Search

International audience; We address the problem of finding reliable dense correspondences between a pair of images. This is a challenging task due to strong appearance differences between the corresponding scene elements and ambiguities generated by repetitive patterns. The contributions of this work are threefold. First, inspired by the classic idea of disambiguating feature matches using semi-local constraints, we develop an end-to-end trainable convolutional neural network architecture that identifies sets of spatially consistent matches by analyzing neighbourhood consensus patterns in the 4D space of all possible correspondences between a pair of images without the need for a global geometric model. Second, we demonstrate that the model can be trained effectively from weak supervision in the form of matching and non-matching image pairs without the need for costly manual annotation of point to point correspondences. Third, we show the proposed neighbourhood consensus network can be applied to a range of matching tasks including both category- and instance-level matching, obtaining the state-of-the-art results on the PF, TSS, InLoc and HPatches benchmarks.

International audience; The current study tries a new approach to simulating interactions between waves and seagrass through Smoothed Particle Hydrodynamics (SPH). In this model, the plants are defined as a solid that respects Hooke's law, and are assumed to have direct interaction with the fluid. Given the characteristics of the SPH method, especially in terms of computational time, the dimensions of the simulations were limited. The first goal of the current study was to optimize the approach to avoid reaching certain limits such as the rupture of the simulated plant. Plant movements under waves and/or currents have been studied by several authors in various in-situ, physical, and numerical experiments concerning various vegetation species, thus proving that plant movements can be successfully reproduced by SPH 2D/3D. Manning's roughness coefficient, n, was calculated to confirm that the results were in accordance with what had been measured in flume studies. Even though there is still room for improvement, it is shown that this method can be used to estimate Manning's coefficient for coastal vegetation (seagrass and saltmarsh vegetation) and to greatly improve the modeling and forecasting of coastal erosion and storm surge risks by including the effects of vegetation in integrated models.

The consolidation of spatial memory depends on the reactivation (‘replay’) of hippocampal place cells that were active during recent behaviour. Such reactivation is observed during sharp-wave ripples (SWRs)—synchronous oscillatory electrical events that occur during non-rapid-eye-movement (non-REM) sleep and whose disruption impairs spatial memory. Although the hippocampus also encodes a wide range of non-spatial forms of declarative memory, it is not yet known whether SWRs are necessary for such memories. Moreover, although SWRs can arise from either the CA3 or the CA2 region of the hippocampus, the relative importance of SWRs from these regions for memory consolidation is unknown. Here we examine the role of SWRs during the consolidation of social memory—the ability of an animal to recognize and remember a member of the same species—focusing on CA2 because of its essential role in social memory. We find that ensembles of CA2 pyramidal neurons that are active during social exploration of previously unknown conspecifics are reactivated during SWRs. Notably, disruption or enhancement of CA2 SWRs suppresses or prolongs social memory, respectively. Thus, SWR-mediated reactivation of hippocampal firing related to recent experience appears to be a general mechanism for binding spatial, temporal and sensory information into high-order memory representations, including social memory., This work was supported by the NVIDIA Corporation, an EMBO Postdoctoral Fellowship (ALTF 120-2017) and a K99 grant from the US National Institutes of Health (NIH; K99MH122582) (to A.O.); a Sir Henry Wellcome Postdoctoral Fellowship and K99 grant (K99MH120343) (to A.F.-R.); a National Alliance for Research on Schizophrenia and Depression (NARSAD) Young Investigator award from the Brain and Behavior Foundation founded by the Osterhaus family (to F.L.); and grants MH-104602 and MH-106629 from the National Institute of Mental Health (NIMH) and a grant from the Zegar Family Foundation (to S.A.S.).

To address the lack of statistical power and interpretability of genome-wide association studies (GWAS), gene-level analyses combine the p-values of individual single nucleotide polymorphisms (SNPs) into gene statistics. However, using all SNPs mapped to a gene, including those with low association scores, can mask the association signal of a gene.We therefore propose a new two-step strategy, consisting in first selecting the SNPs most associated with the phenotype within a given gene, before testing their joint effect on the phenotype. The recently proposed kernelPSI framework for kernel-based post-selection inference makes it possible to model non-linear relationships between features, as well as to obtain valid p-values that account for the selection step.In this paper, we show how we adapted kernelPSI to the setting of quantitative GWAS, using kernels to model epistatic interactions between neighboring SNPs, and post-selection inference to determine the joint effect of selected blocks of SNPs on a phenotype. We illustrate this tool on the study of two continuous phenotypes from the UKBiobank.We show that kernelPSI can be successfully used to study GWAS data and detect genes associated with a phenotype through the signal carried by the most strongly associated regions of these genes. In particular, we show that kernelPSI enjoys more statistical power than other gene-based GWAS tools, such as SKAT or MAGMA.kernelPSI is an effective tool to combine SNP-based and gene-based analyses of GWAS data, and can be used successfully to improve both statistical performance and interpretability of GWAS.

