检索结果-内蒙古大学图书馆

A Robust Deep Learning Method with Uncertainty Estimation for the Pathological Classification of Renal Cell Carcinoma based on CT Images

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yao, Ni Hu, Hang Chen, Kaicong Zhao, Chen Guo, Yuan Li, Boya Nan, Jiaofen Li, Yanting Han, Chuang Zhu, Fubao Zhou, Weihua Tian, Li School of Computer and Communication Engineering Zhengzhou University of Light Industry Henan Zhengzhou450002 China Department of Medical Imaging State Key Laboratory of Oncology in South China Guangdong Provincial Clinical Research Center for Cancer Collaborative Innovation Center for Cancer Medicine Sun Yat-Sen University Cancer Center Guangdong Guangzhou510060 China Department of Applied Computing Michigan Technological University HoughtonMI United States Center for Biocomputing and Digital Health Institute of Computing and Cybersystems Health Research Institute Michigan Technological University HoughtonMI United States Department of Radiology The First People's Hospital of Guangzhou Guangdong Guangzhou510180 China

Objectives To develop and validate a deep learning-based diagnostic model incorporating uncertainty estimation so as to facilitate radiologists in the preoperative differentiation of the pathological subtypes of renal cell carcinoma (RCC) based on CT images. Methods Data from 668 consecutive patients, pathologically proven RCC, were retrospectively collected from Center 1. By using five-fold cross-validation, a deep learning model incorporating uncertainty estimation was developed to classify RCC subtypes into clear cell RCC (ccRCC), papillary RCC (pRCC), and chromophobe RCC (chRCC). An external validation set of 78 patients from Center 2 further evaluated the model's performance. Results In the five-fold cross-validation, the model's area under the receiver operating characteristic curve (AUC) for the classification of ccRCC, pRCC, and chRCC was 0.868 (95% CI: 0.826-0.923), 0.846 (95% CI: 0.812-0.886), and 0.839 (95% CI: 0.802-0.88), respectively. In the external validation set, the AUCs were 0.856 (95% CI: 0.838-0.882), 0.787 (95% CI: 0.757-0.818), and 0.793 (95% CI: 0.758-0.831) for ccRCC, pRCC, and chRCC, respectively. Conclusions The developed deep learning model demonstrated robust performance in predicting the pathological subtypes of RCC, while the incorporated uncertainty emphasized the importance of understanding model confidence, which is crucial for assisting clinical decision-making for patients with renal tumors. Clinical relevance statement Our deep learning approach, integrated with uncertainty estimation, offers clinicians a dual advantage: accurate RCC subtype predictions complemented by diagnostic confidence references, promoting informed decision-making for patients with RCC. © 2023, CC BY.

关键词： Cells

Univariate Forecasting of time series data Based on Seq2Seq Model 4

学校读者我要写书评

暂无评论

Univariate Forecasting of time series data Based on Seq2Seq ...

4th International Conference on Electronic Information Technology and computer engineering, EITCE 2020

作者： Guo, Yanwen Wu, Tao Liu, Yanbing Jiang, Yixin Department of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing China School of Cybersecurity and Information Law Chongqing University of Posts and Telecommunications Chongqing China Chongqing Engineering Laboratory of Internet and Information Sccurity Chongqing University of Posts and Telecommunications Chongqing China Electric Power Research Institute Csg Guangdong Provincial Key Laboratory of Power System Network Security Guangzhou China

ISBN: (纸本)9781450387811

With the release of country's 2020 new infrastructure projects, the increasing demand for electricity in various projects has further promoted the intelligent upgrade of the power grid. The processing of grid data at the edge has become one of the future trends of next generation grid. In the field of power grid edge time series data prediction, the inability to obtain the variable characteristics of other systems and the uncertainty of the length of various edge input data, this paper proposes a Seq2seq prediction model based on a Bi-directional Long Short Term Memory (Bi-LSTM). Bi-LSTM can make full use of contextual information to achieve a more accurate prediction effect. The Seq2Seq model can process variable-length sequence data. In this paper, Seq2Seq model and BI-LSTM model are combined to solve the problem of uncertain edge length and univariate prediction. At the same time, in order to solve the information loss caused by Seq2Seq in the encoding stage, the Attention mechanism is introduced, it can compute the allocation of resources to more important tasks while addressing the problem of information overload. Through experimental research, it is found that the proposed ABSeq2Seq model is superior to other classic models in univariate prediction, and it has the better generalization for various types of data. We have proved its effectiveness in the single variable power prediction. © 2020 ACM.

