检索结果-内蒙古大学图书馆

Large language models for disease diagnosis: a scoping review

学校读者我要写书评

暂无评论

npj Artificial Intelligence

npj Artificial Intelligence 2025年第1期1卷 1-17页

作者： Shuang Zhou Jeremy Yeung Mingquan Lin Rui Zhang Zidu Xu Mian Zhang Chunpu Xu Jiashuo Wang Kaishuai Xu Yawen Guo Zaifu Zhan Yi Fang Sirui Ding Liqiao Xia Daochen Zha Dongming Cai Genevieve B. Melton Division of Computational Health Sciences Department of Surgery University of Minnesota Minneapolis MN USA School of Nursing Columbia University New York New York USA Erik Jonsson School of Engineering and Computer Science University of Texas at Dallas Richardson TX USA Department of Computing The Hong Kong Polytechnic University Hong Kong Hong Kong SAR Department of Informatics University of California Irvine Irvine CA USA Department of Electrical and Computer Engineering University of Minnesota Minneapolis MN USA Department of Computer Science New York University (Shanghai) Shanghai China Bakar Computational Health Sciences Institute University of California San Francisco San Francisco CA USA Department of Industrial and Systems Engineering The Hong Kong Polytechnic University Hong Kong Hong Kong SAR Independent Researcher San Francisco CA USA Department of Neurology University of Minnesota Minneapolis MN USA Institute for Health Informatics and Division of Colon and Rectal Surgery Department of Surgery University of Minnesota Minneapolis MN USA

Automatic disease diagnosis has become increasingly valuable in clinical practice. The advent of large language models (LLMs) has catalyzed a paradigm shift in artificial intelligence, with growing evidence supporting the efficacy of LLMs in diagnostic tasks. Despite the increasing attention in this field, a holistic view is still lacking. Many critical aspects remain unclear, such as the diseases and clinical data to which LLMs have been applied, the LLM techniques employed, and the evaluation methods used. In this article, we perform a comprehensive review of LLM-based methods for disease diagnosis. Our review examines the existing literature across various dimensions, including disease types and associated clinical specialties, clinical data, LLM techniques, and evaluation methods. Additionally, we offer recommendations for applying and evaluating LLMs for diagnostic tasks. Furthermore, we assess the limitations of current research and discuss future directions. To our knowledge, this is the first comprehensive review for LLM-based disease diagnosis.

关键词：

Wide-angle giant photonic spin Hall effect

学校读者我要写书评

暂无评论

Physical Review B 2022年第7期106卷 075409-075409页

作者： Zhihao Chen Yu Chen Yaodong Wu Xinxing Zhou Handong Sun Tony Low Hongsheng Chen Xiao Lin Key Laboratory of Low-Dimensional Quantum Structures and Quantum Control of Ministry of Education Synergetic Innovation Center for Quantum Effects and Applications School of Physics and Electronics Hunan Normal University Changsha 410081 China International Collaborative Laboratory of 2D Materials for Optoelectronics Science and Technology Engineering Technology Research Center for 2D Material Information Function Devices and Systems of Guangdong Province Institute of Microscale Optoelectronics Shenzhen University Shenzhen 518060 China Division of Physics and Applied Physics School of Physical and Mathematical Sciences Nanyang Technological University Singapore 637371 Singapore Centre for Disruptive Photonic Technologies (CDPT) School of Physical and Mathematical Sciences Nanyang Technological University Singapore 637371 Singapore Department of Electrical and Computer Engineering University of Minnesota Minneapolis Minnesota 55455 USA Interdisciplinary Center for Quantum Information State Key Laboratory of Modern Optical Instrumentation ZJU-Hangzhou Global Scientific and Technological Innovation Center College of Information Science and Electronic Engineering Zhejiang University Hangzhou 310027 China International Joint Innovation Center ZJU-UIUC Institute The Electromagnetics Academy at Zhejiang University Zhejiang University Haining 314400 China