Trabajo presentado en la 17th European Conference on Computer Vision, celebrada en Tel Aviv (Israel), los días 23 y 27 de octubre de 2022, A critical limitation of current methods based on Neural Radiance Fields (NeRF) is that they are unable to quantify the uncertainty associated with the learned appearance and geometry of the scene. This information is paramount in real applications such as medical diagnosis or autonomous driving where, to reduce potentially catastrophic failures, the confidence on the model outputs must be included into the decision-making process. In this context, we introduce Conditional-Flow NeRF (CF-NeRF), a novel probabilistic framework to incorporate uncertainty quantification into NeRF-based approaches. For this purpose, our method learns a distribution over all possible radiance fields modelling which is used to quantify the uncertainty associated with the modelled scene. In contrast to previous approaches enforcing strong constraints over the radiance field distribution, CF-NeRF learns it in a flexible and fully data-driven manner by coupling Latent Variable Modelling and Conditional Normalizing Flows. This strategy allows to obtain reliable uncertainty estimation while preserving model expressivity. Compared to previous state-of-the-art methods proposed for uncertainty quantification in NeRF, our experiments show that the proposed method achieves significantly lower prediction errors and more reliable uncertainty values for synthetic novel view and depth-map estimation., This work is supported partly by the Chinese Scholarship Council (CSC) under grant (201906120031), by the Spanish government under project MoHuCo PID2020-120049RB-I00 and the Chistera project IPALM PCI2019-103386. We also thank Nvidia for hardware donation under the GPU grant program.

Glass transition temperature and related dynamics play an essential role in amorphous materials research since many of their properties and functionalities depend on molecular mobility. However, the temperature dependence of the structural relaxation time for a given glass former is only experimentally accessible after synthesizing it, implying a time-consuming and costly process. In this work, we propose combining artificial neural networks and disordered systems theory to estimate the glass transition temperature and the temperature dependence of the main relaxation time based on the knowledge of the molecule's chemical structure. This approach provides a way to assess the dynamics of molecular glass formers, with reasonable accuracy, even before synthesizing them. We expect this methodology to boost industrial development, save time and resources, and accelerate the scientific understanding of structure-properties relationships., We gratefully acknowledge the financial support from the Spanish Government “Ministerio de Ciencia e Innovación" (PID2019-104650GB-C21) and the Basque Government (IT-1566-22). This research was also funded by the Vietnam National Foundation for Science and Technology Development (NAFOSTED) under grant number 103.01-2019.318. We also acknowledge the support of NVIDIA Corporation with the donation of two GPUs used for this research., We also acknowledge support of the publication fee by the CSIC Open Access Publication Support Initiative through its Unit of Information Resources for Research (URICI).

The novel betacoronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused a worldwide pandemic (COVID-19) after emerging in Wuhan, China. Here we analyzed public host and viral RNA sequencing data to better understand how SARS-CoV-2 interacts with human respiratory cells. We identified genes, isoforms and transposable element families that are specifically altered in SARS-CoV-2-infected respiratory cells. Well-known immunoregulatory genes including CSF2, IL32, IL-6 and SERPINA3 were differentially expressed, while immunoregulatory transposable element families were upregulated. We predicted conserved interactions between the SARS-CoV-2 genome and human RNA-binding proteins such as the heterogeneous nuclear ribonucleoprotein A1 (hnRNPA1) and eukaryotic initiation factor 4 (eIF4b). We also identified a viral sequence variant with a statistically significant skew associated with age of infection, that may contribute to intracellular host–pathogen interactions. These findings can help identify host mechanisms that can be targeted by prophylactics and/or therapeutics to reduce the severity of COVID-19., Ferrarini & Lal et al. developed a novel bioinformatic pipeline to explore how SARS-CoV-2 interacts with human respiratory cells using public available host gene expression and viral genome sequence data. Several human genes and proteins were predicted to play a role in the viral life cycle and the host response to SARS-CoV-2 infection.