关键词： Time series

Deep Metric Learning for Few-Shot Image Classification: A Review of Recent Developments

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Xiaoxu Yang, Xiaochen Ma, Zhanyu Xue, Jing-Hao School of Computer and Communication Lanzhou University of Technology China Pattern Recognition and Intelligent System Laboratory School of Information and Communication Engineering Beijing University of Posts and Telecommunications China School of Mathematics and Statistics University of Glasgow United Kingdom Department of Statistical Science University College London United Kingdom

Few-shot image classification is a challenging problem that aims to achieve the human level of recognition based only on a small number of training images. One main solution to few-shot image classification is deep metric learning. These methods, by classifying unseen samples according to their distances to few seen samples in an embedding space learned by powerful deep neural networks, can avoid overfitting to few training images in few-shot image classification and have achieved the state-of-the-art performance. In this paper, we provide an up-to-date review of deep metric learning methods for few-shot image classification from 2018 to 2022 and categorize them into three groups according to three stages of metric learning, namely learning feature embeddings, learning class representations, and learning distance measures. With this taxonomy, we identify the novelties of different methods and problems they face. We conclude this review with a discussion on current challenges and future trends in few-shot image classification. © 2021, CC BY.

关键词： Image classification

Deep learning insights into spatial patterns of stable isotopes in Iran’s precipitation: a novel approach to climatological mapping

学校读者我要写书评

暂无评论

Isotopenpraxis Isotopes in Environmental and Health Studies 2024年第4期60卷

作者： Mojtaba Heydarizad Rogert Sori Masoud Minaei Hamid Ghalibaf Mohammadabadi Elham Mahdipour a State Key Laboratory of Marine Geology Tongji University Shanghai People’s Republic of China b Environmental Physics Laboratory (EPhysLab) Centro de Investigación Mariña Universidade de Vigo Ourense Spain c Department of Geography Faculty of Letters and Humanities Ferdowsi University of Mashhad Mashhad Irand Geographic Information Science/System and Remote Sensing Laboratory (GISSRS: Lab) Faculty of Letters and Humanities Ferdowsi University of Mashhad Mashhad Iran e Department of Computer Engineering Ferdowsi University of Mashhad Mashhad Iran f Computer Engineering Department Khavaran Institute of Higher Education Mashhad Iran

ABSTRACTStable isotope techniques are precise methods for studying various aspects of hydrology, such as precipitation characteristics. However, understanding the variations in the stable isotope content in precipitation is challenging in Iran due to numerous climatic and geographic factors. To address this, forty-two precipitation sampling stations were selected across Iran to assess the fractional importance of these climatic and geographic parameters influencing stable isotopes. Additionally, deep learning models were employed to simulate the stable isotope content, with missing data initially addressed using the predictive mean matching (PMM) method. Subsequently, the recursive feature elimination (RFE) technique was applied to identify influential parameters impacting Iran’s precipitation stable isotope content. Following this, long short-term memory (LSTM) and deep neural network (DNN) models were utilized to predict stable isotope values in precipitation. Interpolated maps of these values across Iran were developed using inverse distance weighting (IDW), while an interpolated reconstruction error (RE) map was generated to quantify deviations between observed and predicted values at study stations, offering insights into model precision. Validation using evaluation metrics demonstrated that the model based on DNN exhibited higher accuracy. Furthermore, RE maps confirmed acceptable accuracy in simulating the stable isotope content, albeit with minor weaknesses observed in simulation maps. The methodology outlined in this study holds promise for application in regions worldwide characterized by diverse climatic conditions.