The photonic spin Hall effect is a manifestation of the spin-orbit interaction of light and can be measured by a transverse shift δ of photons with opposite spins. The precise measurement of transverse shifts can enable many spin-related applications, such as precise metrology and optical sensing. However, this transverse shift is generally small (i.e., δ/λ<10−1, where λ is the wavelength), which impedes its precise measurement. To date, proposals to generate a giant spin Hall effect (namely, with δ/λ>102) have severe limitations, particularly its occurrence only over a narrow angular cone (with a width of Δθ<1∘). Here we propose a universal scheme to realize the wide-angle giant photonic spin Hall effect with Δθ>70∘ by exploiting the interface between free space and uniaxial epsilon-near-zero media. The underlying mechanism is ascribed to the almost-perfect polarization splitting between s and p polarized waves at the designed interface. Remarkably, this almost-perfect polarization splitting does not resort to the interference effect and is insensitive to the incident angle, which then gives rise to the wide-angle giant photonic spin Hall effect.

关键词： Light-matter interaction Photonics Spin Hall effect Spin-orbit coupling

Emerging chiral optics from chiral interfaces

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhang, Xinyan Zhong, Yuhan Low, Tony Chen, Hongsheng Lin, Xiao Interdisciplinary Center for Quantum Information State Key Laboratory of Modern Optical Instrumentation ZJU-Hangzhou Global Science and Technology Innovation Center College of Information Science and Electronic Engineering Zhejiang University Hangzhou310027 China International Joint Innovation Center ZJU-UIUC Institute Zhejiang University Haining314400 China Department of Electrical and Computer Engineering University of Minnesota MinneapolisMN55455 United States

Twisted atomic bilayers are emerging platforms for manipulating chiral light-matter interaction at the extreme nanoscale, due to their inherent magnetoelectric responses induced by the finite twist angle and quantum interlayer coupling between the atomic layers. Recent studies have reported the direct correspondence between twisted atomic bilayers and chiral metasurfaces, which features a chiral surface conductivity, in addition to the electric and magnetic surface conductivities. However, far-field chiral optics in light of these consitututive conductivities remains unexplored. Within the framework of the full Maxwell equations, we find that the chiral surface conductivity can be exploited to realize perfect polarization transformation between linearly polarized light. Remarkably, such an exotic chiral phenomenon can occur either for the reflected or transmitted light. © 2021, CC BY.

关键词： Electric conductivity

Communication-efficient robust federated learning over heterogeneous datasets

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Dong, Yanjie Giannakis, Georgios B. Chen, Tianyi Cheng, Julian Hossain, Md. Jahangir Leung, Victor C.M. Department of Electrical and Computer Engineering University of British Columbia VancouverBCV6T 1Z4 Canada Department of Electrical and Computer Engineering and Digital Technology Center University of Minnesota MinneapolisMN55455 United States Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute TroyNY12180 United States School of Engineering University of British Columbia KelownaBCV1V 1V7 Canada College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China

This work investigates fault-resilient federated learning when the data samples are non-uniformly distributed across workers, and the number of faulty workers is unknown to the central server. In the presence of adversarially faulty workers who may strategically corrupt datasets, the local messages exchanged (e.g., local gradients and/or local model parameters) can be unreliable, and thus the vanilla stochastic gradient descent (SGD) algorithm is not guaranteed to converge. Recently developed algorithms improve upon vanilla SGD by providing robustness to faulty workers at the price of slowing down convergence. To remedy this limitation, the present work introduces a fault-resilient proximal gradient (FRPG) algorithm that relies on Nesterov’s acceleration technique. To reduce the communication overhead of FRPG, a local (L) FRPG algorithm is also developed to allow for intermittent server-workers parameter exchanges. For strongly convex loss functions, FRPG and LFRPG have provably faster convergence rates than a benchmark robust stochastic aggregation algorithm. Moreover, LFRPG converges faster than FRPG while using the same communication rounds. Numerical tests performed on various real datasets confirm the accelerated convergence of FRPG and LFRPG over the robust stochastic aggregation benchmark and competing alternatives. Copyright © 2020, The Authors. All rights reserved.

关键词： Gradient methods

Resource Provisioning for Virtual Network Function Deployment with In-Subnetwork Processing

学校读者我要写书评

暂无评论

Resource Provisioning for Virtual Network Function Deploymen...

