检索结果-内蒙古大学图书馆

From connectomic to task-evoked fingerprints: individualized prediction of task contrasts from resting-state functional connectivity

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ngo, Gia Khosla, Meenakshi Jamison, Keith Kuceyeski, Amy Sabuncu, Mert R. School of Electrical & Computer Engineering Cornell University United States Radiology Weill Cornell Medicine United States Brain and Mind Research Institute Weill Cornell Medicine United States Nancy E. & Peter C. Meinig School of Biomedical Engineering Cornell University United States

Resting-state functional MRI (rsfMRI) yields functional connectomes that can serve as cognitive fingerprints of individuals. Connectomic fingerprints have proven useful in many machine learning tasks, such as predicting subject-specific behavioral traits or task-evoked activity. In this work, we propose a surface-based convolutional neural network (BrainSurfCNN) model to predict individual task contrasts from their resting-state fingerprints. We introduce a reconstructive-contrastive loss that enforces subject-specificity of model outputs while minimizing predictive error. The proposed approach significantly improves the accuracy of predicted contrasts over a well-established baseline. Furthermore, BrainSurfCNN’s prediction also surpasses test-retest benchmark in a subject identification task. 5 Copyright © 2020, The Authors. All rights reserved.

关键词： Forecasting

GRADIENTS OF BRAIN ORGANIZATION: SMOOTH SAILING FROM METHODS DEVELOPMENT TO USER COMMUNITY

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Royer, Jessica Paquola, Casey Valk, Sofie L. Kirschner, Matthias Hong, Seok-Jun Park, Bo-Yong Bethlehem, Richard A.I. Leech, Robert Yeo, B.T. Thomas Jefferies, Elizabeth Smallwood, Jonathan Margulies, Daniel Bernhardt, Boris C. Montreal Neurological Institute and Hospital McGill University MontrealQC Canada Forschungszentrum Julich Julich Germany Max Planck Institute for Human Cognitive and Brain Sciences Leipzig Germany Division of Adult Psychiatry Department of Psychiatry University Hospitals of Geneva Thonex Switzerland Center for Neuroscience Imaging Research Institute for Basic Science Suwon Korea Republic of Center for the Developing Brain Child Mind Institute New York United States Department of Biomedical Engineering Sungkyunkwan University Suwon Korea Republic of Department of Data Science Inha University Incheon Korea Republic of Department of Statistics and Data Science Inha University Incheon Korea Republic of Department of Psychology University of Cambridge Cambridge United Kingdom Centre for Neuroimaging Science King's College London London United Kingdom Centre for Sleep áCognition Centre for Translational Magnetic Resonance Research Yong Loo Lin School of Medicine National University of Singapore Singapore Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore Institute for Health Institute for Digital Medicine National University of Singapore Singapore Singapore National University of Singapore Singapore Singapore Martinos Center for Biomedical Imaging Massachusetts General Hospital CharlestownMA United States Department of Psychology University of York York United Kingdom Department of Psychology Queens University KingstonON Canada Universite de Paris Paris France

Multimodal neuroimaging grants a powerful in vivo window into the structure and function of the human brain. Recent methodological and conceptual advances have enabled investigations of the interplay between large-scale spatial trends - or gradients - in brain structure and function, offering a framework to unify principles of brain organization across multiple scales. Strong community enthusiasm for these techniques has been instrumental in their widespread adoption and implementation to answer key questions in neuroscience. Following a brief review of current literature on this framework, this perspective paper will highlight how pragmatic steps aiming to make gradient methods more accessible to the community propelled these techniques to the forefront of neuroscientific inquiry. More specifically, we will emphasize how interest for gradient methods was catalyzed by data sharing, open-source software development, as well as the organization of dedicated workshops led by a diverse team of early career researchers. To this end, we argue that the growing excitement for brain gradients is the result of coordinated and consistent efforts to build an inclusive community and can serve as a case in point for future innovations and conceptual advances in neuroinformatics. We close this perspective paper by discussing challenges for the continuous refinement of neuroscientific theory, methodological innovation, and real-world translation to maintain our collective progress towards integrated models of brain organization. © 2024, CC BY.