Author Correction: Genomic surveillance of enterovirus associated with aseptic meningitis cases in southern Spain, 2015-2018. Sci Rep. 2021 Nov 18;11(1):22899. doi: 10.1038/s41598-021-02194-2. New circulating Enterovirus (EV) strains often emerge through recombination. Upsurges of recombinant non-polio enteroviruses (NPEVs) associated with neurologic manifestations such as EVA71 or Echovirus 30 (E30) are a growing public health concern in Europe. Only a few complete genomes of EVs circulating in Spain are available in public databases, making it difficult to address the emergence of recombinant EVs, understand their evolutionary relatedness and the possible implication in human disease. We have used metagenomic (untargeted) NGS to generate full-length EV genomes from CSF samples of EV-positive aseptic meningitis cases in Southern Spain between 2015 and 2018. Our analyses reveal the co-circulation of multiple Enterovirus B (EV-B) types (E6, E11, E13 and E30), including a novel E13 recombinant form. We observed a genetic turnover where emergent lineages (C1 for E6 and I [tentatively proposed in this study] for E30) replaced previous lineages circulating in Spain, some concomitant with outbreaks in other parts of Europe. Metagenomic sequencing provides an effective approach for the analysis of EV genomes directly from PCR-positive CSF samples. The detection of a novel, disease-associated, recombinant form emphasizes the importance of genomic surveillance to monitor spread and evolution of EVs. This work was funded by the Junta de Andalucía, Consejería de Salud y Familias (Project number PI-0216-2019) and by the Acción Estratégica en Salud Intramural (Project number PI20CIII/00005). MD Fernandez-Garcia received a Miguel Servet Research Contract (CP18/00067) from the Strategic Action in Health 2018 and funded by National Institute of Health Carlos III (ISCIII). ESL acknowledges funding from the INCEPTION programme (Investissements d’Avenir grant ANR-16-CONV-0005). Sí

This work is based on the seminar titled ``Resiliency in Numerical Algorithm Design for Extreme Scale Simulations'' held March 1-6, 2020 at Schloss Dagstuhl, that was attended by all the authors. Naive versions of conventional resilience techniques will not scale to the exascale regime: with a main memory footprint of tens of Petabytes, synchronously writing checkpoint data all the way to background storage at frequent intervals will create intolerable overheads in runtime and energy consumption. Forecasts show that the mean time between failures could be lower than the time to recover from such a checkpoint, so that large calculations at scale might not make any progress if robust alternatives are not investigated. More advanced resilience techniques must be devised. The key may lie in exploiting both advanced system features as well as specific application knowledge. Research will face two essential questions: (1) what are the reliability requirements for a particular computation and (2) how do we best design the algorithms and software to meet these requirements? One avenue would be to refine and improve on system- or application-level checkpointing and rollback strategies in the case an error is detected. Developers might use fault notification interfaces and flexible runtime systems to respond to node failures in an application-dependent fashion. Novel numerical algorithms or more stochastic computational approaches may be required to meet accuracy requirements in the face of undetectable soft errors. The goal of this Dagstuhl Seminar was to bring together a diverse group of scientists with expertise in exascale computing to discuss novel ways to make applications resilient against detected and undetected faults. In particular, participants explored the role that algorithms and applications play in the holistic approach needed to tackle this challenge., Comment: 45 pages, 3 figures, submitted to The International Journal of High Performance Computing Applications

In the last 2 decades, a great amount of work has been done on data mining and knowledge discovery using complex networks. These works have provided insightful information about the structure and evolution of scientific activity, as well as important biomedical discoveries. However, interdisciplinary knowledge discovery, including disciplines other than science, is more complicated to implement because most of the available knowledge is not indexed. Here, a new method is presented for mining Wikipedia to unveil implicit interdisciplinary knowledge to map and understand how different disciplines (art, science, literature) are related to and interact with each other. Furthermore, the formalism of complex networks allows us to characterise both individual and collective behaviour of the different elements (people, ideas, works) within each discipline and among them. The results obtained agree with well-established interdisciplinary knowledge and show the ability of this method to boost quantitative studies. Note that relevant elements in different disciplines that rarely directly refer to each other may nonetheless have many implicit connections that impart them and their relationship with new meaning. Owing to the large number of available works and to the absence of cross-references among different disciplines, tracking these connections can be challenging. This approach aims to bridge this gap between the large amount of reported knowledge and the limited human capacity to find subtle connections and make sense of them., The author acknowledges the financial support from the Spanish Government ‘Ministerio de Ciencia e Innovación’ (PID2019-104650GB-C21) and from the Donostia International Physics Center (Programa Mestizajes), as well as the support of NVIDIA Corporation with the donation of a Quadro RTX 6000 GPU used for this research.

© 2021 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/), Background: Studies have previously shown evidence for presymptomatic cortical atrophy in genetic FTD. Whilst initial investigations have also identified early deep grey matter volume loss, little is known about the extent of subcortical involvement, particularly within subregions, and how this differs between genetic groups. Methods: 480 mutation carriers from the Genetic FTD Initiative (GENFI) were included (198 GRN, 202 C9orf72, 80 MAPT), together with 298 non-carrier cognitively normal controls. Cortical and subcortical volumes of interest were generated using automated parcellation methods on volumetric 3 T T1-weighted MRI scans. Mutation carriers were divided into three disease stages based on their global CDR® plus NACC FTLD score: asymptomatic (0), possibly or mildly symptomatic (0.5) and fully symptomatic (1 or more). Results: In all three groups, subcortical involvement was seen at the CDR 0.5 stage prior to phenoconversion, whereas in the C9orf72 and MAPT mutation carriers there was also involvement at the CDR 0 stage. In the C9orf72 expansion carriers the earliest volume changes were in thalamic subnuclei (particularly pulvinar and lateral geniculate, 9-10%) cerebellum (lobules VIIa-Crus II and VIIIb, 2-3%), hippocampus (particularly presubiculum and CA1, 2-3%), amygdala (all subregions, 2-6%) and hypothalamus (superior tuberal region, 1%). In MAPT mutation carriers changes were seen at CDR 0 in the hippocampus (subiculum, presubiculum and tail, 3-4%) and amygdala (accessory basal and superficial nuclei, 2-4%). GRN mutation carriers showed subcortical differences at CDR 0.5 in the presubiculum of the hippocampus (8%). Conclusions: C9orf72 expansion carriers show the earliest and most widespread changes including the thalamus, basal ganglia and medial temporal lobe. By investigating individual subregions, changes can also be seen at CDR 0 in MAPT mutation carriers within the limbic system. Our results suggest that subcortical brain volumes may be used as markers of neurodegeneration even prior to the onset of prodromal symptoms., This work was also supported by the MRC UK GENFI grant (MR/M023664/1), the Italian Ministry of Health (CoEN015 and Ricerca Corrente), the Canadian Institutes of Health Research as part of a Centres of Excellence in Neurodegeneration grant, a Canadian Institutes of Health Research operating grant, the Alzheimer's Society grant (AS-PG-16-007), the Bluefield Project and the JPND GENFI-PROX grant (2019-02248). MB is supported by a Fellowship award from the Alzheimer’s Society, UK (AS-JF-19a-004-517). MB’s work was also supported by the UK Dementia Research Institute which receives its funding from DRI Ltd, funded by the UK Medical Research Council, Alzheimer’s Society and Alzheimer’s Research UK. MB acknowledges the support of NVIDIA Corporation with the donation of the Titan V GPU used for part of the analyses in this research. JDR is an MRC Clinician Scientist (MR/M008525/1) and has received funding from the NIHR Rare Diseases Translational Research Collaboration (BRC149/NS/MH), the Bluefield Project and the Association for Frontotemporal Degeneration. JEI is supported by the European Research Council (Starting Grant 677697, project BUNGEE-TOOLS), Alzheimer’s Research UK (ARUK-IRG2019A003) and NIH 1RF1MH123195-01. JBR is funded by the Wellcome Trust (103838) and the National Institute for Health Research Cambridge Biomedical Research Centre. This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy within the framework of the Munich Cluster for Systems Neurology (EXC 2145 SyNergy – ID 390857198). Several authors of this publication (JCvS, MS, RSV, AD, MO, JDR) are members of the European Reference Network for Rare Neurological Diseases (ERN-RND) - Project ID No 739510.