IEEE Workshop on Signal Processing Advances in Wireless Communications (SPAWC)

作者： Navid Reyhanian Hamid Farmanbar Soheil Mohajer Zhi-Quan Luo Department of Electrical and Computer Engineering University of Minnesota Minneapolis MN USA Huawei Canada Research Center Ottawa Canada Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China

ISBN: (数字)9781728154787

ISBN: (纸本)9781728154794

Network Function Virtualization (NFV) is an efficient approach to simplify and accelerate the deployment of diverse network services. In this paper, we study the problem of link capacity and server purchase to provision the network which enables joint Virtual Network Function (VNF) placement on servers and traffic engineering in a network spanning multiple subnetworks. Each subnetwork is owned and controlled by a different administrator. Given flow demands, we formulate the problem from a sparse optimization perspective and propose an efficient approach based on iteratively solving a sequence of group LASSO problems. Using purchased link capacities and servers, each subnetwork is able to carry out VNF placement and traffic engineering locally to meet flow demands. A scalable and decentralized approach based on the proximal Alternating Direction Method of Multipliers (ADMM) is proposed for this problem. The distributed optimization can be locally solved with minimum information shared with other administrators. Extensive numerical evaluations show the efficiency of our approach against existing work.

关键词： Servers Optimization Routing Convex functions Minimization Conferences Signal processing

Resource Reservation in Backhaul and Radio Access Network with Uncertain User Demands

学校读者我要写书评

暂无评论

Resource Reservation in Backhaul and Radio Access Network wi...

IEEE Workshop on Signal Processing Advances in Wireless Communications (SPAWC)

作者： Navid Reyhanian Hamid Farmanbar Zhi-Quan Luo Department of Electrical and Computer Engineering University of Minnesota Minneapolis MN USA Huawei Canada Research Center Ottawa Canada Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China

ISBN: (数字)9781728154787

ISBN: (纸本)9781728154794

We consider the problem of joint resource reservation in the backhaul and Radio Access Network (RAN) based on the user demand statistics and network availability. The goal is to maximize the sum of expected traffic flow rates, subject to link and radio node budget constraints, while minimizing the expected outage of wireless channels. The formulated problem turns out to be non-convex and difficult to solve to global optimality. We propose an efficient Block Coordinate Descent (BCD) algorithm to approximately solve the problem. To optimize with respect to each block of variables, a distributed decomposition approach is proposed. Simulation results verify the efficiency and the efficacy of our proposed approach against two heuristic algorithms.

关键词： Wireless communication Radio access networks Routing Signal processing algorithms Data centers Probability density function Conferences

Convert widespread paraelectric perovskite to ferroelectrics

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Hongwei Tang, Fujie Stengel, Massimiliano Xiang, Hongjun An, Qi Low, Tony Wu, Xifan Department of Physics Temple University PhiladelphiaPA19122 United States Department of Electrical and Computer Engineering University of Minnesota MinneapolisMN55455 United States Campus UAB Bellaterra08193 Spain ICREA-Institució Catalana de Recerca i Estudis Avançats Barcelona08010 Spain State Key Laboratory of Surface Physics Department of Physics Fudan University Shanghai200438 China Department of Chemical and Materials Engineering University of Nevada-Reno RenoNV89557 United States Institute for Computational Molecular Science Temple University PhiladelphiaPA19122 United States

While nature provides a plethora of perovskite materials, only a few exhibits large ferroelectricity and possibly multiferroicity. The majority of perovskite materials have the non-polar CaTiO3(CTO) structure, limiting the scope of their applications. Based on effective Hamiltonian model as well as first-principles calculations, we propose a general thin-film design method to stabilize the functional BiFeO3(BFO)-type structure, which is a common metastable structure in widespread CaTiO3-type perovskite oxides. It is found that the improper antiferroelectricity in CTO-type perovskite and fer-roelectricity in BFO-type perovskite have distinct dependences on mechanical and electric boundary conditions, both of which involve oxygen octahedral rotation and tilt. The above difference can be used to stabilize the highly polar BFO-type structure in many CTO-type perovskite materials. © 2022, CC BY.