关键词： Gradient methods

A Data-Driven Reinforcement Learning Solution Framework for Optimal and Adaptive Personalization of a Hip Exoskeleton

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Tu, Xikai Li, Minhan Liu, Ming Si, Jennie Huang, He Helen NCSU/UNC Department of Biomedical Engineering NC State University University of North Carolina at Chapel Hill Raleigh Chapel HillNC27695-7115 United States Department of Electrical Computer and Energy Engineering Arizona State University TempeAZ85281 United States Department of Mechanical Engineering Hubei University of Technology Wuhan Hubei430068 China

Robotic exoskeletons are exciting technologies for augmenting human mobility. However, designing such a device for seamless integration with the human user and to assist human movement still is a major challenge. This paper aims at developing a novel data-driven solution framework based on reinforcement learning (RL), without first modeling the human-robot dynamics, to provide optimal and adaptive personalized torque assistance for reducing human efforts during walking. Our automatic personalization solution framework includes the assistive torque profile with two control timing parameters (peak and offset timings), the least square policy iteration (LSPI) for learning the parameter tuning policy, and a cost function based on transferred work ratio. The proposed controller was successfully validated on a healthy human subject to assist unilateral hip extension in walking. The results showed that the optimal and adaptive RL controller as a new approach was feasible for tuning assistive torque profile of the hip exoskeleton that coordinated with human actions and reduced activation level of hip extensor muscle in *** Codes 68T40, 93C85 © 2020, CC-BY-NC-SA.

关键词： Reinforcement learning

A review of molecular and genetic factors for determining mild traumatic brain injury severity and recovery

学校读者我要写书评

暂无评论

Brain Disorders 2022年 8卷 100058-100058页

作者： Tajik, Mahnaz Danielli, Ethan Sharma, Bhanu Noseworthy, Michael D School of Biomedical Engineering McMaster University 1280 Main St W Hamilton L8S 4L8 ON Canada Imaging Research Centre St. Joseph's Healthcare Hamilton 50 Charlton Ave E Hamilton L8N 4A6 ON Canada Department of Electrical and Computer Engineering McMaster University 1280 Main St W Hamilton L8S 4L8 ON Canada Department of Radiology McMaster University 1280 Main St W Hamilton L8S 4L8 ON Canada

Mild traumatic brain injuries (mTBI) affect millions of people globally every year. The clinical presentation of this injury is highly variable, and its progression from the acute to chronic stages of injury is driven by a dynamic pathophysiology. More specifically, biomechanical brain damage can trigger complex cellular, molecular, functional, genetic, and metabolomic changes. Recent research has taken aim at understanding the association between such complex changes and clinical outcomes, with the ultimate intent of identifying prognostic indicators. This is important as to date, current diagnostic protocols using patient reported symptom tracking and routine medical imaging are limited, often subjective, and can lead to missed diagnoses. Thus, neither patients nor their physicians can currently predict recovery timeline and whether recovery will be complete. Consequently, biological markers need to be determined that can improve diagnostic and recovery assessments following brain injuries. Possible indicator candidates, based on human and animal research, include the expression of neuroprotective genes and microRNAs (i.e., GFAP, BDNF, MBP) and single nucleotide polymorphisms (i.e., BDNF, COMT, APOE, D2R2). However, these factors are non-specific in terms of injury location and severity. Due to the vast range of physiological, molecular and omics alterations present post-mTBI, it is clear that mTBIs are a highly complex pathophysiological clinical problem. Thus, the purpose of this review is to provide a comprehensive understanding of post-mTBI genetic and metabolic brain changes, both at the cellular and molecular level, to understand how they can affect the symptoms and outcome of mTBIs. We discuss how these changes may be leveraged for improved acute detection of brain injury, and their potential for use in future personalized treatments. © 2022 The Author(s)

关键词： Biomarkers Epigenetic Genetics Metabolomics MicroRNA Mild traumatic brain injury (mTBI)

Reliability of Electrocardiogram Signals during Feature Extraction Stage for Smart Textile Shirts

学校读者我要写书评

暂无评论

Journal of Physics: Conference Series 2021年第1期2071卷

作者： MM Mohd Nawawi Khairul Azami Sidek Amelia Wong Azman Fazli Mohd Nasir Nashrul Department of Electrical and Computer Engineering Universiti Islam Antarabangsa Malaysia Kuala Lumpur Malaysia. Biomedical Electronic Engineering Programme School of Mechatronic Universiti Malaysia Perlis Perlis Malaysia.