关键词： Deep learning DNN isotope hydrology LSTM Iran precipitation stable isotopes simulation

Cloud detection from visual band of satellite image based on variance of fractal dimension

学校读者我要写书评

暂无评论

Journal of systems engineering and Electronics 2019年第3期30卷 485-491页

作者： TIAN Pingfang GUANG Qiang LIU Xing School of Computer Science and Technology Wuhan University of Science and Technology Wuhan 430065 China Key Laboratory of Intelligent Information Processing and Real-Time Industrial System in Hubei Province Wuhan 430065 China Institute of Big Data Science and Engineering Wuhan University of Science and Technology Wuhan 430065 China Key Laboratory of Rich-Media Knowledge Organization and Service of Digital Publishing Content National Press and Publication Administration Beijing 100038 China

Cover ratio of cloud is a very important factor which affects the quality of a satellite image, therefore cloud detection from satellite images is a necessary step in assessing the image quality. The study on cloud detection from the visual band of a satellite image is developed. Firstly, we consider the differences between the cloud and ground including high grey level, good continuity of grey level, area of cloud region, and the variance of local fractal dimension (VLFD) of the cloud region. A single cloud region detection method is proposed. Secondly, by introducing a reference satellite image and by comparing the variance in the dimensions corresponding to the reference and the tested images, a method that detects multiple cloud regions and determines whether or not the cloud exists in an image is described. By using several Ikonos images, the performance of the proposed method is demonstrated.

关键词： cloud detection visual image satellite image variance of local fractal dimension (VLFD)

Resource allocation for millimeter-wave train-ground communications in high-speed railway scenarios

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhang, Xiangfei Niu, Yong Mao, Shiwen Cai, Yunlong He, Ruisi Ai, Bo Zhong, Zhangdui Liu, Yiru State Key Laboratory of Rail Traffic Control and Safety Frontiers Science Center for Smart High-speed Railway System Beijing Jiaotong University Beijing100044 China National Mobile Communications Research Laboratory Southeast University Nanjing211189 China Department of Electrical and Computer Engineering Auburn University AuburnAL36849-5201 United States Department of ISEE Zhejiang University China State Key Laboratory of Rail Traffic Control and Safety Beijing Engineering Research Center of High-speed Railway Broadband Mobile Communications School of Electronic and Information Engineering Beijing Jiaotong University Beijing100044 China

With the development of wireless communication, higher requirements arise for train-ground wireless communications in high-speed railway (HSR) scenarios. The millimeter-wave (mm-wave) frequency band with rich spectrum resources can provide users in HSR scenarios with high performance broadband multimedia services, while the full-duplex (FD) technology has become mature. In this paper, we study train-ground communication system performance in HSR scenarios with mobile relays (MRs) mounted on rooftop of train and operating in the FD mode. We formulate a nonlinear programming problem to maximize network capacity by allocation of spectrum resources. Then, we develop a sequential quadratic programming (SQP) algorithm based on the Lagrange function to solve the bandwidth allocation optimization problem for track-side base station (BS) and MRs in this mm-wave train-ground communication system. Extensive simulation results demonstrate that the proposed SQP algorithm can effectively achieve high network capacity for train-ground communication in HSR scenarios while being robust to the residual self-interference (SI). Copyright © 2021, The Authors. All rights reserved.

关键词： Resource allocation

Single-mode fiber coated with zinc oxide (ZnO) nanorods for H2 gas sensor applications

学校读者我要写书评

暂无评论

Single-mode fiber coated with zinc oxide (ZnO) nanorods for ...

2019 IEEE International Conference on Sensors and Nanotechnology, SENSORS and NANO 2019

作者： Saidin, Norazlina Idris, Nur Farahi Yaacob, Mohd Hanif Harun, Sulaiman Wadi Md Ralib, Aliza Aini Hasbullah, Nurul Fadzlin Department of Electrical and Computer Engineering International Islamic University Malaysia Kuala Lumpur53100 Malaysia Photonics System Engineering Laboratory University Putra Malaysia Seri Kembangan Selangor43300 Malaysia Department of Electrical Engineering University of Malaya Kuala Lumpur50603 Malaysia

ISBN: (纸本)9781538656198

A Hydrogen (H2) gas sensor was successfully developed using optical fiber coated with Zinc Oxide (ZnO) nanorods. The single-mode fiber (SMF) used as a sensing device has been prepared by etching the SMF fiber and coated with ZnO nanorods. The etching of the fiber was performed using hydrofluoric acid (HF) to enhance the evanescent field around the fiber core. The ZnO nanorods were prepared by hydrothermal method through seeding and growth solution technique. The diameter of cladding and core are 125 µm and 8 µm, respectively, before etching and goes down to 11 µm after etching. Around 2 cm of ZnO nanorods were coated in the middle of the etched fiber. The sensing layer was characterized through Scanning Electron Microscopy (SEM), Energy Dispersive X-Ray (EDX) and X-Ray Diffraction (XRD) to verify the properties of ZnO. The developed sensor’s response and recovery time were observed to be 7 min and 3 min, respectively, for a low concentration of 0.25% H2 gas. The aim of this study is to understand the gas sensing properties towards the spectral intensity variations in etched optical fiber coated with ZnO nanorods. © 2019 IEEE.

