检索结果-内蒙古大学图书馆

A scalable generative model for dynamical system reconstruction from neuroimaging data 24

学校读者我要写书评

暂无评论

A scalable generative model for dynamical system reconstruct...

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Eric Volkmann Alena Brändle Daniel Durstewitz Georgia Koppe Department of Theoretical Neuroscience Central Institute of Mental Health (CIMH) Medical Faculty Mannheim Heidelberg University Mannheim Germany and Institute for Machine Learning Johannes Kepler University Linz Austria Department of Theoretical Neuroscience Central Institute of Mental Health (CIMH) Medical Faculty Mannheim Heidelberg University Mannheim Germany and Interdisciplinary Center for Scientific Computing Heidelberg University Heidelberg Germany and Faculty of Physics and Astronomy Heidelberg University Heidelberg Germany Interdisciplinary Center for Scientific Computing Heidelberg University Heidelberg Germany and Hector Institute for AI in Psychiatry & Dept. for Psychiatry and Psychotherapy CIMH and Faculty of Mathematics and Computer Science Heidelberg University Heidelberg Germany

ISBN: (纸本)9798331314385

Data-driven inference of the generative dynamics underlying a set of observed time series is of growing interest in machine learning and the natural sciences. In neuroscience, such methods promise to alleviate the need to handcraft models based on biophysical principles and allow to automatize the inference of inter-individual differences in brain dynamics. Recent breakthroughs in training techniques for state space models (SSMs) specifically geared toward dynamical systems (DS) reconstruction (DSR) enable to recover the underlying system including its geometrical (attractor) and long-term statistical invariants from even short time series. These techniques are based on control-theoretic ideas, like modern variants of teacher forcing (TF), to ensure stable loss gradient propagation while training. However, as it currently stands, these techniques are not directly applicable to data modalities where current observations depend on an entire history of previous states due to a signal's filtering properties, as common in neuroscience (and physiology more generally). Prominent examples are the blood oxygenation level dependent (BOLD) signal in functional magnetic resonance imaging (fMRI) or Ca2+ imaging data. Such types of signals render the SSM's decoder model non-invertible, a requirement for previous TF-based methods. Here, exploiting the recent success of control techniques for training SSMs, we propose a novel algorithm that solves this problem and scales exceptionally well with model dimensionality and filter length. We demonstrate its efficiency in reconstructing dynamical systems, including their state space geometry and long-term temporal properties, from just short BOLD time series.

关键词：

Deep inference of latent dynamics with spatio-temporal super-resolution using selective backpropagation through time

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhu, Feng Sedler, Andrew R. Grier, Harrison A. Ahad, Nauman Davenport, Mark A. Kaufman, Matthew T. Giovannucci, Andrea Pandarinath, Chethan Neuroscience Graduate Program Emory University Center for Machine Learning Georgia Tech Computational Neuroscience Graduate Program The University of Chicago School of Electrical and Computer Engineering Georgia Tech Dept. of Organismal Biology and Anatomy The University of Chicago Neuroscience Institute The University of Chicago Joint Dept. of Biomedical Engineering UNC Chapel Hill NC State University UNC Chapel Hill Neuroscience Center Coulter Dept. of Biomedical Engineering Emory University Georgia Tech Dept. of Neurosurgery Emory University

Modern neural interfaces allow access to the activity of up to a million neurons within brain circuits. However, bandwidth limits often create a trade-off between greater spatial sampling (more channels or pixels) and the temporal frequency of sampling. Here we demonstrate that it is possible to obtain spatio-temporal super-resolution in neuronal time series by exploiting relationships among neurons, embedded in latent low-dimensional population dynamics. Our novel neural network training strategy, selective backpropagation through time (SBTT), enables learning of deep generative models of latent dynamics from data in which the set of observed variables changes at each time step. The resulting models are able to infer activity for missing samples by combining observations with learned latent dynamics. We test SBTT applied to sequential autoencoders and demonstrate more efficient and higher-fidelity characterization of neural population dynamics in electrophysiological and calcium imaging data. In electrophysiology, SBTT enables accurate inference of neuronal population dynamics with lower interface bandwidths, providing an avenue to significant power savings for implanted neuroelectronic interfaces. In applications to two-photon calcium imaging, SBTT accurately uncovers high-frequency temporal structure underlying neural population activity, substantially outperforming the current state-of-the-art. Finally, we demonstrate that performance could be further improved by using limited, high-bandwidth sampling to pretrain dynamics models, and then using SBTT to adapt these models for sparsely-sampled data. © 2021, CC BY-NC-SA.