Wearable smart textiles have garnered significant interest due to their high flexibility, reusability, convenience and ability to work on home-based, real-life and real-time monitoring. Wearable smart textiles are shirts with inbuilt textile sensors that enable electrocardiogram (ECG) data to be collected more comfortably and smoothly outside the laboratory and clinical environment for a continuous and longer duration for ECG data collection. However, the existing ECG wearable smart textile main challenge is maintaining the quality and reliability of data across multiple wearable smart textile shirts. Therefore, this research analyses the capability of ECG morphology during Feature Extraction stages for different wearable smart textile shirts. This paper reports the experiment conducted on eleven healthy volunteers, either wearing the Hexoskin smart shirt or the HeartIn Fit shirt or both. ECG data were recorded while they are doing normal daily routine activities for at least 45 minutes. The study demonstrates a significant possibility of reliability in Feature Extraction stages at different time instances among subject and wearable smart textiles shirts. With R peaks average between 0.543 to 1.194 mV and R-R interval average between 0.625 to 0.799 seconds, the study concludes that both wearable smart textiles do not significantly differ in Feature Extraction stages. Thus, both wearable smart textiles gave a significant result, although both are affected by their wearer's motion artefacts during the shifting of body postures and the wearer's body physical states. Furthermore, the ECG morphology in this study has yielded a promising result in real life and as on-the-go ECG smart textile biometric readiness for future explorations.

关键词：

Robotic Knee Parameter Tuning Using Approximate Policy Iteration 4th

学校读者我要写书评

暂无评论

Robotic Knee Parameter Tuning Using Approximate Policy Itera...

4th International Conference on Cognitive Systems and Information Processing, ICCSIP 2018

作者： Gao, Xiang Wen, Yue Li, Minhan Si, Jennie Huang, He [Helen] Department of Electrical Computer and Energy Engineering Arizona State University TempeAZ85281 United States Joint Department of Biomedical Engineering North Carolina State University RaleighNC27695 United States University of North Carolina at Chapel Hill Chapel HillNC27599 United States

ISBN: (纸本)9789811379826

This paper presents an online model-free reinforcement learning based controller realized by approximate dynamic programming for a robotic knee as part of a human-machine system. Traditionally, prosthesis wearers’ gait performance is improved by manually tuning the impedance parameters. In this paper, we show that the parameter tuning problem can be formulated as an optimal control problem and thus solved by dynamic programming. Toward this goal, we constructed an quadratic instantaneous cost, which resulted in a value function that could be approximated by a neural network. The control policy is then solved by the least-squared method iteratively, a framework of which we refer to as approximate policy iteration. We performed extensive simulations based on prosthetic kinetics and human performance data extracted from real human subjects. Our results show that the proposed parameter tuning algorithm can be readily used for adaptive optimal tuning of prosthetic knee control parameters and the tuning process is time and sample efficient. © 2019, Springer Nature Singapore Pte Ltd.

关键词： Dynamic programming

Perspectives on the modeling of the neuromusculoskeletal system to investigate the influence of neurodegenerative diseases on sensorimotor control

学校读者我要写书评

暂无评论

Research on biomedical engineering 2018年第2期34卷 176-186页

作者： Elias, Leonardo Abdala Matoso, Débora Elisa da Costa Watanabe, Renato Naville Kohn, André Fabio Neural Engineering Research Laboratory Department of Biomedical Engineering School of Electrical and Computer Engineering University of Campinas CampinasSP Brazil Center for Biomedical Engineering University of Campinas CampinasSP Brazil Biomedical Engineering Laboratory Escola Politécnica University of São Paulo São PauloSP Brazil Biomedical Engineering Federal University of ABC Santo AndréSP Brazil University of São Paulo São PauloSP Brazil

Introduction: The understanding of the neurophysiological mechanisms underlying movement control can be much furthered using computational models of the neuromusculoskeletal system. Biologically based multi-scale neuromusculoskeletal models have a great potential to provide new theories and explanations related to mechanisms behind muscle force generation at the molecular, cellular, synaptic, and systems levels. Albeit some efforts have been made to investigate how neurodegenerative diseases alter the dynamics of individual elements of the neuromuscular system, such diseases have not been analyzed from a systems viewpoint using multi-scale models. Overview and Perspectives: This perspective article synthesizes what has been done in terms of multi-scale neuromuscular development and points to a few directions where such models could be extended so that they can be useful in the future to discover early predictors of neurodegenerative diseases, as well as to propose new quantitative clinical neurophysiology approaches to follow the course of improvements associated with different therapies (drugs or others). Concluding Remarks: Therefore, this article will present how existing biologically based multi-scale models of the neuromusculoskeletal system could be expanded and adapted for clinical applications. It will point to mechanisms operating at different levels that would be relevant to be considered during model development, along with implications for interpreting experimental results from neurological patients. © 2018, Brazilian Society of biomedical engineering. All rights reserved.