关键词： Perovskite

Reproducible brain-wide association studies require thousands of individuals (Mar, 10.1038/s41586-022-04492-9, 2022)

学校读者我要写书评

暂无评论

NATURE 2022年第7911期605卷 E11-E11页

作者： Marek, Scott Tervo-Clemmens, Brenden Calabro, Finnegan J. Montez, David F. Kay, Benjamin P. Hatoum, Alexander S. Donohue, Meghan Rose Foran, William Miller, Ryland L. Hendrickson, Timothy J. Malone, Stephen M. Kandala, Sridhar Feczko, Eric Miranda-Dominguez, Oscar Graham, Alice M. Earl, Eric A. Perrone, Anders J. Cordova, Michaela Doyle, Olivia Moore, Lucille A. Conan, Gregory M. Uriarte, Johnny Snider, Kathy Lynch, Benjamin J. Wilgenbusch, James C. Pengo, Thomas Tam, Angela Chen, Jianzhong Newbold, Dillan J. Zheng, Annie Seider, Nicole A. Van, Andrew N. Metoki, Athanasia Chauvin, Roselyne J. Laumann, Timothy O. Greene, Deanna J. Petersen, Steven E. Garavan, Hugh Thompson, Wesley K. Nichols, Thomas E. Yeo, B. T. Thomas Barch, Deanna M. Luna, Beatriz Fair, Damien A. Dosenbach, Nico U. F. Department of Psychiatry Washington University School of Medicine St Louis MO USA Department of Neurology Washington University School of Medicine St Louis MO USA Department of Psychological and Brain Sciences Washington University in St Louis St Louis MO USA Department of Psychiatry Massachusetts General Hospital Harvard Medical School Boston MA USA Department of Psychology University of Pittsburgh Pittsburgh PA USA Department of Psychiatry University of Pittsburgh Pittsburgh PA USA Department of Bioengineering University of Pittsburgh Pittsburgh PA USA Department of Biomedical Engineering Washington University in St Louis St Louis MO USA Department of Radiology Washington University School of Medicine St Louis MO USA Department of Neurological Surgery Washington University School of Medicine St Louis MO USA Program in Occupational Therapy Washington University School of Medicine St Louis MO USA Department of Pediatrics Washington University School of Medicine St Louis MO USA University of Minnesota Informatics Institute University of Minnesota Minneapolis MN USA Department of Psychology University of Minnesota Minneapolis MN USA Masonic Institute for the Developing Brain University of Minnesota Medical School Minneapolis MN USA Department of Pediatrics University of Minnesota Medical School Minneapolis MN USA Department of Psychiatry Oregon Health and Science University Portland OR USA Minnesota Supercomputing Institute University of Minnesota Minneapolis MN USA Institute of Child Development University of Minnesota Medical School Minneapolis MN USA Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore Centre for Sleep and Cognition National University of Singapore Singapore Singapore Centre for Translational MR Research National University of Singapore Singapore Singapore N.1 Institute for Health Institute for Digital Medicine National University of Singapore Singapore Sing

Gauss-Newton unrolled neural networks and data-driven priors for regularized PSSE with robustness

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Yang, Qiuling Sadeghi, Alireza Wang, Gang Giannakis, Georgios B. Sun, Jian State Key Lab of Intelligent Control and Decision of Complex Systems School of Automation Beijing Institute of Technology Beijing100081 China Department of Electrical and Computer Engineering University of Minnesota MinneapolisMN55455 United States

Renewable energy sources, elastic loads, and purposeful manipulation of meter readings challenge the monitoring and control of today’s power systems (PS). In this context, fast and robust state estimation (SE) is timely and of major importance to maintain a comprehensive view of the system in real time. Conventional PSSE solvers typically entail minimizing a nonlinear and nonconvex least-squares cost using e.g., the Gauss-Newton method. Those iterative solvers however, are sensitive to initialization, and may converge to local minima. To overcome these hurdles, the present paper adapts and leverages recent advances on image denoising to introduce a PSSE approach with a regularizer capturing a deep neural network (DNN) prior. For the resultant regularized PSSE objective, a "Gauss-Newton-type" alternating minimization solver is developed first. To accommodate real-time monitoring, a novel end-to-end DNN is constructed subsequently by unrolling the proposed alternating minimization solver. The deep PSSE architecture can further account for the power network topology through a graph neural network (GNN) based prior. To further endow the physics-based DNN with robustness against bad data, an adversarial DNN training method is put forth. Numerical tests using real load data on the IEEE 118-bus benchmark system showcase the improved estimation and robustness performance of the proposed scheme compared with several state-of-the-art alternatives. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks