Search

Analysis of phylogenetic trees has become an essential tool in epidemiology. Likelihood-based methods fit models to phylogenies to draw inferences about the phylodynamics and history of viral transmission. However, these methods are often computationally expensive, which limits the complexity and realism of phylodynamic models and makes them ill-suited for informing policy decisions in real-time during rapidly developing outbreaks. Likelihood-free methods using deep learning are pushing the boundaries of inference beyond these constraints. In this paper, we extend, compare, and contrast a recently developed deep learning method for likelihood-free inference from trees. We trained multiple deep neural networks using phylogenies from simulated outbreaks that spread among 5 locations and found they achieve close to the same levels of accuracy as Bayesian inference under the true simulation model. We compared robustness to model misspecification of a trained neural network to that of a Bayesian method. We found that both models had comparable performance, converging on similar biases. We also implemented a method of uncertainty quantification called conformalized quantile regression that we demonstrate has similar patterns of sensitivity to model misspecification as Bayesian highest posterior density (HPD) and greatly overlap with HPDs, but have lower precision (more conservative). Finally, we trained and tested a neural network against phylogeographic data from a recent study of the SARS-Cov-2 pandemic in Europe and obtained similar estimates of region-specific epidemiological parameters and the location of the common ancestor in Europe. Along with being as accurate and robust as likelihood-based methods, our trained neural networks are on average over 3 orders of magnitude faster after training. Our results support the notion that neural networks can be trained with simulated data to accurately mimic the good and bad statistical properties of the likelihood functions of generative phylogenetic models. [ABSTRACT FROM AUTHOR]

New propulsion systems in aircrafts must meet strict regulations and emission limitations. The Flightpath 2050 goals set by the Advisory Council for Aviation Research and Innovation in Europe (ACARE) include reductions of 75%, 90%, and 65% in CO2, NOx, and noise, respectively. These goals are not fully satisfied by marginal improvements in gas turbine technology or aircraft design. A novel control design procedure for the next generation of turbofan engines is proposed in this paper to improve Full Authority Digital Engine Control (FADEC) systems and reduce the emission levels to meet the Flightpath 2050 regulations. Hence, an Adaptive Network–based Fuzzy Inference System (ANFIS), nonlinear autoregressive network with exogenous inputs (NARX) techniques, and the block-structure Hammerstein–Wiener approach are used to develop a model for a turbofan engine. The Min–Max control structure is chosen as the most widely used practical control algorithm for gas turbine aero engines. The objective function is considered to minimize the emission level for the engine in a pre-defined maneuver while keeping the engine performance in different aspects. The Genetic Algorithm (GA) is applied to find the optimized control structure. The results confirm the effectiveness of the proposed approach in emission reduction for the next generation of turbofan engines. [ABSTRACT FROM AUTHOR]

Historical maps are popular reference tools for historical, archaeological and temporal analysis, and there has recently been an increase in their use. However, for various reasons, the planimetric accuracy of maps produced before the nineteenth century is usually considered to be lower than today. In this paper, a new method for assessing maps is proposed, using a series of processes, such as radial-based function artificial neural network, magnetic declination, and also MapAnalyst software. The map used in the current study (of Europe and the Mediterranean Sea) is a small-scale map; therefore, control points were produced by taking reference from large-scale maps drawn by the same cartographer, Piri Reis (c.1465–1553). While developing this method, affine transformation (six parameters) was compared in terms of planimetric accuracy. The results indicate that Piri Reis's Mediterranean map offers us unique information in many areas. [ABSTRACT FROM AUTHOR]

Depression is ranked as the largest contributor to global disability and is also a major reason for suicide. Still, many individuals suffering from forms of depression are not treated for various reasons. Previous studies have shown that depression also has an effect on language usage and that many depressed individuals use social media platforms or the internet in general to get information or discuss their problems. This paper addresses the early detection of depression using machine learning models based on messages on a social platform. In particular, a convolutional neural network based on different word embeddings is evaluated and compared to a classification based on user-level linguistic metadata. An ensemble of both approaches is shown to achieve state-of-the-art results in a current early detection task. Furthermore, the currently popular $ERDE$ E R D E score as metric for early detection systems is examined in detail and its drawbacks in the context of shared tasks are illustrated. A slightly modified metric is proposed and compared to the original score. Finally, a new word embedding was trained on a large corpus of the same domain as the described task and is evaluated as well. [ABSTRACT FROM AUTHOR]