关键词： Neurodegenerative diseases

Scan Specific Artifact Reduction in K-space (SPARK) Neural Networks Synergize with Physics-based Reconstruction to Accelerate MRI

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Arefeen, Yamin Beker, Onur Cho, Jaejin Yu, Heng Adalsteinsson, Elfar Bilgic, Berkin Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA United States Computer and Communication Sciences École Polytechnique Fédérale de Lausanne Lausanne Switzerland Athinoula A. Martinos Center for Biomedical Imaging CharlestownMA United States Department of Automation Tsinghua University Beijing China Harvard-MIT Health Sciences and Technology Massachusetts Institute of Technology CambridgeMA United States Institute for Medical Engineering and Science Massachusetts Institute of Technology CambridgeMA United States Department of Radiology Harvard Medical School BostonMA United States

Purpose: To develop a scan-specific model that estimates and corrects k-space errors made when reconstructing accelerated Magnetic Resonance Imaging (MRI) data. Methods: Scan-Specific Artifact Reduction in k-space (SPARK) trains a convolutional-neural-network to estimate and correct k-space errors made by an input reconstruction technique by back-propagating from the mean-squared-error loss between an auto-calibration signal (ACS) and the input technique's reconstructed ACS. First, SPARK is applied to GRAPPA and demonstrates improved robustness over other scan-specific models, such as RAKI and residual-RAKI. Subsequent experiments demonstrate that SPARK synergizes with residual-RAKI to improve reconstruction performance. SPARK also improves reconstruction quality when applied to advanced acquisition and reconstruction techniques like 2D virtual coil (VC-) GRAPPA, 2D LORAKS, 3D GRAPPA without an integrated ACS region, and 2D/3D wave-encoded imaging. Results: SPARK yields SSIM improvement and 1.5 - 2x RMSE reduction when applied to GRAPPA and improves robustness to ACS size for various acceleration rates in comparison to other scan-specific techniques. When applied to advanced reconstruction techniques such as residual-RAKI, 2D VC-GRAPPA and LORAKS, SPARK achieves up to 20% RMSE improvement. SPARK with 3D GRAPPA also improves RMSE performance by ~2x, SSIM performance, and perceived image quality without a fully sampled ACS region. Finally, SPARK synergizes with non-cartesian, 2D and 3D wave-encoding imaging by reducing RMSE between 20 - 25% and providing qualitative improvements. Conclusion: SPARK synergizes with physics-based acquisition and reconstruction techniques to improve accelerated MRI by training scan-specific models to estimate and correct reconstruction errors in k-space. Copyright © 2021, The Authors. All rights reserved.

关键词： Image reconstruction

EndoL2H: Deep super-resolution for capsule endoscopy

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Almalioglu, Yasin Gokce, Abdulkadir Incetan, Kagan Simsek, Muhammed Ali Ararat, Kivanc Chen, Richard J. Durr, Nichalos J. Mahmood, Faisal Turan, Mehmet Institute of Biomedical Engineering Bogazici University Turkey Computer Science Department University of Oxford Oxford United Kingdom Electrical and Electronics Engineering Bogazici University Turkey BaltimoreMD United States Department of Computational Engineering Friedrich Alexander Universität Erlangen-Nürnberg Germany Department of Pathology Harvard Medical School BostonMA United States

Wireless capsule endoscopy is the preferred modality for diagnosis and assessment of small bowel disease. However, the poor resolution is a limitation for both subjective and automated diagnostics. Enhanced-resolution endoscopy has shown to improve adenoma detection rate for conventional endoscopy and is likely to do the same for capsule endoscopy. In this work, we propose and quantitatively validate a novel framework to learn a mapping from low-to-high resolution endoscopic images. We use conditional adversarial networks and spatial attention to improve the resolution by up to a factor of 8×. Our quantitative study demonstrates the superiority of our proposed approach over Super-Resolution Generative Adversarial Network (SRGAN) and bicubic interpolation. For qualitative analysis, visual Turing tests were performed by 16 gastroenterologists to confirm the clinical utility of the proposed approach. Our approach is generally applicable to any endoscopic capsule system and has the potential to improve diagnosis and better harness computational approaches for polyp detection and characterization. Our code and trained models are available at https://***/akgokce/EndoL2H. Copyright © 2020, The Authors. All rights reserved.

关键词： Endoscopy