMK identifies optimal catalyst-sorbent configurations that maximize apparent activity and reactor-scale performance. More broadly, DNMK establishes a high-resolution, ML-driven platform for digital catalytic experimentation, enabling predictive, in silico optimization of catalyst scaling, utilization, and process intensification. By allowing rapid, physically consistent evaluation of complex catalytic systems, DNMK reduces reliance on costly experimental trials and opens new pathways for data-driven reactor and process design across diverse chemical engineering applications
Deep learning from imperfectly labeled malware data
No full textDeep learning approaches have achieved remarkable performance in malware classification and detection. However, their success relies on the availability of large, accurately labeled datasets: a critical yet challenging requirement in the malware domain. In practice, most malware datasets are automatically labeled using outputs from antivirus engines, a process that often introduces significant label noise. Such imperfections can severely degrade the performance and generalizability of deep learning models. To address this challenge, we introduce SLB, a framework designed to robustly train deep learning–based malware systems while simultaneously refining dataset labels. SLB begins by partitioning the dataset into two subsets: a clean set containing samples with reliable labels, and a noisy set with samples that may be mislabeled, to which pseudo labels are assigned. As training progresses, SLB continuously monitors the model's predictions to dynamically update both sets. Specifically, samples in the noisy set that consistently receive predictions aligning with their (observed or pseudo) labels are promoted to the clean set, whereas samples in the clean set that exhibit unstable predictions are reclassified as noisy. This iterative process not only enhances model performance but also progressively corrects labeling errors. We evaluated SLB on multiple security datasets with both synthetic and real-world label noise across various deep learning architectures and ML algorithms. Experimental results show that SLB significantly improves malware detection performance and reduces overall noise. For example, on the Android binary dataset with 25% injected label noise, SLB reduced the noise to below 1.5% while increasing the macro F1 score from 74.51% to 96.03% and the accuracy score from 87.66% to 98.68%
Probabilistically robust counterfactual explanations under model changes
No full textWe study the problem of generating robust counterfactual explanations for deep learning models subject to model changes. We focus on plausible model changes altering model parameters and propose a novel framework to reason about the robustness property in this setting. To motivate our solution, we begin by showing for the first time that computing the robustness of counterfactuals with respect to model changes is NP-hard. As this (practically) rules out the existence of scalable algorithms for exactly computing robustness, we propose a novel probabilistic approach which is able to provide tight estimates of robustness with strong guarantees while preserving scalability. Remarkably, and differently from existing solutions targeting plausible model changes, our approach does not impose requirements on the network to be analysed, thus enabling robustness analysis on a wider range of architectures, including state-of-the-art tabular transformers. A thorough experimental analysis on four binary classification datasets reveals that our method improves the state of the art in generating robust explanations, outperforming existing methods