École Centrale de Nantes

Abstract The genetic make-up of an individual contributes to the susceptibility and response to viral infection. Although environmental, clinical and social factors have a role in the chance of exposure to SARS-CoV-2 and the severity of COVID-19 1,2 , host genetics may also be important. Identifying host-specific genetic factors may reveal biological mechanisms of therapeutic relevance and clarify causal relationships of modifiable environmental risk factors for SARS-CoV-2 infection and outcomes. We formed a global network of researchers to investigate the role of human genetics in SARS-CoV-2 infection and COVID-19 severity. Here we describe the results of three genome-wide association meta-analyses that consist of up to 49,562 patients with COVID-19 from 46 studies across 19 countries. We report 13 genome-wide significant loci that are associated with SARS-CoV-2 infection or severe manifestations of COVID-19. Several of these loci correspond to previously documented associations to lung or autoimmune and inflammatory diseases 3–7 . They also represent potentially actionable mechanisms in response to infection. Mendelian randomization analyses support a causal role for smoking and body-mass index for severe COVID-19 although not for type II diabetes. The identification of novel host genetic factors associated with COVID-19 was made possible by the community of human genetics researchers coming together to prioritize the sharing of data, results, resources and analytical frameworks. This working model of international collaboration underscores what is possible for future genetic discoveries in emerging pandemics, or indeed for any complex human disease.

Biped robots form a subclass of legged or walking robots. The study of mechanical legged motion has been motivated by its potential use as a means of locomotion in rough terrain, as well as its potential benefits to prothesis development and testing. The paper concentrates on issues related to the automatic control of biped robots. More precisely, its primary goal is to contribute a means to prove asymptotically-stable walking in planar, underactuated biped robot models. Since normal walking can be viewed as a periodic solution of the robot model, the method of Poincare sections is the natural means to study asymptotic stability of a walking cycle. However, due to the complexity of the associated dynamic models, this approach has had limited success. The principal contribution of the present work is to show that the control strategy can be designed in a way that greatly simplifies the application of the method of Poincare to a class of biped models, and, in fact, to reduce the stability assessment problem to the calculation of a continuous map from a subinterval of R to itself. The mapping in question is directly computable from a simulation model. The stability analysis is based on a careful formulation of the robot model as a system with impulse effects and the extension of the method of Poincare sections to this class of models.

Abstract The present paper proposes a thorough comparison of twenty hyperelastic models for rubber-like materials. The ability of these models to reproduce different types of loading conditions is analyzed thanks to two classical sets of experimental data. Both material parameters and the stretch range of validity of each model are determined by an efficient fitting procedure. Then, a ranking of these twenty models is established, highlighting new efficient constitutive equations that could advantageously replace well-known models, which are widely used by engineers for finite element simulation of rubber parts.

Abstract A methodology for solving three‐dimensional crack problems with geometries that are independent of the mesh is described. The method is based on the extended finite element method, in which the crack discontinuity is introduced as a Heaviside step function via a partition of unity. In addition, branch functions are introduced for all elements containing the crack front. The branch functions include asymptotic near‐tip fields that improve the accuracy of the method. The crack geometry is described by two signed distance functions, which in turn can be defined by nodal values. Consequently, no explicit representation of the crack is needed. Examples for three‐dimensional elastostatic problems are given and compared to analytic and benchmark solutions. The method is readily extendable to inelastic fracture problems. Copyright © 2002 John Wiley & Sons, Ltd.

This letter describes algorithms for nonnegative matrix factorization (NMF) with the β-divergence (β-NMF). The β-divergence is a family of cost functions parameterized by a single shape parameter β that takes the Euclidean distance, the Kullback-Leibler divergence, and the Itakura-Saito divergence as special cases (β = 2, 1, 0 respectively). The proposed algorithms are based on a surrogate auxiliary function (a local majorization of the criterion function). We first describe a majorization-minimization algorithm that leads to multiplicative updates, which differ from standard heuristic multiplicative updates by a β-dependent power exponent. The monotonicity of the heuristic algorithm can, however, be proven for β ∈ (0, 1) using the proposed auxiliary function. Then we introduce the concept of the majorization-equalization (ME) algorithm, which produces updates that move along constant level sets of the auxiliary function and lead to larger steps than MM. Simulations on synthetic and real data illustrate the faster convergence of the ME approach. The letter also describes how the proposed algorithms can be adapted to two common variants of NMF: penalized NMF (when a penalty function of the factors is added to the criterion function) and convex NMF (when the dictionary is assumed to belong to a known subspace).