Trabajo presentado en el ICPR International Workshops and Challenges, celebrado de forma virtual del 10 al 15 de enero de 2021, The recognition of human activities captured by a wearable photo-camera is especially suited for understanding the behavior of a person. However, it has received comparatively little attention with respect to activity recognition from fixed cameras.In this work, we propose to use segmented events from photo-streams as temporal boundaries to improve the performance of activity recognition. Furthermore, we robustly measure its effectiveness when images of the evaluated person have been seen during training, and when the person is completely unknown during testing. Experimental results show that leveraging temporal boundary information on pictures of seen people improves all classification metrics, particularly it improves the classification accuracy up to 85.73%., Lecture Notes in Computer Science 12663, This work was partially funded by projects RTI2018-095232-B-C2, SGR 1742, CERCA, Nestore Horizon2020 SC1-PM-15-2017 (n 769643), Validithi EIT Health Program, and the Spanish Ministry of Economy and Competitiveness and the European Regional Development Fund (MINECO/ERDF, EU) through the program Ramon y Cajal, the national Spanish project PID2019-110977GA-I00 and the Spanish national network RED2018-102511-T. A. Cartas supported by a doctoral fellowship from the Mexican Council of Science and Technology (CONACYT) (grant-no. 366596). The authors acknowledge the support of NVIDIA Corporation for hardware donation.

International audience; We present the design and results of recent competitions in Automated Deep Learning (AutoDL). In the AutoDL challenge series 2019, we organized 5 machine learning challenges: AutoCV, AutoCV2, AutoNLP, AutoSpeech and AutoDL. The first 4 challenges concern each a specific application domain, such as computer vision, natural language processing and speech recognition. At the time of March 2020, the last challenge AutoDL is still ongoing and we only present its design. 1 Some highlights of this work include: (1) a benchmark suite of baseline AutoML solutions, with emphasis on domains for which Deep Learning methods have had prior success (image, video, text, speech, etc); (2) a novel "anytime learning" framework, which opens doors for further theoretical consideration; (3) a repository of around 100 datasets (from all above domains) over half of which are released as public datasets to enable research on meta-learning; (4) analyses revealing that winning solutions generalize to new unseen datasets, validating progress towards universal AutoML 1. Its results will be presented in future work together with detailed introduction of winning solutions of each challenge.

Automatic speaker verification (ASV) is one of the most natural and convenient means of biometric person recognition. Unfortunately, just like all other biometric systems, ASV is vulnerable to spoofing, also referred to as "presentation attacks." These vulnerabilities are generally unacceptable and call for spoofing countermeasures or "presentation attack detection" systems. In addition to impersonation, ASV systems are vulnerable to replay, speech synthesis, and voice conversion attacks. The ASVspoof 2019 edition is the first to consider all three spoofing attack types within a single challenge. While they originate from the same source database and same underlying protocol, they are explored in two specific use case scenarios. Spoofing attacks within a logical access (LA) scenario are generated with the latest speech synthesis and voice conversion technologies, including state-of-the-art neural acoustic and waveform model techniques. Replay spoofing attacks within a physical access (PA) scenario are generated through carefully controlled simulations that support much more revealing analysis than possible previously. Also new to the 2019 edition is the use of the tandem detection cost function metric, which reflects the impact of spoofing and countermeasures on the reliability of a fixed ASV system. This paper describes the database design, protocol, spoofing attack implementations, and baseline ASV and countermeasure results. It also describes a human assessment on spoofed data in logical access. It was demonstrated that the spoofing data in the ASVspoof 2019 database have varied degrees of perceived quality and similarity to the target speakers, including spoofed data that cannot be differentiated from bona-fide utterances even by human subjects., Accepted, Computer Speech and Language. This manuscript version is made available under the CC-BY-NC-ND 4.0. For the published version on Elsevier website, please visit https://doi.org/10.1016/j.csl.2020.101114

