922017 research outputs found
Sort by
Denotational and Algebraic Semantics for Cyber-physical Systems
No full textThe cyber-physical system (CPS) is a dynamic system that contains both continuous and discrete behaviors. It has a wide range of applications in fields such as healthcare equipment, intelligent traffic control and environmental monitoring. However, the combination of continuous physical behaviorand discrete control behavior may complicate the design of systems further. It is of great necessity to give an explicit formal language and its semantics for CPS. In this paper, we elaborate the modeling language for CPS based on our previous work.This language supports shared variables to model the interaction between the physical and the cyber. Additionally, we give it denotational semantics and algebraic semantics, especially focus on the continuous behavior and its composition with the discrete behavior. Throughout this paper, we also present some examples to illustrate the feasibility of the language and its semantics intuitively.Index Terms—Cyber-physical system (CPS), Unifying Theories of Programming (UTP), Denotational semantics, Algebraic semantic
Does the design of the NHS Low Calorie Diet Programme have fidelity to the programme specification? A documentary review of service parameters and behaviour change content in a Type 2 Diabetes intervention
No full textBackgroundNHS England commissioned four independent service providers to pilot low-calorie diet programmes to drive weight loss, improve glycaemia and potentially achieve remission of Type 2 Diabetes across 10 localities. Intervention fidelity might contribute to programme success. Previous research has illustrated a drift in fidelity in the design and delivery of other national diabetes programmes.Aims: 1) To describe and compare the programme designs across the four service providers; 2) To assess the fidelity of programme designs to the NHS England service specification.MethodsThe NHS England service specification documents and each provider’s programme design documents were double coded for key intervention content using the Template for Intervention Description and Replication Framework and the Behaviour Change Technique (BCT) Taxonomy. ResultsThe four providers demonstrated fidelity to most but not all the service parameters stipulated in the NHS England service specification. Providers included between 74% and 87% of the 23 BCTs identified in the NHS specification. Twelve of these BCTs were included by all four providers; two BCTs were consistently absent. An additional seven to 24 BCTs were included across providers. ConclusionsA loss of fidelity for some service parameters and BCTs was identified across the provider’s designs; this may have important consequences for programme delivery and thus programme outcomes. Furthermore, there was a large degree of variation between providers in the presence and dosage of additional BCTs. How these findings relate to the fidelity of programme delivery and variation in programme outcomes and experiences across providers will be examined.KeywordsType 2 diabetes, low calorie diet, behaviour change, intervention design, intervention fidelity, total diet replacement, weight management <br/
Affective Human-Robot Interaction with Multimodal Explanations
No full textFacial expressions are one of the most practical and straightforward ways to communicate emotions. Facial Expression Recognition has been used in lots of fields such as human behaviour understanding and health monitoring. Deep learning models can achieve excellent performance in facial expression recognition tasks. As these deep neural networks have very complex nonlinear structures, when the model makes a prediction, it is not easy for human users to understand what is the basis for the model’s prediction. Specifically, we do not know which facial units contribute to the classification more or less. Developing affective computing models with more explainable and transparent feedback for human interactors is essential for a trustworthy human-robot interaction. Comparing to “white-box” approaches, “black-box” approaches using deep neural networks, which have advantages in terms of overall accuracy but lack reliability and explainability. In this work, we introduce a multimodal affective human-robot interaction framework, with visualbased and verbal-based explanation, by Layer Wise Relevance Propagation (LRP) and Local Intepretable Mode-Agnostic Explanation (LIME). The proposed framework has been tested on the KDEF dataset, and in human-robot interaction experiments with the Pepper robot. This experimental evaluation shows the benefits of linking deep learning emotion recognition systems with explainable strategies
THE AVOIDING LATE DIAGNOSIS OF OVARIAN CANCER (ALDO) PROJECT; A PILOT NATIONAL SURVEILLANCE PROGRAM FOR WOMEN WITH PATHOGENIC GERMLINE VARIANTS IN BRCA1 AND BRCA2.
No full textBackgrounds To establish ‘real-world’ performance and cost-effectiveness of Ovarian Cancer (OC) surveillance in women with pathogenic germline BRCA1/2 variants deferring risk-reducing salpingo-oophorectomy (RRSO). Methods875 female BRCA1/2-heterozygotes were recruited at 13 UK centres and via an online media campaign, with 767 undergoing at least one 4-monthly surveillance test with the Risk of Ovarian Cancer Algorithm (ROCA® Test). Surveillance performance was calculated with modelling of occult cancers detected at RRSO. Incremental cost-effectiveness ratio (ICER) was calculated using Markov population cohort simulation. Results 8 OCs occurred during 1277 women screen years: 2 occult OCs at RRSO (both stage 1a), and 6 screen-detected; 3 of 6 (50%) were ≤stage 3a and 5 of 6 (83%) were completely surgically cytoreduced. Modelled sensitivity, specificity, PPV and NPV for OC were 87.5% (95%CI, 47.3-99.7), 99.9% (99.9-100), 75% (34.9-96.8) and 99.9% (99.9-100) respectively. The predicted number of quality-adjusted life-years gained by surveillance was 0.179 with an ICER cost-saving of -£102,496/QALY. ConclusionOC surveillance for women deferring RRSO in a ‘real-world’ setting is feasible and demonstrates similar performance to research trials; it down-stages OC, leading to a high complete cytoreduction rate and is cost-saving in the UK setting. Whilst RRSO remains recommended management, ROCA-based surveillance may be considered for BRCA-heterozygotes deferring such surgery
Bias-Variance Decompositions for Margin Losses
No full textWe introduce a novel bias-variance decomposition for a range of strictly convex margin losses, including the logistic loss (minimized by the classic LogitBoost algorithm), as well as the squared margin loss and canonical boosting loss. Furthermore, we show that, for all strictly convex margin losses, the expected risk decomposes into the risk of a “central” model and a term quantifying variation in the functional margin with respect to variations in the training data. These decompositions provide a diagnostic tool for practitioners to understand model overfitting/underfitting, and have implications for additive ensemble models—for example, when our bias-variance decomposition holds, there is a corresponding “ambiguity” decomposition, which can be used to quantify model diversity.<br/
Prognostic impact of bronchoalveolar lavage galactomannan and Aspergillus culture results on survival in COVID-19 ICU patients: a post-hoc analysis from the European Confederation of Medical Mycology (ECMM) COVID-19-associated pulmonary aspergillosis (CAPA) Study
No full textCritically ill patients with coronavirus disease 2019 (COVID-19) may develop COVID-19-associated pulmonary aspergillosis (CAPA), that impact their chances of survival. Whether positive bronchoalveolar lavage fluid (BALF) mycological tests can be used as a survival proxy remains unknown. We conducted a post-hoc analysis of a previous multicenter, multinational observational study with the aim of assessing the differential prognostic impact of BALF mycological tests, namely positive (≥ 1.0 optical density index) BALF galactomannan (GM) and positive BALF Aspergillus culture alone or in combination in critically ill patients with COVID-19. Of the 592 patients critically ill patients with COVID-19 enrolled in the main study, 218 were included in this post-hoc analysis as they had both test results available. CAPA was diagnosed in 56/218 patients (26%). Most cases were probable CAPA (51/56, 91%) and fewer were proven CAPA (5/56, 9%). In the final multivariable model adjusted for between-center heterogeneity, an independent association with 90-day mortality was observed for the combination of positive BALF GM and positive BALF Aspergillus culture in comparison with both tests negative (hazard ratio 2.53, 95% CI 1.28-5.02, p = 0.008). The other independent predictors of 90-day mortality were increasing age and active malignant disease. In conclusion, the combination of positive BALF GM and positive BALF Aspergillus culture was associated with increased 90-day mortality in critically ill patients with COVID-19. Additional study is needed to explore the possible prognostic value of other BALF markers.Key words: CAPA; GM; biomarker; galactomannan; Aspergillus; COVID-19; BALF.<br/
Impact and control of fouling in radioactive environments.