Deep learning techniques (in particular convolutional neural networks, CNNs) have recently emerged as a promising approach for statistical downscaling due to their ability to learn spatial features from huge spatio-temporal datasets. However, existing studies are based on complex models, applied to particular case studies and using simple validation frameworks, which makes difficult a proper assessment of the (possible) added value offered by these techniques. As a result, these models are usually seen as black-boxes generating distrust among the climate community, particularly in climate change problems. In this paper we undertake a comprehensive assessment of deep learning techniques for continental-scale statistical downscaling, building on the VALUE validation framework. In particular, different CNN models of increasing complexity are applied for downscaling temperature and precipitation over Europe, comparing them with a few standard benchmark methods from VALUE (linear and generalized linear models) which have been traditionally used for this purpose. Besides analyzing the adequacy of different components and topologies, we also focus on their extrapolation capability, a critical point for their possible application in climate change studies. To do this, we use a warm test period as surrogate of possible future climate conditions. Our results show that, whilst the added value of CNNs is mostly limited to the reproduction of extremes for temperature, these techniques do outperform the classic ones for the case of precipitation for most aspects considered. This overall good performance, together with the fact that they can be suitably applied to large regions (e.g. continents) without worrying about the spatial features being considered as predictors, can foster the use of statistical approaches in international initiatives such as CORDEX. [ABSTRACT FROM AUTHOR]

BACKGROUND: The assessment of bone age and skeletal maturity and its comparison to chronological age is an important task in the medical environment for the diagnosis of pediatric endocrinology, orthodontics and orthopedic disorders, and legal environment in what concerns if an individual is a minor or not when there is a lack of documents. Being a time-consuming activity that can be prone to inter- and intra-rater variability, the use of methods which can automate it, like Machine Learning techniques, is of value. OBJECTIVE: The goal of this paper is to present the state of the art evidence, trends and gaps in the research related to bone age assessment studies that make use of Machine Learning techniques. METHOD: A systematic literature review was carried out, starting with the writing of the protocol, followed by searches on three databases: Pubmed, Scopus and Web of Science to identify the relevant evidence related to bone age assessment using Machine Learning techniques. One round of backward snowballing was performed to find additional studies. A quality assessment was performed on the selected studies to check for bias and low quality studies, which were removed. Data was extracted from the included studies to build summary tables. Lastly, a meta-analysis was performed on the performances of the selected studies. RESULTS: 26 studies constituted the final set of included studies. Most of them proposed automatic systems for bone age assessment and investigated methods for bone age assessment based on hand and wrist radiographs. The samples used in the studies were mostly comprehensive or bordered the age of 18, and the data origin was in most of cases from United States and West Europe. Few studies explored ethnic differences. CONCLUSIONS: There is a clear focus of the research on bone age assessment methods based on radiographs whilst other types of medical imaging without radiation exposure (e.g. magnetic resonance imaging) are not much explored in the literature. Also, socioeconomic and other aspects that could influence in bone age were not addressed in the literature. Finally, studies that make use of more than one region of interest for bone age assessment are scarce. [ABSTRACT FROM AUTHOR]

Motivated by the increasing integration among electricity markets, in this paper we propose two different methods to incorporate market integration in electricity price forecasting and to improve the predictive performance. First, we propose a deep neural network that considers features from connected markets to improve the predictive accuracy in a local market. To measure the importance of these features, we propose a novel feature selection algorithm that, by using Bayesian optimization and functional analysis of variance, evaluates the effect of the features on the algorithm performance. In addition, using market integration, we propose a second model that, by simultaneously predicting prices from two markets, improves the forecasting accuracy even further. As a case study, we consider the electricity market in Belgium and the improvements in forecasting accuracy when using various French electricity features. We show that the two proposed models lead to improvements that are statistically significant. Particularly, due to market integration, the predictive accuracy is improved from 15.7% to 12.5% sMAPE (symmetric mean absolute percentage error). In addition, we show that the proposed feature selection algorithm is able to perform a correct assessment, i.e. to discard the irrelevant features. [ABSTRACT FROM AUTHOR]

