检索结果-内蒙古大学图书馆

Supervised speech separation based on deep learning: An overview

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Wang, DeLiang Chen, Jitong Department of Computer Science and Engineering Center for Cognitive and Brain Sciences Ohio State University ColumbusOH43210 United States Center of Intelligent Acoustics and Immersive Communications Northwestern Polytechnical University Xi’an China Department of Computer Science and Engineering Ohio State University ColumbusOH43210 United States Silicon Valley AI Lab at Baidu Research 1195 Bordeaux Drive SunnyvaleCA94089 United States

Speech separation is the task of separating target speech from background interference. Traditionally, speech separation is studied as a signal processing problem. A more recent approach formulates speech separation as a supervised learning problem, where the discriminative patterns of speech, speakers, and background noise are learned from training data. Over the past decade, many supervised separation algorithms have been put forward. In particular, the recent introduction of deep learning to supervised speech separation has dramatically accelerated progress and boosted separation performance. This article provides a comprehensive overview of the research on deep learning based supervised speech separation in the last several years. We first introduce the background of speech separation and the formulation of supervised separation. Then we discuss three main components of supervised separation: learning machines, training targets, and acoustic features. Much of the overview is on separation algorithms where we review monaural methods, including speech enhancement (speech-nonspeech separation), speaker separation (multi-talker separation), and speech dereverberation, as well as multi-microphone techniques. The important issue of generalization, unique to supervised learning, is discussed. This overview provides a historical perspective on how advances are made. In addition, we discuss a number of conceptual issues, including what constitutes the target source. Copyright © 2017, The Authors. All rights reserved.

关键词： Separation

Deep learning for automatic stereotypical motor movement detection using wearable sensors in autism spectrum disorders

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Rad, Nastaran Mohammadian Kia, Seyed Mostafa Zarbo, Calogero van Laarhoven, Twan Jurman, Giuseppe Venuti, Paola Marchiori, Elena Furlanello, Cesare Fondazione Bruno Kessler Via Sommarive 18 Povo Trento38123 Italy Fondazione Bruno Kessler Trento Italy Institute for Computing and Information Sciences Radboud University Nijmegen Netherlands Department of Information Engineering and Computer Science University of Trento Trento Italy Donders Centre for Cognitive Neuroimaging Donders Institute for Brain Cognition and Behaviour Radboud University Nijmegen Netherlands Department of Cognitive Neuroscience Radboud University Medical Centre Nijmegen Netherlands Department of Psychology and Cognitive Science University of Trento Trento Italy

Autism Spectrum Disorders are associated with atypical movements, of which stereotypical motor movements (SMMs) interfere with learning and social interaction. The automatic SMM detection using inertial measurement units (IMU) remains complex due to the strong intra and inter-subject variability, especially when handcrafted features are extracted from the signal. We propose a new application of the deep learning to facilitate automatic SMM detection using multi-axis IMUs. We use a convolutional neural network (CNN) to learn a discriminative feature space from raw data. We show how the CNN can be used for parameter transfer learning to enhance the detection rate on longitudinal data. We also combine the long short-term memory (LSTM) with CNN to model the temporal patterns in a sequence of multi-axis signals. Further, we employ ensemble learning to combine multiple LSTM learners into a more robust SMM detector. Our results show that: 1) feature learning outperforms handcrafted features;2) parameter transfer learning is beneficial in longitudinal settings;3) using LSTM to learn the temporal dynamic of signals enhances the detection rate especially for skewed training data;4) an ensemble of LSTMs provides more accurate and stable detectors. These findings provide a significant step toward accurate SMM detection in real-time scenarios. Copyright © 2017, The Authors. All rights reserved.

关键词： Long short-term memory

Learning from label proportions in brain-computer interfaces: Online unsupervised learning with guarantees

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Hübner, David Verhoeven, Thibault Schmid, Konstantin Müller, Klaus-Robert Tangermann, Michael Kindermans, Pieter-Jan Brain State Decoding Lab Cluster of Excellence BrainLinks-BrainTools Department of Computer Science Albert-Ludwigs-University Freiburg Germany Machine Learning Group Berlin Institute of Technology Berlin Germany Electronics and Information Systems Ghent University Ghent Belgium Department of Brain and Cognitive Engineering Korea University Seoul Korea Republic of

Objective: Using traditional approaches, a brain-computer Interface (BCI) requires the collection of calibration data for new subjects prior to online use. Calibration time can be reduced or eliminated e.g. by subject-to-subject transfer of a pre-trained classifier or unsupervised adaptive classification methods which learn from scratch and adapt over time. While such heuristics work well in practice, none of them can provide theoretical guarantees. Our objective is to modify an event-related potential (ERP) paradigm to work in unison with the machine learning decoder, and thus to achieve a reliable unsupervised calibration-less decoding with a guarantee to recover the true class means. Copyright © 2017, The Authors. All rights reserved.