No full textFouling and scaling of equipment in the nuclear industry is a significant and challenging problem that effects multiple areas across the entire nuclear fuel cycle. Consequences such as the blockage of fluid flow, accumulation of radionuclides, reduction of heat-transfer energy and enhancement of corrosion, all can have detrimental effects on safety and performance as well as incurring substantial damage and maintenance costs amounting to billions of pounds a year. This review focuses on pipelines and understanding the mechanisms of formation and radionuclide incorporation of inorganic and biological fouling, and microbially influenced corrosion (MIC) mechanisms, as well as exploring prevalent examples in the nuclear industry and parallels in the oil and gas industries. The review will also cover advancements in fouling and scale mitigation and treatment strategies, which are imperative to reduce economic loses and avoid safety hazards in nuclear as well as many other industries
Adrenergic prolongation of action potential duration in rainbow trout myocardium via inhibition of the delayed rectifier potassium current, IKr.
No full textCatecholamines mediate the ‘fight or flight’ response in a wide variety of vertebrates.The endogenous catecholamine adrenaline increases heart rate and contractilestrength to raise cardiac output. The increase in contractile force is driven in large partby an increase in myocyte Ca 2+ influx on the L-type Ca current (I CaL ) during thecardiac action potential (AP). Here, we report a K + - based mechanism that prolongsAP duration (APD) in fish hearts following adrenergic stimulation. We show thatadrenergic stimulation inhibits the delayed rectifier K + current (I Kr ) in rainbow trout( Oncorhynchus mykiss ) cardiomyocytes. This slows repolarization and prolongsAPD which may contribute to positive inotropy following adrenergic stimulation in fishhearts. The endogenous ligand, adrenaline (10 -6 M), which activates both α- and βARs reduced maximal I Kr tail current to 61.4±3.9% of control in atrial and ventricularmyocytes resulting in an APD prolongation of ~ 20% at both 50 and 90%repolarization. This effect was reproduced by the α-specific adrenergic agonist,phenylephrine (10 -6 M), but not the β-specific adrenergic agonist isoproterenol (10 -6 M). Adrenaline (10 -6 M) in the presence of β 1 and β 2 -blockers (10 -6 Matenolol and 10 -6 M ICI-118551, respectively) also inhibited I Kr . Thus, I Krsuppression following adrenergic stimulation leads to APD prolongation in the rainbowtrout heart. This is the first time this mechanism has been identified in fish and may actin unison with the well-known enhancement of I CaL following adrenergic stimulationto prolong APD and increase cardiac inotropy
Fairness-Aware Data Integration
No full textMachine learning can be applied in applications that take decisions that impact people’s lives. Such techniques have the potential to make decision making more objective, but there also is a risk that the decisions can discriminate against certain groups as a result of bias in the underlying data. Reducing bias, or promoting fairness, has been a focus of significant investigation in machine learning, for example based on pre-processing the training data, changing the learning algorithm, or post-processing the results of the learning. However, prior to these activities, data integration discovers and integrates the data that is used for training, and data integration processes have the potential to produce data that leads to biased conclusions. In this paper, we propose an approach that generates schema mappings in ways that take into account: (i) properties that are intrinsic to mapping results that may give rise to bias in analyses; and (ii) bias observed in classifiers trained on the results of different sets of mappings. The approach explores a space of different ways of integrating the data, using a tabu search algorithm, guided by bias-aware objective functions that represent different types of bias.The resulting approach is evaluated using Adult Census and German Credit datasets, to explore the extent to which and the circumstances in which the approach can increase the fairness of the results of the data integration process