Buildings, energy, and the environment are key issues facing construction around the world. The energy efficiency of buildings is a key topic when it comes to reducing the world's energy consumption, releasing harmful gases, and global climate change, as they consume about 40% of the world's energy supplies. Heat losses in buildings reduce the energy performance of buildings and are basically important to them. In the paper, the authors focus on the main problems related to heat losses generated by chimney systems, which are inseparable equipment of building structures, resulting in lower energy efficiency and, at the same time, technical efficiency and durability of the building partitions themselves. Authors present thermal imaging with its contribution to the detection of heat losses, thermal bridges, insulation problems, and other performance disturbances, and then verifications using appropriate simulation models. The mathematical apparatus of artificial neural networks was implemented to predict the temperature distributions on the surfaces of prefabricated chimney solutions. In Europe, we can often find a large building substance equipped with traditional chimneys, which disrupts the current trend of striving to reduce energy consumption, especially that derived from fossil fuels. Speaking of energy-efficient buildings, one should not ignore those that, without additional security and modern installations, are constantly used in a very wide range. Therefore, the article deals with an essential problem that is not perceived in design studies and during the operation period as having a basis in incorrect architectural solutions and which can be easily eliminated. It concerns the cooling of internal partitions of buildings on their last storeys, in places where chimneys are located, regardless of their function. The authors of the paper decided to take a closer look at this phenomenon, which may allow the limiting of its effects and at the same time reduce its impact on the energy performance of technologically older buildings. [ABSTRACT FROM AUTHOR]

The aim of this paper is to present the development and evaluation of neural network based identification algorithms for the prediction of outdoor air temperature using acquired data from four European cities (Ancona – Italy, Chania – Greece, Granada – Spain and Mollet – Spain). Different neural network topologies (feed forward, cascade and elman) have been tested to identify the most suitable for each city. The efficiency of the prediction is validated by comparing predicted and measured outdoor air temperature. Furthermore, statistical tools such as R 2 , and root mean square error (rmse) are used to evaluate the annual performance of the neural network. The comparison of measured and predicted outdoor air temperature ( R 2 > 0.9, rmse <2 °C) confirms the accurate training of the neural network for all four European cities. All work has been contacted using Matlab's environment. [ABSTRACT FROM AUTHOR]

Satellite-derived high-resolution precipitation products (HRPP) have been developed to address the needs of the user community and are now available with 0.25° × 0.25° (or less) subdaily resolutions. This paper evaluates a number of commonly available satellite-derived HRPPs covering northwest Europe over a 6-yr period. Precipitation products include the Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA), the Climate Prediction Center (CPC) morphing (CMORPH) technique, the CPC merged microwave technique, the Naval Research Laboratory (NRL) blended technique, and the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN) technique. In addition, the Geosynchronous Operational Environmental Satellite (GOES) precipitation index (GPI) and the European Centre for Medium-Range Weather Forecasting (ECMWF) operational forecast model products are included for comparison. Surface reference data from the European radar network is used as ground truth, supported by the Global Precipitation Climatology Centre (GPCC) precipitation gauge analysis and gauge data over the United Kingdom. Measures of correlation, bias ratio, probability of detection, and false alarm ratio are used to evaluate the products. Results show that satellite products generally exhibit a seasonal cycle in correlation, bias ratio, probability of detection, and false alarm ratio, with poorer statistics during the winter. The ECMWF model also shows a seasonal cycle in the correlation, although the results are poorer during the summer, while the bias ratio, probability of detection, and false alarm ratio are consistent through all seasons. Importantly, all the satellite HRPPs underestimate precipitation over northwest Europe in all seasons. [ABSTRACT FROM AUTHOR]

We present an approach to estimate gross primary production (GPP) using a remotely sensed biophysical vegetation product (fraction of absorbed photosynthetically active radiation, FAPAR) from the European Commission Joint Research Centre (JRC) in conjunction with GPP estimates from eddy covariance measurement towers in Europe. By analysing the relationship between the cumulative growing season FAPAR and annual GPP by vegetation type, we find that the former can be used to accurately predict the latter. The root mean square error of prediction is of the order of 250 gC m−2 yr−1. The cumulative growing season FAPAR integrates over a number of effects relevant for GPP such as the length of the growing season, the vegetation's response to environmental conditions and the amount of light harvested that is available for photosynthesis. We corroborate the proposed GPP estimate (noted FAPAR-based productivity assessment+land cover, FPA+LC) on the continental scale with results from the MOD17+radiation-use efficiency model, an artificial neural network up-scaling approach (ANN) and the Lund–Potsdam–Jena managed Land biosphere model (LPJmL). The closest agreement of the mean spatial GPP pattern among the four models is between FPA+LC and ANN ( R2= 0.74). At least some of the discrepancy between FPA-LC and the other models result from biases of meteorological forcing fields for MOD17+, ANN and LPJmL. Our analysis further implies that meteorological information is to a large degree redundant for GPP estimation when using the JRC-FAPAR. A major advantage of the FPA+LC approach presented in this paper lies in its simplicity and that it requires no additional meteorological input driver data that commonly introduce substantial uncertainty. We find that results from different data-oriented models may be robust enough to evaluate process-oriented models regarding the mean spatial pattern of GPP, while there is too little consensus among the diagnostic models for such purpose regarding inter-annual variability. [ABSTRACT FROM AUTHOR]