关键词： brain computer interface

The application of retinal fundus camera imaging in dementia: A systematic review

学校读者我要写书评

暂无评论

Alzheimer's and Dementia: Diagnosis, Assessment and Disease Monitoring 2017年第0期6卷 91-107页

作者： McGrory, Sarah Cameron, James R. Pellegrini, Enrico Warren, Claire Doubal, Fergus N. Deary, Ian J. Dhillon, Baljean Wardlaw, Joanna M. Trucco, Emanuele MacGillivray, Thomas J. Centre for Clinical Brain Sciences College of Medicine and Veterinary Medicine University of Edinburgh Edinburgh United Kingdom Anne Rowling Regenerative Neurology Clinic College of Medicine and Veterinary Medicine University of Edinburgh Edinburgh United Kingdom College of Medicine and Veterinary Medicine University of Edinburgh Edinburgh United Kingdom Department of Psychology University of Edinburgh Edinburgh United Kingdom Department of Psychology Centre for Cognitive Ageing and Cognitive Epidemiology University of Edinburgh Edinburgh United Kingdom Scottish Imaging Network: A Platform for Scientific Excellence (SINAPSE) Collaboration Edinburgh United Kingdom VAMPIRE Project and Computer Vision and Image Processing Group School of Science and Engineering (Computing) University of Dundee Dundee United Kingdom VAMPIRE Project and Edinburgh Clinical Research Facility University of Edinburgh Edinburgh United Kingdom

Introduction The ease of imaging the retinal vasculature, and the evolving evidence suggesting this microvascular bed might reflect the cerebral microvasculature, presents an opportunity to investigate cerebrovascular disease and the contribution of microvascular disease to dementia with fundus camera imaging. Methods A systematic review and meta-analysis was carried out to assess the measurement of retinal properties in dementia using fundus imaging. Results Ten studies assessing retinal properties in dementia were included. Quantitative measurement revealed significant yet inconsistent pathologic changes in vessel caliber, tortuosity, and fractal dimension. Retinopathy was more prevalent in dementia. No association of age-related macular degeneration with dementia was reported. Discussion Inconsistent findings across studies provide tentative support for the application of fundus camera imaging as a means of identifying changes associated with dementia. The potential of fundus image analysis in differentiating between dementia subtypes should be investigated using larger well-characterized samples. Future work should focus on refining and standardizing methods and measurements. © 2016 The Authors

关键词： Alzheimer's disease Dementia Fundus Retinal imaging Retinal vasculature

Heterogeneous defect prediction via exploiting correlation subspace 28

学校读者我要写书评

暂无评论

Heterogeneous defect prediction via exploiting correlation s...

28th International Conference on Software engineering and Knowledge engineering, SEKE 2016

作者： Cheng, Ming Wu, Guoqing Jiang, Min Wan, Hongyan You, Guoan Yuan, Mengting State Key Laboratory of Software Engineering School of Computer Wuhan University Wuhan430072 China Department of Cognitive Science and Technology Fujian Key Laboratory of Brain-Like Intelligent Systems Xiamen University Xiamen361005 China

ISBN: (纸本)189170639X

Software defect prediction generally builds models from intra-project data. Lack of training data at the early stage of software testing limits the efficiency of prediction in practice. Thereby researchers proposed cross-project defect prediction using the data from other projects. Most previous efforts assumed the cross-project defect data have the same metrics set which means the metrics used and size of metrics set are same in the data of projects. However, in real scenarios, this assumption may not hold. In addition, software defect datasets have the class imbalance problem increasing the difficulty for the learner to predict defects. In this paper, we advance canonical correlation analysis for deriving a joint feature space for associating crossproject data and propose a novel support vector machine algorithm which incorporates the correlation transfer information into classifier design for cross-project prediction. Moreover, we take different misclassification costs into consideration to make the classification inclining to classify a module as a defective one, alleviating the impact of imbalanced data. Experiments on public heterogeneous datasets from different projects show that our method is more effective, compared to state-of-the-art methods. Copyright © 2016 by KSI Research Inc. and Knowledge Systems Institute Graduate School.

关键词： Support vector machines

Penetrance of neurodevelopmental copy number variants is associated with variations in cortical morphology

学校读者我要写书评

暂无评论

Biological psychiatry. cognitive neuroscience and neuroimaging 2025年

作者： Ana I Silva Ida E Sønderby George Kirov Abdel Abdellaoui Ingrid Agartz David Ames Nicola J Armstrong Eric Artiges Tobias Banaschewski Anne S Bassett Carrie E Bearden John Blangero Rune Boen Dorret I Boomsma Robin Bülow Nancy J Butcher Vince Calhoun Linda E Campbell Eva W C Chow Simone Ciufolini Michael C Craig Benedicto Crespo-Farroco Adam C Cunningham Shareefa Dalvie Eileen Daly Paola Dazzan Eco J C de Geus Greig I de Zubicaray Joanne L Doherty Gary Donohoe Mark Drakesmith Thomas Espeseth Vincent Frouin Hugh Garavan David C Glahn Naomi J Goodrich-Hunsaker Penny A Gowland Hans J Grabe Antoine Grigis Maria Gudbrandsen Boris A Gutman Jan Haavik Asta K Håberg Jeremy Hall Andreas Heinz Sarah Hohmann Jouke-Jan Hottenga Sébastien Jacquemont Neda Jahanshad Rachel K Jonas Derek K Jones Erik G Jönsson Sanne Koops Kuldeep Kumar Stephanie Le Hellard Herve Lemaitre Jingyu Liu Astri J Lundervold Jean-Luc Martinot Karen A Mather Donna M McDonald-McGinn Katie L McMahon Allan F McRae Sarah E Medland Clara A Moreau Kieran C Murphy Declan Murphy Robin M Murray Frauke Nees Michael J Owen Marie-Laure Paillère Martinot Diimitri Papadopoulos Orfanos Tomas Paus Luise Poustka Tiago Reis Marques David R Roalf Perminder S Sachdev Freda Scheffler J Eric Schmitt Gunter Schumann Vidar M Steen Dan J Stein Lachlan T Strike Alexander Teumer Anbupalam Thalamuthu Sophia I Thomopoulos Diana Tordesillas-Gutiérrez Julian N Trollor Anne Uhlmann Ariana Vajdi Dennis van 't Ent Therese van Amelsvoort Marianne B M van den Bree Dennis van der Meer Javier Vázquez-Bourgon Julio E Villalón-Reina Uwe Völker Henry Völzke Jacob A S Vorstman Lars T Westlye Nigel Williams Katharina Wittfeld Margaret J Wright Paul M Thompson Ole A Andreassen David E J Linden School for Mental Health and Neuroscience Department of Psychiatry and Neuropsychology Faculty of Health Medicine and Life Sciences Maastricht University Maastricht The Netherlands Neuroscience and Mental Health Innovation Institute Cardiff University Cardiff United Kingdom Center for Magnetic Resonance Research Department of Radiology University of Minnesota Minneapolis MN USA. Electronic address: silvaai@um.edu. Centre for Precision Psychiatry Institute of Clinical Medicine University of Oslo Oslo Norway Department of Medical Genetics Oslo University Hospital Oslo Norway KG Jebsen Centre for Neurodevelopmental Disorders University of Oslo Oslo Norway. Division of Psychological Medicine and Clinical Neurosciences Cardiff University Cardiff United Kingdom Centre for Neuropsychiatric Genetics and Genomics School of Medicine Cardiff University Cardiff United Kingdom. Department of Psychiatry Amsterdam UMC University of Amsterdam Amsterdam the Netherlands. Norwegian Centre for Mental Disorders Research (NORMENT) Institute of Clinical Medicine University of Oslo Oslo Norway Centre for Psychiatry Research Department of Clinical Neuroscience Karolinska Institutet & Stockholm Health Care Services Stockholm Region Stockholm Sweden Department of Psychiatric Research Diakonhjemmet Hospital Oslo Norway. University of Melbourne Academic Unit for Psychiatry of Old Age Kew Victoria Australia. Mathematics and Statistics Curtin University Perth WA Australia. Institut National de la Santé et de la Recherche Médicale INSERM U1299 "Trajectoires développementales et psychiatrie" Université Paris-Saclay Ecole Normale supérieure Paris-Saclay CNRS Centre Borelli Gif-sur-Yvette France Psychiatry Department EPS Barthélémy Durand Etampes Île-de-France France. Department of Child and Adolescent Psychiatry and Psychotherapy Central Institute of Mental Health Medical Faculty Mannheim University of Heidelberg Mannheim Germany. Clinical Genetics Research Program Centre for

BACKGROUND:Copy number variants (CNVs) increase risk for neurodevelopmental conditions. The neurobiological mechanisms linking these high-risk genetic variants to clinical phenotypes are largely unknown. An important question is whether brain abnormalities in individuals carrying CNVs are associated with their degree of penetrance. METHODS:We investigated if increased CNV-penetrance for schizophrenia and other developmental disorders was associated with variations in cortical and subcortical morphology. We pooled T1-weighted brain magnetic resonance imaging and genetic data from 22 cohorts from the ENIGMA-CNV consortium. In the main analyses, we included 9,268 individuals (aged 7 to 90 years, 54% females), from which we identified 398 carriers of 36 neurodevelopmental CNVs at 20 distinct loci. A secondary analysis was performed including additional neuroimaging data from the ENIGMA-22q consortium, including 274 carriers of the 22q11.2 deletion and 291 non-carriers. CNV-penetrance was estimated through penetrance scores that were previously generated from large cohorts of patients and controls. These scores represent the probability risk to develop either schizophrenia or other developmental disorders (including developmental delay, autism spectrum disorder and congenital malformations). RESULTS:For both schizophrenia and developmental disorders, increased penetrance scores were associated with lower surface area in the cerebral cortex and lower intracranial volume. For both conditions, associations between CNV-penetrance scores and cortical surface area were strongest in regions of the occipital lobes, specifically in the cuneus and lingual gyrus. CONCLUSIONS:Our findings link global and regional cortical morphometric features with CNV-penetrance, providing new insights into neurobiological mechanisms of genetic risk for schizophrenia and other developmental disorders.

关键词： autism neuroimaging psychiatric disorders schizophrenia structural imaging

Author Correction: A multi-modal single-cell and spatial expression map of metastatic breast cancer biopsies across clinicopathological features

学校读者我要写书评

暂无评论

Nature medicine 2025年 2025 Mar 28页

作者： Johanna Klughammer Daniel L Abravanel Åsa Segerstolpe Timothy R Blosser Yury Goltsev Yi Cui Daniel R Goodwin Anubhav Sinha Orr Ashenberg Michal Slyper Sébastien Vigneau Judit Jané-Valbuena Shahar Alon Chiara Caraccio Judy Chen Ofir Cohen Nicole Cullen Laura K DelloStritto Danielle Dionne Janet Files Allison Frangieh Karla Helvie Melissa E Hughes Stephanie Inga Abhay Kanodia Ana Lako Colin MacKichan Simon Mages Noa Moriel Evan Murray Sara Napolitano Kyleen Nguyen Mor Nitzan Rebecca Ortiz Miraj Patel Kathleen L Pfaff Caroline B M Porter Asaf Rotem Sarah Strauss Robert Strasser Aaron R Thorner Madison Turner Isaac Wakiro Julia Waldman Jingyi Wu Jorge Gómez Tejeda Zañudo Diane Zhang Nancy U Lin Sara M Tolaney Eric P Winer Edward S Boyden Fei Chen Garry P Nolan Scott J Rodig Xiaowei Zhuang Orit Rozenblatt-Rosen Bruce E Johnson Aviv Regev Nikhil Wagle Klarman Cell Observatory Broad Institute of Harvard and MIT Cambridge MA USA. klughammer@genzentrum.lmu.de. Gene Center and Department of Biochemistry Ludwig Maximilians Universität München Munich Germany. klughammer@genzentrum.lmu.de. Department of Medical Oncology Dana-Farber Cancer Institute Boston MA USA. daniel_abravanel@dfci.harvard.edu. Harvard Medical School Boston MA USA. daniel_abravanel@dfci.harvard.edu. Klarman Cell Observatory Broad Institute of Harvard and MIT Cambridge MA USA. Department of Chemistry and Chemical Biology Harvard University Cambridge MA USA. Baxter Laboratory in Stem Cell Biology Department of Microbiology and Immunology Stanford University School of Medicine Stanford CA USA. Department of Media Arts and Sciences McGovern Institute Massachusetts Institute of Technology Cambridge MA USA. Center for Cancer Genomics Dana-Farber Cancer Institute Boston MA USA. Faculty of Engineering Gonda Brain Research Center and Institute of Nanotechnology Bar-Ilan University Ramat Gan Israel. Department of Medical Oncology Dana-Farber Cancer Institute Boston MA USA. Broad Institute of Harvard and MIT Cambridge MA USA. Department of Microbiology Immunology and Genetics Faculty of Health Sciences Ben-Guiron University Beersheba Israel. Center for Immuno-Oncology Dana-Farber Cancer Institute Boston MA USA. Gene Center and Department of Biochemistry Ludwig Maximilians Universität München Munich Germany. School of Computer Science and Engineering The Hebrew University of Jerusalem Jerusalem Israel. Racah Institute of Physics The Hebrew University of Jerusalem Jerusalem Israel. Faculty of Medicine The Hebrew University of Jerusalem Jerusalem Israel. AstraZeneca R&D Boston MA USA. Department of Microbiology Immunology and Cancer Biology University of Virginia Charlottesville VA USA. Department of Media Arts and Sciences Massachusetts Institute of Technology Cambridge MA USA. Department of Biological Engineering Massachusetts I

Explainable artificial intelligence: Understanding, visualizing and interpreting deep learning models

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Samek, Wojciech Wiegand, Thomas Müller, Klaus-Robert Dept. of Video Coding & Analytics Fraunhofer Heinrich Hertz Institute Berlin10587 Germany Dept. of Computer Science Technische Universität Berlin Berlin10587 Germany Dept. of Brain & Cognitive Engineering Korea University Seoul136-713 Korea Republic of Max Planck Institute for Informatics Saarbrücken66123 Germany

With the availability of large databases and recent improvements in deep learning methodology, the performance of AI systems is reaching or even exceeding the human level on an increasing number of complex tasks. Impressive examples of this development can be found in domains such as image classification, sentiment analysis, speech understanding or strategic game playing. However, because of their nested non-linear structure, these highly successful machine learning and artificial intelligence models are usually applied in a black box manner, i.e., no information is provided about what exactly makes them arrive at their predictions. Since this lack of transparency can be a major drawback, e.g., in medical applications, the development of methods for visualizing, explaining and interpreting deep learning models has recently attracted increasing attention. This paper summarizes recent developments in this field and makes a plea for more interpretability in artificial intelligence. Furthermore, it presents two approaches to explaining predictions of deep learning models, one method which computes the sensitivity of the prediction with respect to changes in the input and one approach which meaningfully decomposes the decision in terms of the input variables. These methods are evaluated on three classification tasks. Copyright © 2017, The Authors. All rights reserved.

关键词： Sensitivity analysis

Toward goal-driven neural network models for the rodent whisker-trigeminal system

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Zhuang, Chengxu Kubilius, Jonas Hartmann, Mitra Yamins, Daniel Department of Psychology Stanford University StanfordCA94305 United States Department of Brain and Cognitive Sciences Massachusetts Institute of Technology CambridgeMA02139 United States Departments of Biomedical Engineering and Mechanical Engineering Northwestern University EvanstonIL60208 United States Departments of Psychology and Computer Science Stanford Neurosciences Institute Stanford University StanfordCA94305 United States

In large part, rodents "see" the world through their whiskers, a powerful tactile sense enabled by a series of brain areas that form the whisker-trigeminal system. Raw sensory data arrives in the form of mechanical input to the exquisitely sensitive, actively-controllable whisker array, and is processed through a sequence of neural circuits, eventually arriving in cortical regions that communicate with decision-making and memory areas. Although a long history of experimental studies has characterized many aspects of these processing stages, the computational operations of the whisker-trigeminal system remain largely unknown. In the present work, we take a goal-driven deep neural network (DNN) approach to modeling these computations. First, we construct a biophysically-realistic model of the rat whisker array. We then generate a large dataset of whisker sweeps across a wide variety of 3D objects in highly-varying poses, angles, and speeds. Next, we train DNNs from several distinct architectural families to solve a shape recognition task in this dataset. Each architectural family represents a structurally-distinct hypothesis for processing in the whisker-trigeminal system, corresponding to different ways in which spatial and temporal information can be integrated. We find that most networks perform poorly on the challenging shape recognition task, but that specific architectures from several families can achieve reasonable performance levels. Finally, we show that Representational Dissimilarity Matrices (RDMs), a tool for comparing population codes between neural systems, can separate these higher-performing networks with data of a type that could plausibly be collected in a neurophysiological or imaging experiment. Our results are a proof-of-concept that DNN models of the whiskertrigeminal system are potentially within reach. Copyright © 2017, The Authors. All rights reserved.

关键词： Mammals