For intelligent systems to make best use of the audio modality, it is important that they can recognize not just speech and music, which have been researched as specific tasks, but also general sounds in everyday environments. To stimulate research in this field we conducted a public research challenge: the IEEE Audio and Acoustic Signal Processing Technical Committee challenge on Detection and Classification of Acoustic Scenes and Events (DCASE). In this paper, we report on the state of the art in automatically classifying audio scenes, and automatically detecting and classifying audio events. We survey prior work as well as the state of the art represented by the submissions to the challenge from various research groups. We also provide detail on the organization of the challenge, so that our experience as challenge hosts may be useful to those organizing challenges in similar domains. We created new audio datasets and baseline systems for the challenge; these, as well as some submitted systems, are publicly available under open licenses, to serve as benchmarks for further research in general-purpose machine listening.

Abstract A large number of urban surface energy balance models now exist with different assumptions about the important features of the surface and exchange processes that need to be incorporated. To date, no comparison of these models has been conducted; in contrast, models for natural surfaces have been compared extensively as part of the Project for Intercomparison of Land-surface Parameterization Schemes. Here, the methods and first results from an extensive international comparison of 33 models are presented. The aim of the comparison overall is to understand the complexity required to model energy and water exchanges in urban areas. The degree of complexity included in the models is outlined and impacts on model performance are discussed. During the comparison there have been significant developments in the models with resulting improvements in performance (root-mean-square error falling by up to two-thirds). Evaluation is based on a dataset containing net all-wave radiation, sensible heat, and latent heat flux observations for an industrial area in Vancouver, British Columbia, Canada. The aim of the comparison is twofold: to identify those modeling approaches that minimize the errors in the simulated fluxes of the urban energy balance and to determine the degree of model complexity required for accurate simulations. There is evidence that some classes of models perform better for individual fluxes but no model performs best or worst for all fluxes. In general, the simpler models perform as well as the more complex models based on all statistical measures. Generally the schemes have best overall capability to model net all-wave radiation and least capability to model latent heat flux.

Describes the design, construction and control of an experimental bipedal robot platform for the study of walking.

Abstract We present a level set method for treating the growth of non‐planar three‐dimensional cracks.The crack is defined by two almost‐orthogonal level sets (signed distance functions). One of them describes the crack as a two‐dimensional surface in a three‐dimensional space, and the second is used to describe the one‐dimensional crack front, which is the intersection of the two level sets. A Hamilton–Jacobi equation is used to update the level sets. A velocity extension is developed that preserves the old crack surface and can accurately generate the growing surface. The technique is coupled with the extended finite element method which approximates the displacement field with a discontinuous partition of unity. This displacement field is constructed directly in terms of the level sets, so the discretization by finite elements requires no explicit representation of the crack surface. Numerical experiments show the robustness of the method, both in accuracy and in treating cracks with significant changes in topology. Copyright © 2002 John Wiley & Sons, Ltd.

This paper proposes a new algorithm for automatic crack detection from 2D pavement images. It strongly relies on the localization of minimal paths within each image, a path being a series of neighboring pixels and its score being the sum of their intensities. The originality of the approach stems from the proposed way to select a set of minimal paths and the two postprocessing steps introduced to improve the quality of the detection. Such an approach is a natural way to take account of both the photometric and geometric characteristics of pavement images. An intensive validation is performed on both synthetic and real images (from five different acquisition systems), with comparisons to five existing methods. The proposed algorithm provides very robust and precise results in a wide range of situations, in a fully unsupervised manner, which is beyond the current state of the art.

Few theoretical methods and experimental data exist for the analysis of tunnel‐face stability in cohesionless soils. The present paper addresses a series of practical questions by using centrifugal‐model tests. The values of limit internal‐support pressures are given for various conditions (density of the sand, position of the tunnel with respect to the ground surface). These values are shown to be low as predicted by the latest limit‐calculation models and collapse is shown to be sudden. The geometry of the failure zone is depicted for different embedment depths. The initial mechanism appears to be of a bulk shape with a limited extent in front of the face. The presence of a short unlined length of tunnel at the face is also investigated. Data are presented about its effect on the failure mechanism, on the limit pressure, and on stress transfer onto the tunnel lining at collapse. The results obtained in the model tests are in general agreement with present knowledge of full‐scale situations.

Numerical crack propagation schemes were augmented in an elegant manner by the X-FEM method. The use of special tip enrichment functions, as well as a discontinuous function along the sides of the crack allows one to do a complete crack analysis virtually without modifying the underlying mesh, which is of industrial interest, especially when a numerical model for crack propagation is desired. This paper improves the implementation of the X-FEM method for stress analysis around cracks in three ways. First, the enrichment strategy is revisited. The conventional approach uses a ‘topological’ enrichment (only the elements touching the front are enriched). We suggest a ‘geometrical’ enrichment in which a given domain size is enriched. The improvements obtained with this enrichment are discussed. Second, the conditioning of the X-FEM both for topological and geometrical enrichments is studied. A preconditioner is introduced so that ‘off the shelf’ iterative solver packages can be used and perform as well on X-FEM matrices as on standard FEM matrices. The preconditioner uses a local (nodal) Cholesky based decomposition. Third, the numerical integration scheme to build the X-FEM stiffness matrix is dramatically improved for tip enrichment functions by the use of an ad hoc integration scheme. A 2D benchmark problem is designed to show the improvements and the robustness. Copyright © 2005 John Wiley & Sons, Ltd.

Abstract Urban land surface schemes have been developed to model the distinct features of the urban surface and the associated energy exchange processes. These models have been developed for a range of purposes and make different assumptions related to the inclusion and representation of the relevant processes. Here, the first results of Phase 2 from an international comparison project to evaluate 32 urban land surface schemes are presented. This is the first large‐scale systematic evaluation of these models. In four stages, participants were given increasingly detailed information about an urban site for which urban fluxes were directly observed. At each stage, each group returned their models' calculated surface energy balance fluxes. Wide variations are evident in the performance of the models for individual fluxes. No individual model performs best for all fluxes. Providing additional information about the surface generally results in better performance. However, there is clear evidence that poor choice of parameter values can cause a large drop in performance for models that otherwise perform well. As many models do not perform well across all fluxes, there is need for caution in their application, and users should be aware of the implications for applications and decision making. Copyright © 2010 Royal Meteorological Society

This paper studies a fully Bayesian algorithm for endmember extraction and abundance estimation for hyperspectral imagery. Each pixel of the hyperspectral image is decomposed as a linear combination of pure endmember spectra following the linear mixing model. The estimation of the unknown endmember spectra is conducted in a unified manner by generating the posterior distribution of abundances and endmember parameters under a hierarchical Bayesian model. This model assumes conjugate prior distributions for these parameters, accounts for nonnegativity and full-additivity constraints, and exploits the fact that the endmember proportions lie on a lower dimensional simplex. A Gibbs sampler is proposed to overcome the complexity of evaluating the resulting posterior distribution. This sampler generates samples distributed according to the posterior distribution and estimates the unknown parameters using these generated samples. The accuracy of the joint Bayesian estimator is illustrated by simulations conducted on synthetic and real AVIRIS images.

Nature-based Solutions (NBS) are increasingly promoted to support sustainable and resilient urban planning. However, design and planning urban NBS targeted at the needs of the local context require knowledge about the causal relationships between NBS, ecosystem services (ES) and urban challenges (UC) This paper aims at contributing to this knowledge, by systematically identifying nexuses (i.e. qualitative links) between UC, ES and NBS, and describing plausible causal relationships. A conceptual UC-ES-NBS criteria framework was built, and used to guide a two-step systematic literature review on current UC and on the supply of ES by urban NBS. This was followed by a non-systematic literature review, which complemented the previous one by unveiling knowledge gaps on the biophysical and social processes and attributes on which specific ES classes depend. The non-systematic review was also used to identify additional NBS. The UC review identified 18 UC and 58 sub-challenges, and illustrated which UC were more studied, according to the type of literature and environmental and socio-economic attributes of urban contexts. The ES review led to the development of an urban NBS classification, and supported the identification of UC-ES and ES-NBS nexuses, which were analysed and classified into four groups of causal relationship. For the nexuses identified as direct plausible causal relationship, the main processes and attributes on which the supply of specific ES depend were pointed out. Relationships between UC, ES, NBS, processes, and attributes were represented in the form of network diagrams. Our results can be used to support urban policies aimed at mainstreaming NBS and as a basis to further understand UC-ES-NBS relationships.

This article is a survey of deep learning methods for single and multiple sound source localization, with a focus on sound source localization in indoor environments, where reverberation and diffuse noise are present. We provide an extensive topography of the neural network-based sound source localization literature in this context, organized according to the neural network architecture, the type of input features, the output strategy (classification or regression), the types of data used for model training and evaluation, and the model training strategy. Tables summarizing the literature survey are provided at the end of the paper, allowing a quick search of methods with a given set of target characteristics.

Surface electromyography for noninvasive characterization of muscle. Exerc. Sport Sci. Rev., Vol. 29, No. 1, pp 20-25, 2001. Linear electrode arrays are used for noninvasive muscle characterization to study individual motor unit properties and the myoelectric manifestations of muscle fatigue during sustained contractions. The location of an electrode pair with respect to the innervation zone(s), the deterministic rather than stochastic nature of the signal, and the possibility of noninvasive fiber typing are discussed.

During isometric contractions of increasing strength, motor units (MUs) are recruited by the central nervous system in an orderly manner starting with the smallest, with muscle fibers that usually show the lowest conduction velocity (CV). Theory predicts that the higher the velocity of propagation of the action potential, the higher the power at high frequencies of the detected surface signal. These considerations suggest that the power spectral density of the surface detected electromyogram (EMG) signal may give indications about the MU recruitment process. The purpose of this paper is to investigate the potential and limitations of spectral analysis of the surface EMG signal as a technique for the investigation of muscle force control. The study is based on a simulation approach and on an experimental investigation of the properties of surface EMG signals detected from the biceps brachii during isometric linearly increasing torque contractions. Both simulation and experimental data indicate that volume conductor properties play an important role as confounding factors that may mask any relation between EMG spectral variables and estimated CV as a size principle parameter during ramp contractions. The correlation between spectral variables and CV is thus significantly lower when the MU pool is not stable than during constant-torque isometric contractions. Our results do not support the establishment of a general relationship between spectral EMG variables and torque or recruitment strategy.

Public evaluation campaigns and datasets promote active development in target research areas, allowing direct comparison of algorithms. The second edition of the challenge on detection and classification of acoustic scenes and events (DCASE 2016) has offered such an opportunity for development of the state-of-the-art methods, and succeeded in drawing together a large number of participants from academic and industrial backgrounds. In this paper, we report on the tasks and outcomes of the DCASE 2016 challenge. The challenge comprised four tasks: acoustic scene classification, sound event detection in synthetic audio, sound event detection in real-life audio, and domestic audio tagging. We present each task in detail and analyze the submitted systems in terms of design and performance. We observe the emergence of deep learning as the most popular classification method, replacing the traditional approaches based on Gaussian mixture models and support vector machines. By contrast, feature representations have not changed substantially throughout the years, as mel frequency-based representations predominate in all tasks. The datasets created for and used in DCASE 2016 are publicly available and are a valuable resource for further research.

Abstract A paradigm is developed for generating structured finite element models from solid models by means of implicit surface definitions. The implicit surfaces are defined by radial basis functions. Internal features, such as material interfaces, sliding interfaces and cracks are treated by enrichment techniques developed in the extended finite element method. Methods for integrating the weak form for such models are proposed. These methods simplify the generation of finite element models. Results presented for several examples show that the accuracy of this method is comparable to standard unstructured finite element methods. Copyright © 2002 John Wiley & Sons, Ltd.