检索结果-内蒙古大学图书馆

CloudBrain-ReconAI: A Cloud Computing Platform for MRI Reconstruction and Radiologists’ Image Quality Evaluation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhou, Yirong Qian, Chen Li, Jiayu Wang, Zi Hu, Yu Qu, Biao Zhu, Liuhong Zhou, Jianjun Kang, Taishan Lin, Jianzhong Hong, Qing Dong, Jiyang Guo, Di Qu, Xiaobo The Department of Electronic Science Biomedical Intelligent Cloud R&D Center Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance National Institute for Data Science in Health and Medicine Xiamen University Xiamen361102 China The Department of Instrumental and Electrical Engineering Xiamen University Xiamen361005 China Fudan University Xiamen361015 China The Department of Radiology Zhongshan Hospital Xiamen University School of Medicine Xiamen University Xiamen361004 China The China Mobile Group Xiamen361000 China The School of Computer and Information Engineering Xiamen University of Technology Xiamen361024 China

Efficient collaboration between engineers and radiologists is important for image reconstruction algorithm development and image quality evaluation in magnetic resonance imaging (MRI). Here, we develop CloudBrain-ReconAI, an online cloud computing platform, for algorithm deployment, fast and blind reader study. This platform supports online image reconstruction using state-of-the-art artificial intelligence and compressed sensing algorithms with applications for fast imaging (Cartesian and non-Cartesian sampling) and high-resolution diffusion imaging. Through visiting the website, radiologists can easily score and mark images. Then, automatic statistical analysis will be provided. CloudBrain-ReconAI is now open accessed at https://***/*** and will be continually improved to serve the MRI research community. Copyright © 2022, The Authors. All rights reserved.

关键词： Cloud platforms

Federated Cross Learning for Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xu, Xuanang Deng, Hannah H. Chen, Tianyi Kuang, Tianshu Barber, Joshua C. Kim, Daeseung Gateno, Jaime Xia, James J. Yan, Pingkun Department of Biomedical Engineering Center for Biotechnology and Interdisciplinary Studies Rensselaer Polytechnic Institute 110 8th St TroyNY12180 United States Department of Oral and Maxillofacial Surgery Houston Methodist Research Institute 6560 Fannin St HoustonTX77030 United States Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute 110 8th St TroyNY12180 United States Weill Medical College Cornell University 407 E 61st St New YorkNY10065 United States

Federated learning (FL) can collaboratively train deep learning models using isolated patient data owned by different hospitals for various clinical applications, including medical image segmentation. However, a major problem of FL is its performance degradation when dealing with data that are not independently and identically distributed (non-iid), which is often the case in medical images. In this paper, we first conduct a theoretical analysis on the FL algorithm to reveal the problem of model aggregation during training on non-iid data. With the insights gained through the analysis, we propose a simple yet effective method, federated cross learning (FedCross), to tackle this challenging problem. Unlike the conventional FL methods that combine multiple individually trained local models on a server node, our FedCross sequentially trains the global model across different clients in a round-robin manner, and thus the entire training procedure does not involve any model aggregation steps. To further improve its performance to be comparable with the centralized learning method, we combine the FedCross with an ensemble learning mechanism to compose a federated cross ensemble learning (FedCrossEns) method. Finally, we conduct extensive experiments using a set of public datasets. The experimental results show that the proposed FedCross training strategy outperforms the mainstream FL methods on non-iid data. In addition to improving the segmentation performance, our FedCrossEns can further provide a quantitative estimation of the model uncertainty, demonstrating the effectiveness and clinical significance of our designs. Source code is publicly available at https://***/DIAL-RPI/FedCross. © 2022, CC BY.