[EN] Whole brain segmentation of fine-grained structures using deep learning (DL) is a very challenging task since the number of anatomical labels is very high compared to the number of available training images. To address this problem, previous DL methods proposed to use a single convolution neural network (CNN) or few independent CNNs. In this paper, we present a novel ensemble method based on a large number of CNNs processing different overlapping brain areas. Inspired by parliamentary decision-making systems, we propose a framework called AssemblyNet, made of two "assemblies" of U-Nets. Such a parliamentary system is capable of dealing with complex decisions, unseen problem and reaching a relevant consensus. AssemblyNet introduces sharing of knowledge among neighboring U-Nets, an "amendment" procedure made by the second assembly at higher-resolution to refine the decision taken by the first one, and a final decision obtained by majority voting. During our validation, AssemblyNet showed competitive performance compared to state-of-the-art methods such as U-Net, Joint label fusion and SLANT. Moreover, we investigated the scan-rescan consistency and the robustness to disease effects of our method. These experiences demonstrated the reliability of AssemblyNet. Finally, we showed the interest of using semi-supervised learning to improve the performance of our method., This work benefited from the support of the project DeepvolBrain of the French National Research Agency (ANR-18-CE45-0013). This study was achieved within the context of the Laboratory of Excellence TRAIL ANR-10-LABX-57 for the BigDataBrain project. Moreover, we thank the Investments for the future Program IdEx Bordeaux (ANR-10-IDEX-0 3-02, HL-MRI Project), Cluster of excellence CPU and the CNRS/INSERM for the DeepMultiBrain project. This study has been also supported by the DPI2017-87743-R grant from the Spanish Ministerio de Economia, Industria Competitividad. The C-MIND data used in the preparation of this article were obtained from the C-MIND Data Repository (accessed in Feb 2015) created by the C-MIND study of Normal Brain Development. This is a multisite, longitudinal study of typically developing children from ages newborn through young adulthood conducted by Cincinnati Children's Hospital Medical Center and UCLA and supported by the National Institute of Child Health and Human Development (Contract #s HHSN275200900018C). The NDAR data used in the preparation of this manuscript were obtained from the NIH-supported National Database for Autism Research (NDAR). This is a multisite, longitudinal study of typically developing children from ages newborn through young adulthood conducted by the Brain Development Cooperative Group and supported by the National Institute of Child Health and Human Development, the National Institute on Drug Abuse, the National Institute of Mental Health, and the National Institute of Neurological Disorders and Stroke (Contract #s N01- HD02-3343, N01-MH9-0002, and N01-NS-9-2314, -2315, -2316, -2317, -2319 and -2320). A listing of the participating sites and a complete listing of the study investigators can be found at http://pediatricmri.nih.gov/nihpd/info/participating_centers.html. The ADNI data used in the preparation of this manuscript were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904). ADNI data are disseminated by the Laboratory for NeuroImaging at the University of California, Los Angeles. This research was also supported by NIH grants P30AG010129, K01 AG030514 and the Dana Foundation. The OASIS data used in the preparation of this manuscript were obtained from the OASIS project funded by grants P50 AG05681, P01 AG03991, R01 AG021910, P50 MH071616, U24 RR021382, R01 MH56584. The AIBL data used in the preparation of this manuscript were obtained from the AIBL study of ageing funded by the Common-wealth Scientific Industrial Research Organization (CSIRO; a publicly funded government research organization), Science Industry Endowment Fund, National Health and Medical Research Council of Australia (project grant 1011689), Alzheimer's Association, Alzheimer's Drug Discovery Foundation, and an anonymous foundation. The ICBM data used in the preparation of this manuscript were supported by Human Brain Project grant PO1MHO52176-11 (ICBM, P.I. Dr John Mazziotta) and Canadian Institutes of Health Research grant MOP-34996. The IXI data used in the preparation of this manuscript were supported by the U.K. Engineering and Physical Sciences Research Council (EPSRC) GR/S21533/02. ABIDE primary support for the work by Adriana Di Martino was provided by the NIMH (K23MH087770) and the Leon Levy Foundation. Primary support for the work by Michael P. Milham and the INDI team was provided by gifts from Joseph P. Healy and the Stavros Niarchos Foundation to the Child Mind Institute, as well as by an NIMH award to MPM (R03MH096321).

24 pags., 7 figs., 3 tabs., New treatments for diseases caused by antimicrobial-resistant microorganisms can be developed by identifying unexplored therapeutic targets and by designing efficient drug screening protocols. In this study, we have screened a library of compounds to find ligands for the flavin-adenine dinucleotide synthase (FADS) -a potential target for drug design against tuberculosis and pneumonia- by implementing a new and efficient virtual screening protocol. The protocol has been developed for the in silico search of ligands of unexplored therapeutic targets, for which limited information about ligands or ligand-receptor structures is available. It implements an integrative funnel-like strategy with filtering layers that increase in computational accuracy. The protocol starts with a pharmacophore-based virtual screening strategy that uses ligand-free receptor conformations from molecular dynamics (MD) simulations. Then, it performs a molecular docking stage using several docking programs and an exponential consensus ranking strategy. The last filter, samples the conformations of compounds bound to the target using MD simulations. The MD conformations are scored using several traditional scoring functions in combination with a newly-proposed score that takes into account the fluctuations of the molecule with a Morse-based potential. The protocol was optimized and validated using a compound library with known ligands of the Corynebacterium ammoniagenes FADS. Then, it was used to find new FADS ligands from a compound library of 14,000 molecules. A small set of 17 in silico filtered molecules were tested experimentally. We identified five inhibitors of the activity of the flavin adenylyl transferase module of the FADS, and some of them were able to inhibit growth of three bacterial species: C. ammoniagenes, Mycobacterium tuberculosis, and Streptococcus pneumoniae, where the last two are human pathogens. Overall, the results show that the integrative VS protocol is a cost-effective solution for the discovery of ligands of unexplored therapeutic targets., The authors would like to acknowledge the use of Servicios Generales de Apoyo a la Investigacion-SAI, Universidad de Zaragoza. Some computations were performed in a local server with an NVIDIA Titan X GPU. P.C. gratefully acknowledges the support of NVIDIA Corporation for the donation of this GPU.