关键词： Neurons

Spectroscopic detection of pigments in tissues: correlation with tissue aging and cancer development

学校读者我要写书评

暂无评论

Spectroscopic detection of pigments in tissues: correlation ...

2022 International Conference Laser Optics, ICLO 2022

作者： Oliveira, L.M. Goncalves, T.M. Botelho, A.R. Martins, I.S. Silva, H.F. Carneiro, I. Carvalho, S. Henrique, R. Tuchin, V.V. Porto Portugal Center of Innovation in Engineering and Industrial Technology Isep Porto Portugal Department of Electrical and Computer Engineering Porto University-School of Engineering Porto Portugal Department of Pathology and Cancer Biology and Epigenetics Group Portuguese Oncology Institute of Porto Porto Portugal Porto Portugal Viana do Castelo Portugal Dept. of Pathol. and Molecular Immunology Porto University-Institute of Biomedical Sciences Abel Salazar Porto Portugal Science Medical Center Saratov State University Saratov Russia Laboratory of Laser Molecular Imaging and Machine Learning National Research Tomsk State University Tomsk Russia Laboratory of Laser Diagnostics of Technical and Living Systems Institute of Precision Mechanics and Control Frc-Saratov Scientific Centre-of the Russian Academy of Sciences Saratov Russia

ISBN: (纸本)9781665466646

The direct calculation of the absorption coefficient spectra of various tissues from spectral measurements allowed to retrieve the contents of melanin and lipofuscin. In the rabbit brain cortex, 1.8 times higher melanin content is explained by the neuron degeneration process. Similar melanin and lipofuscin contents were found in the rabbit pancreas as a result of the tissue aging process. The conversion of 83 % of the melanin in the human normal kidney into lipofuscin in the cancer kidney indicates that lipofuscin can be considered a kidney cancer marker in humans. © 2022 IEEE.

关键词： Tissue

Benefit-aware Early Prediction of Health Outcomes on Multivariate EEG Time Series

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Shekhar, Shubhranshu Eswaran, Dhivya Hooi, Bryan Elmer, Jonathan Faloutsos, Christos Akoglu, Leman Heinz College Machine Learning Dept. Carnegie Mellon University Department of Computer Science Carnegie Mellon University School of Computing National University of Singapore Departments of Emergency Medicine Critical Care Medicine and Neurology University of Pittsburgh Heinz College of Info. Sys. Carnegie Mellon University Amazon Inc.

Given a cardiac-arrest patient being monitored in the ICU (intensive care unit) for brain activity, how can we predict their health outcomes as early as possible? Early decision-making is critical in many applications, e.g. monitoring patients may assist in early intervention and improved care. On the other hand, early prediction on EEG data poses several challenges: (i) earliness-accuracy trade-off;observing more data often increases accuracy but sacrifices earliness, (ii) large-scale (for training) and streaming (online decision-making) data processing, and (iii) multi-variate (due to multiple electrodes) and multi-length (due to varying length of stay of patients) time series. Motivated by this real-world application, we present BeneFitter that infuses the incurred savings from an early prediction as well as the cost from misclassification into a unified domain-specific target called benefit. Unifying these two quantities allows us to directly estimate a single target (i.e. benefit), and importantly, dictates exactly when to output a prediction: when benefit estimate becomes positive. BeneFitter (a) is efficient and fast, with training time linear in the number of input sequences, and can operate in real-time for decision-making, (b) can handle multi-variate and variable-length time-series, suitable for patient data, and (c) is effective, providing up to 2× time-savings with equal or better accuracy as compared to competitors. Copyright © 2021, The Authors. All rights reserved.