关键词： Learning systems

Visual detection of misfolded alpha-synuclein and prions via capillary-based quaking-induced conversion assay (Cap-QuIC)

学校读者我要写书评

暂无评论

npj Biosensing

npj Biosensing 2024年第1期1卷 1-8页

作者： Peter R. Christenson Hyeonjeong Jeong Hyerim Ahn Hye Yoon Park Sang-Hyun Oh Manci Li Gage Rowden Rachel L. Shoemaker Peter A. Larsen Department of Electrical and Computer Engineering University of Minnesota Minneapolis MN 55455 USA Minnesota Center for Prion Research and Outreach University of Minnesota St. Paul MN 55108 USA Department of Veterinary and Biomedical Sciences University of Minnesota St. Paul MN 55108 USA

Neurodegenerative protein misfolding diseases impact tens of millions of people worldwide, contributing to millions of deaths and economic hardships across multiple scales. The prevalence of neurodegenerative disease is predicted to greatly increase over the coming decades, yet effective diagnostics for such diseases are limited. Most diagnoses come from the observation of external symptoms in clinical settings, which typically manifest during relatively advanced stages of disease, thus limiting potential therapeutic applications. While progress is being made on biomarker testing, the underlying methods largely rely on fragile and expensive equipment that limits their point-of-care potential, especially in developing countries. Here we present Capillary-based Quaking Induced Conversion (Cap-QuIC) as a visual diagnostic assay based on simple capillary action for the detection of neurodegenerative disease without necessitating expensive and complex capital equipment. We demonstrate that Cap-QuIC has the potential to be a detection tool for a broad range of misfolded proteins by successfully distinguishing misfolded versus healthy proteins associated with Parkinson’s disease (α-synuclein) and Chronic Wasting Disease (prions). Additionally, we show that Cap-QuIC can accurately classify biological tissue samples from wild white-tailed deer infected with Chronic Wasting Disease. Our findings elucidate the underlying mechanism that enables the Cap-QuIC assay to distinguish misfolded protein, highlighting its potential as a diagnostic technology for neurodegenerative diseases.

关键词：

Visual-tactile Fusion for Transparent Object Grasping in Complex Backgrounds

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Li, Shoujie Yu, Haixin Ding, Wenbo Liu, Houde Ye, Linqi Xia, Chongkun Wang, Xueqian Zhang, Xiao-Ping Tsinghua Shenzhen International Graduate School Shenzhen518055 China The RISC-V International Open Source Laboratory Tsinghua-Berkeley Shenzhen Institute Shenzhen518055 China The Institute of Artificial Intelligence Collaborative Innovation Center for the Marine Artificial Intelligence Shanghai University Shanghai200444 China The Department of Electrical Computer and Biomedical Engineering Ryerson University TorontoONM5B 2K3 Canada

The grasping of transparent objects is challenging but of significance to robots. In this article, a visual-tactile fusion framework for transparent object grasping in complex backgrounds is proposed, which synergizes the advantages of vision and touch, and greatly improves the grasping efficiency of transparent objects. First, we propose a multi-scene synthetic grasping dataset named SimTrans12K together with a Gaussian-Mask annotation method. Next, based on the TaTa gripper, we propose a grasping network named transparent object grasping convolutional neural network (TGCNN) for grasping position detection, which shows good performance in both synthetic and real scenes. Inspired by human grasping, a tactile calibration method and a visual-tactile fusion classification method are designed, which improve the grasping success rate by 36.7% compared to direct grasping and the classification accuracy by 39.1%. Furthermore, the Tactile Height Sensing (THS) module and the Tactile Position Exploration (TPE) module are added to solve the problem of grasping transparent objects in irregular and visually undetectable scenes. Experimental results demonstrate the validity of the framework. Copyright © 2022, The Authors. All rights reserved.

关键词： Calibration

Identification of brain states, transitions, and communities using functional MRI

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Bian, Lingbin Cui, Tiangang Yeo, B.T. Thomas Fornito, Alex Razi, Adeel Keith, Jonathan School of Mathematics Monash University Australia Turner Institute for Brain and Mental Health School of Psychological Sciences and Monash Biomedical Imaging Monash University Australia Department of Electrical and Computer Engineering National University of Singapore Singapore Wellcome Centre for Human Neuroimaging University College London United Kingdom

Brain function relies on a precisely coordinated and dynamic balance between the functional integration and segregation of distinct neural systems. Characterizing the way in which neural systems reconfigure their interactions to give rise to distinct but hidden brain states remains an open challenge. In this paper, we propose a Bayesian model-based characterization of latent brain states and showcase a novel method based on posterior predictive discrepancy using the latent block model to detect transitions between latent brain states in blood oxygen level-dependent (BOLD) time series. The set of estimated parameters in the model includes a latent label vector that assigns network nodes to communities, and also block model parameters that reflect the weighted connectivity within and between communities. Besides extensive in-silico model evaluation, we also provide empirical validation (and replication) using the Human Connectome Project (HCP) dataset of 100 healthy adults. Our results obtained through an analysis of task-fMRI data during working memory performance show appropriate lags between external task demands and change-points between brain states, with distinctive community patterns distinguishing fixation, low-demand and high-demand task conditions. © 2021, CC BY.

关键词： Bayesian networks

MT51 Cost-Consequences Analysis (CCA) of Small Bowel Video Capsule Endoscopy (VCE) Digital Artificial Intelligence (AI) Monitoring Application Developed at Hosmartai (Horizon 2020 Funded)

学校读者我要写书评

暂无评论

Value in Health 2024年第12期27卷 S495-S495页

作者： Chatzikou, M Latsou, D Apostolidis, G Charisis, V Hadjidimitriou, S Hadjileontiadis, L Lazaridis, N Germanidis, G Pharmecons Easy Access LtD York UK Signal Processing & Biomedical Technology Unit Department of Electrical & Computer Engineering Aristotle University of Thessaloniki Thessaloniki Greece Department of Biomedical Engineering Khalifa University Abu Dhabi Abu Dhabi United Arab Emirates 1st Department of Internal Medicine University General Hospital of Thessaloniki AHEPA Thessaloniki 54 Greece 1st Department of Internal Medicine University General Hospital of Thessaloniki AHEPA Thessaloniki Greece

On-scalp printing of personalized electroencephalography e-tattoos

学校读者我要写书评

暂无评论

Cell Biomaterials 2025年第1期1卷

作者： Luize Scalco de Vasconcelos Yichen Yan Pukar Maharjan Satyam Kumar Minsu Zhang Bowen Yao Hongbian Li Sidi Duan Eric Li Eric Williams Sandhya Tiku Pablo Vidal R. Sergio Solorzano-Vargas Wen Hong Yingjie Du Zixiao Liu Fumiaki Iwane Charles Block Andrew T. Repetski Philip Tan Nanshu Lu Department of Aerospace Engineering and Engineering Mechanics The University of Texas at Austin Austin TX 78712 USA Department of Materials Science and Engineering University of California Los Angeles Los Angeles CA 90095 USA Chandra Family Department of Electrical and Computer Engineering The University of Texas at Austin Austin TX 78712 USA David Geffen School of Medicine University of California Los Angeles Los Angeles CA 90095 USA Department of Biomedical Engineering The University of Texas at Austin Austin TX 78712 USA Department of Mechanical Engineering The University of Texas at Austin Austin TX 78712 USA Texas Materials Institute The University of Texas at Austin Austin TX 78712 USA

Electroencephalography (EEG) is crucial for diagnosing neurological disorders and facilitating brain-computer interfaces. Traditional EEG setups with wet gels and cumbersome cables are labor intensive, uncomfortable, and degrade over time. Dry, skin-conformable e-tattoos offer a comfortable and user-friendly alternative but struggle with hairy scalps. To tackle this problem, we introduce non-contact digital printing of e-tattoos directly on the hairy scalp. Biocompatible inks are crafted for low-impedance electrodes and high-conductivity interconnects. The fabrication system includes a custom sensor layout design algorithm based on individual head scans and a 5-axis robot controlling a microjet printhead for safe and precise on-scalp ink delivery. After printing, these inks rapidly self-dry into conductive films perfectly conformed to scalp skin, with enhanced wearability, longevity, and skin adhesion compared with gel electrodes and transferred e-tattoos. Motion imagery and error-related potentials are effectively measured by these printed e-tattoos. This innovation heralds a new era in on-body manufacturing of personalized e-tattoos.

关键词： additive manufacture inkjet printing conformable printing on-body manufacture conductive ink flexible electronics stretchable electronics wearable electronics personalized e-tattoo brain-computer interface BCI

cytoNet: Spatiotemporal network analysis of cell communities (vol 18, e1009846, 2022)

学校读者我要写书评

暂无评论

PLOS COMPUTATIONAL BIOLOGY 2022年第11期18卷 e1009846页

作者： Mahadevan, A. S. Long, B. L. Hu, C. W. Ryan, D. T. Grandel, N. E. Department of Bioengineering University of Pennsylvania Philadelphia Pennsylvania United States of America. Department of Bioengineering Rice University Houston Texas United States of America. Systems Synthetic and Physical Biology Program Rice University Houston Texas United States of America. Department of Biomedical Engineering University of Texas at San Antonio San Antonio Texas United States of America. Biophysics Graduate Program University of California Berkeley California United States of America. Department of Oral & Maxillofacial Surgery University of Texas Health Science Center at San Antonio San Antonio Texas United States of America. Department of Electrical and Computer Engineering Rice University Houston Texas United States of America.

We introduce cytoNet, a cloud-based tool to characterize cell populations from microscopy images. cytoNet quantifies spatial topology and functional relationships in cell communities using principles of network science. Capturing multicellular dynamics through graph features, cytoNet also evaluates the effect of cell-cell interactions on individual cell phenotypes. We demonstrate cytoNet's capabilities in four case studies: 1) characterizing the temporal dynamics of neural progenitor cell communities during neural differentiation, 2) identifying communities of pain-sensing neurons in vivo, 3) capturing the effect of cell community on endothelial cell morphology, and 4) investigating the effect of laminin α4 on perivascular niches in adipose tissue. The analytical framework introduced here can be used to study the dynamics of complex cell communities in a quantitative manner, leading to a deeper understanding of environmental effects on cellular behavior. The versatile, cloud-based format of cytoNet makes the image analysis framework accessible to researchers across domains.

关键词： Calcium imaging Fluorescence imaging Growth factors Centrality Cell cycle and cell division Endothelial cells Network analysis Stem cells

Double-containment coil with enhanced winding mounting for transcranial magnetic stimulation with reduced acoustic noise

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Koponen, Lari M. GoetzAngel, Stefan M. Peterchev, V. Department of Psychiatry & Behavioral Sciences Duke University DurhamNC United States Department of Psychiatry & Behavioral Sciences Department of Electrical & Computer Engineering Department of Neurosurgery Duke University DurhamNC United States Department of Psychiatry & Behavioral Sciences Department of Electrical & Computer Engineering Department of Neurosurgery Department of Biomedical Engineering Duke University DurhamNC United States

Objective: This work aims to reduce the acoustic noise level of transcranial magnetic stimulation (TMS) coils. TMS requires high currents (several thousand amperes) to be pulsed through the coil, which generates a loud acoustic impulse whose peak sound pressure level (SPL) can exceed 130 dB(Z). This sound poses a risk to hearing and elicits unwanted neural activation of auditory brain circuits. Methods: We propose a new double-containment coil with enhanced winding mounting (DCC), which utilizes acoustic impedance mismatch to contain and dissipate the impulsive sound within an air-tight outer casing. The coil winding is potted in a rigid block, which is mounted to the outer casing by its acoustic nodes that are subject to minimum vibration during the pulse. The rest of the winding block is isolated from the casing by an air gap, and sound is absorbed by foam within the casing. The casing thickness under the winding center is minimized to maximize the coil electric field output. Results: Compared to commercial figure-of-eight TMS coils, the DCC prototype has 10–33 dB(Z) lower SPL at matched stimulation strength, whilst providing 22% higher maximum stimulation strength than equally focal commercial coils. Conclusion: The DCC design greatly reduces the acoustic noise of TMS while increasing the achievable stimulation strength. Significance: The acoustic noise reduction from our coil design is comparable to that provided by typical hearing protection devices. This coil design approach can enhance hearing safety and reduce auditory co-activations in the brain and other detrimental effects of TMS sound. Copyright © 2020, The Authors. All rights reserved.

关键词： Acoustic noise