[EN] Affine registration of one or several brain image(s) onto a common reference space is a necessary prerequisite for many image processing tasks, such as brain segmentation or functional analysis. Manual assessment of registration quality is a tedious and time-consuming task, especially in studies comprising a large amount of data. Automated and reliable quality control (QC) becomes mandatory. Moreover, the computation time of the QC must be also compatible with the processing of massive datasets. Therefore, automated deep neural network approaches have emerged as a method of choice to automatically assess registration quality. In the current study, a compact 3D convolutional neural network, referred to as RegQCNET, is introduced to quantitatively predict the amplitude of an affine registration mismatch between a registered image and a reference template. This quantitative estimation of registration error is expressed using the metric unit system. Therefore, a meaningful task-specific threshold can be manually or automatically defined in order to distinguish between usable and non-usable images. The robustness of the proposed RegQCNET is first analyzed on lifespan brain images undergoing various simulated spatial transformations and intensity variations between training and testing. Secondly, the potential of RegQCNET to classify images as usable or non-usable is evaluated using both manual and automatic thresholds. During our experiments, automatic thresholds are estimated using several computer-assisted classification models (logistic regression, support vector machine, Naive Bayes and random forest) through cross-validation. To this end we use an expert's visual QC estimated on a lifespan cohort of 3953 brains. Finally, the RegQCNET accuracy is compared to usual image features such as image correlation coefficient and mutual information. The results show that the proposed deep learning QC is robust, fast and accurate at estimating affine registration error in the processing pipeline., The experiments presented in this paper were carried out using the PlaFRIM experimental testbed, supported by Inria, CNRS (LABRI and IMB), Universite de Bordeaux, Bordeaux INP and Conseil Regional d'Aquitaine (see https://www.plafrim.fr/). This work benefited from the support of the project DeepvolBrain of the French National Research Agency (ANR-18-CE45-0013). This study was achieved within the context of the Laboratory of Excellence TRAIL ANR-10-LABX-57 for the BigDataBrain project. Moreover, we thank the Investments for the Future Program IdEx Bordeaux (ANR-10-IDEX-03-02, HL-MRI Project), the Cluster of Excellence CPU and the CNRS/INSERM for the DeepMultiBrain project. This study has been also supported by a DPI2017-87743-R grant from the Spanish Ministerio de Economia, Industria Competitividad. The authors gratefully acknowledge the support of NVIDIA Corporation with their donation of a TITAN X GPU used in this research.

Breast cancer is the most frequent cancer in women and the second most frequent overall after lung cancer. Although the 5-year survival rate of breast cancer is relatively high, recurrence is also common which often involves metastasis with its consequent threat for patients. DNA methylation-derived databases have become an interesting primary source for supervised knowledge extraction regarding breast cancer. Unfortunately, the study of DNA methylation involves the processing of hundreds of thousands of features for every patient. DNA methylation is featured by High Dimension Low Sample Size which has shown well-known issues regarding feature selection and generation. Autoencoders (AEs) appear as a specific technique for conducting nonlinear feature fusion. Our main objective in this work is to design a procedure to summarize DNA methylation by taking advantage of AEs. Our proposal is able to generate new features from the values of CpG sites of patients with and without recurrence. Then, a limited set of relevant genes to characterize breast cancer recurrence is proposed by the application of survival analysis and a pondered ranking of genes according to the distribution of their CpG sites. To test our proposal we have selected a dataset from The Cancer Genome Atlas data portal and an AE with a single-hidden layer. The literature and enrichment analysis (based on genomic context and functional annotation) conducted regarding the genes obtained with our experiment confirmed that all of these genes were related to breast cancer recurrence., This work has been supported by the Spanish Ministry of Economy and Competitiveness under projects TIN2014-55894-C2-R and TIN2017-88209-C2-2-R. J.M. Luna-Romera holds a FPI scholarship from the Spanish Ministry of Economy and Competitiveness. We also gratefully acknowledge the support of the NVIDIA Corporation with the donation of the Titan Xp and RTX 2080 Ti GPUs used in this research.

