68 research outputs found
Exploring European Consensus About the Remaining Treatment Challenges and Subsequent Opportunities to Improve the Management of Invasive Fungal Infection (IFI) in the Intensive Care Unit
No full textBackground: The global prevalence of invasive fungal infections (IFI) is increasing, particularly within Intensive Care Units (ICU), where Candida spp. and Aspergillus spp. represent the most important pathogens. Diagnosis and management of IFIs becomes progressively challenging, with increasing antifungal resistance and the emergence of rare fungal species. Through a consensus survey focused on assessing current views on how IFI should be managed, the aim of this project was to identify challenges around diagnosing and managing IFIs in the ICU. The current status in different countries and perceived challenges to date amongst a multidisciplinary cohort of healthcare professionals involved in the care of IFI in the ICU was assessed. Methods: Using a modified Delphi approach, an expert panel developed 44 Likert-scale statements across 6 key domains concerning patient screening and minimal standards for diagnosis of IFIs in ICU; initiation and termination of antifungal treatments and how to minimise their side effects and insights for future research on this topic. These were used to develop an online survey which was distributed on a convenience sampling basis utilising the subscriber list held by an independent provider (M3 Global). This survey was distributed to intensivists, infectious disease specialists, microbiologists and antimicrobial/ICU pharmacists within the UK, Germany, Spain, France and Italy. The threshold for consensus was set at 75%. Results: A total of 335 responses were received during the five-month collection period. From these, 29/44 (66%) statements attained very high agreement (≥ 90%), 11/44 (25%) high agreement (< 90% and ≥ 75%), and 4/44 (9%) did not meet threshold for consensus (< 75%). Conclusion: The results outline the need for physicians to be aware of the local incidence of IFI and the associated rate of azole resistance in their ICUs. Where high clinical suspicion exists, treatment should start immediately and prior to receiving the results from any diagnostic test. Beta-D-glucan testing should be available to all ICU centres, with results available within 48 h to inform the cessation of empirical antifungal therapy. These consensus statements and proposed measures may guide future areas for further research to optimise the management of IFIs in the ICU
State space modelling of extreme values with particle filters
No full textState space models are a flexible class of Bayesian model that can be used to smoothly capture non-stationarity. Observations are assumed independent given a latent state process so that their distribution can change gradually over time. Sequential Monte Carlo methods known as particle filters provide an approach to inference for such models whereby observations are added to the fit sequentially. Though originally developed for on-line inference, particle filters, along with related particle smoothers, often provide the best approach for off-line inference. This thesis develops new results for particle filtering and in particular develops a new particle smoother that has a computational complexity that is linear in the number of Monte Carlo samples. This compares favourably with the quadratic complexity of most of its competitors resulting in greater accuracy within a given time frame. The statistical analysis of extremes is important in many fields where the largest or smallest values have the biggest effect. Accurate assessments of the likelihood of extreme events are crucial to judging how severe they could be. While the extreme values of a stationary time series are well understood, datasets of extremes often contain varying degrees of non-stationarity. How best to extend standard extreme value models to account for non-stationary series is a topic of ongoing research. The thesis develops inference methods for extreme values of univariate and multivariate non-stationary processes using state space models fitted using particle methods. Though this approach has been considered previously in the univariate case, we identify problems with the existing method and provide solutions and extensions to it. The application of the methodology is illustrated through the analysis of a series of world class athletics running times, extreme temperatures at a site in the Antarctic, and sea-level extremes on the east coast of England
Spatial analysis and simulation of extreme coastal flooding scenarios for national-scale emergency planning
No full textThe UK has a long history of coastal flooding, driven by large-scale low-pressure weather systems which can result in flooding over large spatial areas. Traditional coastal flood risk analysis is, however, often undertaken at local scales and hence does not consider the likelihood of simultaneous flooding over larger areas. The flooding within the UK over the Winter of 2013/2014 was notable both for its long duration, lasting over two months, and its spatial extent, affecting many different areas of England and Wales. It is thus apparent that to plan and prepare for these types of extreme event it is necessary to consider the likelihood of flood events arising at different locations simultaneously (i.e. to consider the spatial dependence of extreme flood events). This paper describes the application of a state-of-the-art multivariate extreme value methodology to extreme sea levels and wave conditions around the coast of England and Wales. The output of the analysis comprises a synthetic set of extreme but plausible events that explicitly captures the dependence between sea conditions at different spatial locations around the coast. These simulated extreme events can be used for emergency management and advanced flood risk analysis
A sequential smoothing algorithm with linear computational cost.
No full textIn this paper we propose a new particle smoother that has a computational complexity of O(N), where N is the number of particles. This compares favourably with the O(N2) computational cost of most smoothers. The new method also overcomes some degeneracy problems in existing algorithms. Through simulation studies we show that substantial gains in efficiency are obtained for practical amounts of computational cost. It is shown both through these simulation studies, and by the analysis of an athletics dataset, that our new method also substantially outperforms the simple filter-smoother, the only other smoother with computational cost that is O(N)
Applying emulators for improved flood risk analysis
No full textFlood risk analysis often involves the integration of multivariate probability distributions over a domain defined by a consequence function. Often, solutions of this risk integral involves Monte-Carlo sampling techniques, whereby 1000’s of potential flood events are generated. It is necessary to evaluate the consequence of flooding for each sampled event. A significant computational time is required in running flood related physical process models, making it computationally impractical to evaluate flood risk using this approach. To overcome the computational challenges, this paper focusses on the Gaussian Process Emulator (GPE) meta-modelling approach. Traditionally, a “look-up table” method is used when a large number of simulations from a numerical model are required. This approach typically involves simulating conditions defined across a regular matrix, and then linearly interpolating intermediate conditions. In this paper we compare a traditional “look-up table” approach to the GPE and analyse their performance in approximating SWAN wave transformation model. In both cases, selecting an appropriate training design set is important and is taken into consideration in the analysis. The analysis shows that the GPE approach requires significantly fewer SWAN runs to obtain similar (or better) accuracies, enabling a substantial reduction in computation time, hence aiding the practicality of Monte-Carlo sampling techniques in advanced flood risk modelling
The challenges of including historical events using Bayesian methods to improve flood flow estimates in the United Kingdom: A practitioner's point of view
No full textEstimates of design flood flows; are important for the design of a wide variety of civil engineering structures. In the United Kingdom, the Flood Estimation Handbook (FEH) methodologies are used by practitioners to estimate flood flows. Until recently it was challenging for practitioners to include additional non‐continuous flood information, such as historical flood descriptions from historical archives, in their analyses, to reduce the uncertainty within the FEH approaches. This paper shows how Bayesian statistical methods can be applied to historical data which have degrees of uncertainty associated with then to improve design flood flows. The paper uses the River Avon at Evesham in the United Kingdom as a case study to illustrate the advantages of the method. The inclusion of historical information at this site improves the estimates of the one in 100 year flood flow compared to the values generated by the FEH pooling group method. The paper makes recommendations as to how practitioners could be encouraged to use historical flood water levels in their analysis of floods more regularly
- …