Abstract Recently flux tower data have become available for a variety of ecosystems under different climatic and edaphic conditions. Although Flux tower data represent point measurements with a footprint of typically 1 km × 1 km they can be used to validate models and to spatialize biospheric fluxes at regional and continental scales. In this paper we present a study where biospheric flux data collected in the EUROFLUX project were used to train a neural network simulator to provide spatial (1 km × 1 km) and temporal (weekly) estimates of carbon fluxes of European forests at continental scale. The novelty of the approach is that flux data were used to constrain and parameterize the neural network structure using a limited number of input driving variables. The overall European carbon uptake from this analysis was 0.47 Gt C yr-1 with distinctive differences between boreal and temperate regions. The length of the growing season is longer in the south of Europe (about 32 weeks), compared with north and central Europe, which have a similar length-growing season (about 27 weeks). A peak in respiration was depicted in spring at continental scale as a coherent signal which parallel the construction respiration increase at the onset of the season as usually shown by leaf level measurements. [ABSTRACT FROM AUTHOR]

ABSTRACT The paper focuses on the Transalpine Freight Transport systems in the light of the future integration of single national transport systems into the European transport network. The environmental, social and institutional peculiarities of this 'region' have favoured--in the past--the development of strong nationally-oriented policies, which are largely in contrast with the goals promoted by the European Union. The present analysis aims to highlight opportunities and limits inherent in the implementation of various new network projects, with a particular view on the planned changes of the Alpine transport system. In this framework, a concise description of the existing and 'planned' situation will be offered. In addition, some new forecasting analyses for road transport will be provided on the basis of environmentally-based transport scenarios. In particular, given the large size of our database on European transport flows, two different approaches will be compared, viz. the logit model and the neural network model. Logit models are well-known in the literature; however, applications of logit analysis to large samples are more rare. Neural networks are nowadays receiving a considerable attention as a new approach that is able to capture major patterns of spatial flows, on the basis of fuzzy and incomplete information. The tentative results of both approaches in this context may then be used as a benchmark for judging the results of other transport flow models and offer also a more 'flexible' range of results to policy actors. Furthermore, our study will present the assessment of trans-European freight flows based on interesting future scenarios related to further congestion and the introduction of eco-taxes on transport in Europe. [ABSTRACT FROM AUTHOR]

Fusion performance in tokamaks depends on the core and edge regions as well as on their nonlinear feedbacks. The achievable degree of edge confinement under the constraints of power handling in presence of a metallic wall is still an open question. Therefore, any improvement in the core temperature and density peaking is crucial for achieving target performance. This has motivated further progress in understanding core turbulent transport mechanisms, to help scenario development in present devices and improve predictive tools for ITER operations. In the last two decades, detailed experiments and their interpretation via the gyrokinetic theory of turbulent transport have led to a satisfactory level of understanding of the heat, particle, and momentum transport channels and of their mutual interactions. This paper presents some highlights of the progress, which stems from joint work of several devices and theory groups, in Europe and worldwide within the International Tokamak Physics Activities framework. On the other hand, the achievement of predictive capabilities of plasma profiles via integrated modeling, which also accounts for the nonlinear interactions inherent to the multi-channel nature of transport, is a priority in view of ITER. This requires using faster, reduced models, and the extent to which they capture the complex physics described by nonlinear gyrokinetics must be carefully evaluated. Present quasi-linear models match well experiments in baseline scenarios, and thus offer reliable predictions for the ITER reference scenario, but have issues in advanced scenarios. Some of these challenges are examined and discussed. In the longer term, advances in high performance computing will continue to drive physics discovery through increasingly complex gyrokinetic simulations, allowing also further development of reduced models. The development of neural network surrogate models is another recent advance that bridges the gap towards physics-based fast models for optimization and control applications. [ABSTRACT FROM AUTHOR]