A recurring problem faced when training neural networks is that there is typically not enough data to maximize the generalization capability of deep neural networks. There are many techniques to address this, including data augmentation, dropout, and transfer learning. In this paper, we introduce an additional method, which we call smart augmentation and we show how to use it to increase the accuracy and reduce over fitting on a target network. Smart augmentation works, by creating a network that learns how to generate augmented data during the training process of a target network in a way that reduces that networks loss. This allows us to learn augmentations that minimize the error of that network. Smart augmentation has shown the potential to increase accuracy by demonstrably significant measures on all data sets tested. In addition, it has shown potential to achieve similar or improved performance levels with significantly smaller network sizes in a number of tested cases. This research is funded under the SFI Strategic Partnership Program by Science Foundation Ireland (SFI) and FotoNation Ltd. Project ID: 13/SPP/I2868 on Next Generation Imaging for Smartphone and Embedded Platforms. This work is also supported by an Irish Research Council Employment Based Programme Award. The authors gratefully acknowledge the support of NVIDIA Corporation with the donation of a Titan X GPU used for this research. peer-reviewed

Trabajo presentado en la International Conference on Computer Vision Workshop (ICCVW), celebrada en Seúl (Corea del Sur), los días 27 y 28 de octubre de 2019, Our interaction with the world is an inherently multi-modal experience. However, the understanding of human-to-object interactions has historically been addressed focusing on a single modality. In particular, a limited number of works have considered to integrate the visual and audio modalities for this purpose. In this work, we propose a multimodal approach for egocentric action recognition in a kitchen environment that relies on audio and visual information. Our model combines a sparse temporal sampling strategy with a late fusion of audio, spatial,and temporal streams. Experimental results on the EPIC-Kitchens dataset show that multimodal integration leads to better performance than unimodal approaches. In particular, we achieved a5.18%improvement over the state of the art on verb classification., This work was partially funded by TIN2018-095232-B-C21, 2017 SGR 1742, Nestore, Validithi, 20141510 (La MaratoTV3) and CERCA Programme/Generalitat de Catalunya. A.C. is supported by a doctoral fellowship from the Mexican Council of Science and Technology (CONACYT) (grant-no. 366596). We acknowledge the support of NVIDIA Corporation with the donation of Titan Xp GPUs

Unsupervised image-to-image translation methods learn to map images in a given class to an analogous image in a different class, drawing on unstructured (non-registered) datasets of images. While remarkably successful, current methods require access to many images in both source and destination classes at training time. We argue this greatly limits their use. Drawing inspiration from the human capability of picking up the essence of a novel object from a small number of examples and generalizing from there, we seek a few-shot, unsupervised image-to-image translation algorithm that works on previously unseen target classes that are specified, at test time, only by a few example images. Our model achieves this few-shot generation capability by coupling an adversarial training scheme with a novel network design. Through extensive experimental validation and comparisons to several baseline methods on benchmark datasets, we verify the effectiveness of the proposed framework. Our implementation and datasets are available at https://github.com/NVlabs/FUNIT ., The paper will be presented at the International Conference on Computer Vision (ICCV) 2019

We present a novel rotation invariant architecture operating directly on point cloud data. We demonstrate how rotation invariance can be injected into a recently proposed point-based PCNN architecture, on all layers of the network. This leads to invariance to both global shape transformations, and to local rotations on the level of patches or parts, useful when dealing with non-rigid objects. We achieve this by employing a spherical harmonics-based kernel at different layers of the network, which is guaranteed to be invariant to rigid motions. We also introduce a more efficient pooling operation for PCNN using space-partitioning data-structures. This results in a flexible, simple and efficient architecture that achieves accurate results on challenging shape analysis tasks, including classification and segmentation, without requiring data-augmentation typically employed by non-invariant approaches. Code and data are provided on the project page https://github.com/adrienPoulenard/SPHnet.