关键词： Etching

Enhancing wide-band trajectory tracking and finite-time control of Dielectric Elastomer Actuators using Nonsingular terminal sliding mode control scheme

学校读者我要写书评

暂无评论

Mechatronics 2025年 109卷

作者： Kassim, Shakiru Olajide Zou, Jiang Behnamgol, Vahid Vaziri, Vahid Aphale, Sumeet S. Group School of Engineering University of Aberdeen ScotlandAB24 3UE United Kingdom The state Key Laboratory of Mechanical System and Vibration Shanghai Jiao Tong University Shanghai China The Robotics Institute School of Mechanical Engineering Shanghai Jiao Tong University Shanghai China Renewable Energy Research Centre Damavand Branch Islamic Azad University Tehran Iran Department of Electrical and Computer Engineering The Federal Polytechnic Damaturu Yobe State 620221 Nigeria

The utilization of Dielectric Elastomer Actuators (DEAs) in soft robotics is becoming increasingly popular due to their distinctive properties. However, controlling these actuators is challenging due to their nonlinear nature. This study addresses precise control of DEAs while minimizing nonlinear effects over a wide frequency range using a Sliding Mode Control (SMC) scheme. Traditional SMC approaches face two main issues: they do not ensure finite-time convergence during the sliding phase and compromise tracking accuracy due to approximations used to reduce chattering. This work presents a non-singular terminal sliding mode control combined with an enhanced boundary layer switching function to improve precision and finite-time stability. Using stability analysis, extensive MATLAB/SIMULINK simulations, and experimental validation on a conical DEA across diverse reference trajectories and its resilience against uncertainties, the proposed scheme demonstrates superior performance in trajectory tracking compared to PID-based and conventional SMC schemes. The results highlight the scheme's effectiveness in high-frequency trajectory tracking and its robustness to possible uncertainties and disturbances, offering a robust theoretical framework for dielectric elastomer actuator control and a promising approach for advancing soft robotic technologies. © 2025 The Authors

关键词： Intelligent robots

DeceFL: A principled decentralized federated learning framework

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Yuan, Ye Liu, Jun Jin, Dou Yue, Zuogong Chen, Ruijuan Wang, Maolin Sun, Chuan Xu, Lei Hua, Feng He, Xin Yi, Xinlei Yang, Tao Zhang, Hai-Tao Sui, Shaochun Ding, Han School of Artificial Intelligence and Automation Huazhong University of Science and Technology School of Mechanical Science and Engineering Huazhong University of Science and Technology Department of Applied Mathematics University of Waterloo Canada State Key Laboratory of Synthetical Automation for Process Industries Northeastern University United States School of Electrical Engineering and Computer Science Digital Futures Kth Royal Institute of Technology Co. Ltd.

Traditional machine learning relies on a centralized data pipeline, i.e., data are provided to a central server for model training. In many applications, however, data are inherently fragmented. Such a decentralized nature of these databases presents the biggest challenge for collaboration: sending all decentralized datasets to a central server raises serious privacy concerns. Although there has been a joint effort in tackling such a critical issue by proposing privacy-preserving machine learning frameworks, such as federated learning, most state-of-the-art frameworks are built still in a centralized way, in which a central client is needed for collecting and distributing model information (instead of data itself) from every other client, leading to high communication pressure and high vulnerability when there exists a failure at or attack on the central client. Here we propose a principled decentralized federated learning algorithm (DeceFL), which does not require a central client and relies only on local information transmission between clients and their neighbors, representing a fully decentralized learning framework. It has been further proven that every client reaches the global minimum with zero performance gap and achieves the same convergence rate O(1/T) (where T is the number of iterations in gradient descent) as centralized federated learning when the loss function is smooth and strongly convex. Finally, the proposed algorithm has been applied to a number of applications to illustrate its effectiveness for both convex and nonconvex loss functions, demonstrating its applicability to a wide range of real-world medical and industrial applications. Copyright © 2021, The Authors. All rights reserved.

关键词： Gradient methods