关键词： Time series

Evaluating deep transfer learning for whole-brain cognitive decoding

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Thomas, Armin W. Lindenberger, Ulman Samek, Wojciech Müller, Klaus-Robert Machine Learning Group Dept. of Computer Science and Electrical Engineering Technische Universität Berlin Berlin Germany Center for Lifespan Psychology Max Planck Institute for Human Development Berlin Germany Stanford Data Science Stanford University StanfordCA United States Dept. of Psychology Stanford University StanfordCA United States Max Planck UCL Centre for Computational Psychiatry and Ageing Research Berlin Germany Dept. of Artificial Intelligence Fraunhofer Heinrich Hertz Institute Berlin Germany BIFOLD – Berlin Institute for the Foundations of Learning and Data Berlin Germany Dept. of Artificial Intelligence Korea University Seoul Korea Republic of Max Planck Institute for Informatics Saarbrücken Germany

Research in many fields has shown that transfer learning (TL) is well-suited to improve the performance of deep learning (DL) models in datasets with small numbers of samples. This empirical success has triggered interest in the application of TL to cognitive decoding analyses with functional neuroimaging data. Here, we systematically evaluate TL for the application of DL models to the decoding of cognitive states (e.g., viewing images of faces or houses) from whole-brain functional Magnetic Resonance Imaging (fMRI) data. We first pre-train two DL architectures on a large, public fMRI dataset and subsequently evaluate their performance in an independent experimental task and a fully independent dataset. The pre-trained models consistently achieve higher decoding accuracies and generally require less training time and data than model variants that were not pre-trained, clearly underlining the benefits of pre-training. We demonstrate that these benefits arise from the ability of the pre-trained models to reuse many of their learned features when training with new data, providing deeper insights into the mechanisms giving rise to the benefits of pre-training. Yet, we also surface nuanced challenges for whole-brain cognitive decoding with DL models when interpreting the decoding decisions of the pre-trained models, as these have learned to utilize the fMRI data in unforeseen and counterintuitive ways to identify individual cognitive states. © 2021, CC BY.

关键词： Decoding

The advantage of conditional meta-learning for biased regularization and fine-tuning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Denevi, Giulia Pontil, Massimiliano Ciliberto, Carlo Computational Statistics and Machine Learning Istituto Italiano di Tecnologia Genova Italy Computer Science Dept. University College of London London United Kingdom Electrical and Electronic Engineering Dept. Imperial College of London London United Kingdom

Biased regularization and fine-tuning are two recent meta-learning approaches. They have been shown to be effective to tackle distributions of tasks, in which the tasks’ target vectors are all close to a common meta-parameter vector. However, these methods may perform poorly on heterogeneous environments of tasks, where the complexity of the tasks’ distribution cannot be captured by a single meta-parameter vector. We address this limitation by conditional meta-learning, inferring a conditioning function mapping task’s side information into a meta-parameter vector that is appropriate for that task at hand. We characterize properties of the environment under which the conditional approach brings a substantial advantage over standard meta-learning and we highlight examples of environments, such as those with multiple clusters, satisfying these properties. We then propose a convex meta-algorithm providing a comparable advantage also in practice. Numerical experiments confirm our theoretical findings. Copyright © 2020, The Authors. All rights reserved.

关键词： machine learning

Maximal Closed Set and Half-Space Separations in Finite Closure Systems

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Seiffarth, Florian Horváth, Tamás Wrobel, Stefan Dept. of Computer Science University of Bonn Bonn Germany Fraunhofer IAIS Schloß Birlinghoven Sankt Augustin Germany Fraunhofer Center for Machine Learning Schloß Birlinghoven Sankt Augustin Germany

Several concept learning problems can be regarded as special cases of half-space separation in abstract closure systems over finite ground sets. For the typical scenario that the closure system is implicitly given via a closure operator, we show that the half-space separation problem is NP-complete. As a first approach to overcome this negative result, we relax the problem to maximal closed set separation, give a generic greedy algorithm solving this problem with a linear number of closure operator calls, and show that this bound is sharp. For a second direction, we consider Kakutani closure systems and prove that they are algorithmically characterized by the greedy algorithm. As a first special case of the general problem setting, we consider Kakutani closure systems over graphs and give a sufficient condition for this kind of closure systems in terms of forbidden graph minors. For a second special case, we then focus on closure systems over finite lattices, give an improved adaptation of the generic greedy algorithm, and present an application concerning subsumption lattices. Copyright © 2020, The Authors. All rights reserved.

关键词： machine learning

Automated Diagnosis of Cardiovascular Diseases from Cardiac Magnetic Resonance Imaging Using Deep learning Models: A Review

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Jafari, Mahboobeh Shoeibi, Afshin Khodatars, Marjane Ghassemi, Navid Moridian, Parisa Delfan, Niloufar Alizadehsani, Roohallah Khosravi, Abbas Ling, Sai Ho Zhang, Yu-Dong Wang, Shui-Hua Gorriz, Juan M. Rokny, Hamid Alinejad Acharya, U. Rajendra Internship in BioMedical Machine Learning Lab The Graduate School of Biomedical Engineering UNSW Sydney SydneyNSW2052 Australia Data Science and Computational Intelligence Institute University of Granada Spain Department of Medical Engineering Mashhad Branch Islamic Azad University Mashhad Iran Faculty of Computer Engineering Dept. of Artificial Intelligence Engineering K. N. Toosi University of Technology Tehran Iran Deakin University VIC3217 Australia Australia School of Computing and Mathematical Sciences University of Leicester Leicester United Kingdom Department of Psychiatry University of Cambridge United Kingdom BioMedical Machine Learning Lab The Graduate School of Biomedical Engineering UNSW Sydney SydneyNSW2052 Australia UNSW Data Science Hub The University of New South Wales SydneyNSW2052 Australia Research Centre Macquarie University Sydney2109 Australia Ngee Ann Polytechnic Singapore599489 Singapore Dept. of Biomedical Informatics and Medical Engineering Asia University Taichung Taiwan Dept. of Biomedical Engineering School of Science and Technology Singapore University of Social Sciences Singapore

In recent years, cardiovascular diseases (CVDs) have become one of the leading causes of mortality globally. CVDs appear with minor symptoms and progressively get worse. The majority of people experience symptoms such as exhaustion, shortness of breath, ankle swelling, fluid retention, and other symptoms when starting CVD. Coronary artery disease (CAD), arrhythmia, cardiomyopathy, congenital heart defect (CHD), mitral regurgitation, and angina are the most common CVDs. Clinical methods such as blood tests, electrocardiography (ECG) signals, and medical imaging are the most effective methods used for the detection of CVDs. Among the diagnostic methods, cardiac magnetic resonance imaging (CMR) is increasingly used to diagnose, monitor the disease, plan treatment and predict CVDs. Coupled with all the advantages of CMR data, CVDs diagnosis is challenging for physicians due to many slices of data, low contrast, etc. To address these issues, deep learning (DL) techniques have been employed to the diagnosis of CVDs using CMR data, and much research is currently being conducted in this field. This review provides an overview of the studies performed in CVDs detection using CMR images and DL techniques. The introduction section examined CVDs types, diagnostic methods, and the most important medical imaging techniques. In the following, investigations to detect CVDs using CMR images and the most significant DL methods are presented. Another section discussed the challenges in diagnosing CVDs from CMR data. Next, the discussion section discusses the results of this review, and future work in CVDs diagnosis from CMR images and DL techniques are outlined. The most important findings of this study are presented in the conclusion section. © 2022, CC BY-NC-ND.

关键词： Electrocardiography

Deep inference of latent dynamics with spatio-temporal super-resolution using selective backpropagation through time 21

学校读者我要写书评

暂无评论

Deep inference of latent dynamics with spatio-temporal super...

Proceedings of the 35th International Conference on Neural Information Processing Systems

作者： Feng Zhu Andrew R. Sedler Harrison A. Grier Nauman Ahad Mark A. Davenport Matthew T. Kaufman Andrea Giovannucci Chethan Pandarinath Neuroscience Graduate Program Emory University Center for Machine Learning Georgia Tech Computational Neuroscience Graduate Program The University of Chicago School of Electrical and Computer Engineering Georgia Tech Dept. of Organismal Biology and Anatomy The University of Chicago and Neuroscience Institute The University of Chicago Joint Dept. of Biomedical Engineering UNC Chapel Hill and NC State University and UNC Chapel Hill Neuroscience Center and NC State/UNC Closed-Loop Engineering for Advanced Rehabilitation (CLEAR) Coulter Dept. of Biomedical Engineering Emory University and Georgia Tech and Dept. of Neurosurgery Emory University

ISBN: (纸本)9781713845393

关键词：