检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2022年第4期3卷 595-608页

作者： Fan, Feng-Lei Wang, Dayang Guo, Hengtao Zhu, Qikui Yan, Pingkun Wang, Ge Yu, Hengyong Rensselaer Polytechnic Institute Department of Biomedical Engineering TroyNY12180 United States University of Massachusetts Department of Electrical and Computer Engineering LowellMA01854 United States

In established network architectures, shortcut connections are often used to take the outputs of earlier layers as additional inputs to later layers. Despite the extraordinary effectiveness of shortcuts, there remain open questions on the mechanism and characteristics. For example, why are shortcuts powerful? Why do shortcuts generalize well? In this article, we investigate the expressivity and generalizability of a novel sparse shortcut topology. First, we demonstrate that this topology can empower a one-neuron-wide deep network to approximate any univariate continuous function. Then, we present a novel width-bounded universal approximator in contrast to depth-bounded universal approximators and extend the approximation result to a family of equally competent networks. Furthermore, with generalization bound theory, we show that the proposed shortcut topology enjoys excellent generalizability. Finally, we corroborate our theoretical analyses by comparing the proposed topology with popular architectures, including ResNet and DenseNet, on well-known benchmarks and perform a saliency map analysis to interpret the proposed topology. Our work helps understand the role of shortcuts and suggests further opportunities to innovate neural architectures. © 2020 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Segmentation Approach for Enhanced DLP Bioprinting 12

Segmentation Approach for Enhanced DLP Bioprinting

引用

Laser 3D Manufacturing XII 2025

作者： Babayigit, Ceren Tavares-Negrete, Jorge Alfonso Esfandyarpour, Rahim Boyraz, Ozdal Department of Electrical Engineering and Computer Science The University of California IrvineCA92697 United States Department of Biomedical Engineering The University of California IrvineCA92697 United States Department of Mechanical and Aerospace Engineering The University of California IrvineCA92697 United States

ISBN: (纸本)9781510684560

Achieving high-resolution outputs in Digital Light Processing (DLP) bioprinting requires precise control over photocrosslinking, particularly for dense, intricate models with fine gaps. Over-crosslinking is affected by factors such as light scattering, limited depth of focus, and molecular diffusion. Traditionally, these issues are managed by altering the material and light source. This study presents a novel segmentation approach designed to address these challenges by leveraging the inherent diffusion characteristics of the DLP system rather than altering its core components. The generation and diffusion of free radicals increase with light intensity and the density of the illuminated area, which alters the risk of over-crosslinking in these regions. The proposed technique involves segmenting the photomask into discrete regions and employing an on-off activation strategy to enhance control over photopolymerization. By strategically segmenting the mask and alternating the activation of adjacent segments, the technique effectively manages the distribution and concentration of free radicals and allows time for the termination phase of photopolymerization. It creates spatial gaps, helps manage free radical diffusion, prevents unwanted combinations, and filling of small gaps, thus improving resolution. Demonstrated by printing a bronchial model, this approach achieved a 100 µm gap between bronchial branches at a 14.43x compression ratio to anatomical size. The live-dead assay shows sustained cell viability of up to 90% for GelMA and up to 85% for GelMA-PEGDA constructs across seven days. This study shows that the segmentation strategy significantly reduces over-crosslinking, advancing DLP bioprinting for applications requiring precise control over complex features and detailed features. © 2025 SPIE.

关键词： Photopolymerization

来源：评论

学校读者我要写书评

暂无评论

Quaternion double random phase encoding for privacy-preserving cancelable biometrics

引用

Multimedia Tools and Applications 2024年第37期83卷 85065-85078页

作者： Nasr, Mahmoud Piórkowski, Adam Abd El-Samie, Fathi E. Department of Biocybernetics and Biomedical Engineering AGH University of Krakow Krakow Poland Faculty of Engineering and Technology Future University in Egypt Cairo Egypt Computer Vision Sano Centre for Computational Medicine Kraków Poland Department of Electronics and Electrical Communications Engineering Faculty of Electronic Engineering Manoufia University Menouf32952 Egypt Department of Information Technology College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia

Biometrics has become a fundamental aspect in the constantly-changing field of security solutions. Nevertheless, the ongoing risk of hacking and the possibility of biometric data being compromised in security breaches emphasize the urgent need for inventive and reliable biometric authentication methods. Introducing cancelable biometrics is an advanced approach to enhance the security of biometric templates. Our research paper presents a unique variation of this approach that depends on the powerful Double Random Phase Encoding (DRPE) algorithm. By incorporating an auxiliary input biometric template and using quaternion mathematics, we get an exceptional level of performance that surpasses previous performance standards. The cancelable biometric recognition system we present demonstrates exceptional performance in simulation experiments, demonstrating its ability to remain robust even in the presence of noise. More precisely, the proposed cancelable face recognition system attains a notable area under the Receiver Operating Characteristic (ROC) curve of 0.994 and an exceptionally-low Equal Error Rate (EER) of 0.00174. The efficiency of our solution is amazing, with an average time of 0.056263 seconds for each generated secure template. This demonstrates both speed and efficacy in addressing current security concerns. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

Deep Learning of Biomechanical Dynamics With Spatial Variability for Lifestyle Management 4

Deep Learning of Biomechanical Dynamics With Spatial Variabi...

引用

4th IEEE Global Conference on Life Sciences and Technologies, LifeTech 2022

作者： Gangadharan, Kiirthanaa Zhang, Qingxue Purdue School of Engineering and Technology Department of Electrical and Computer Engineering IndianapolisIN United States Purdue School of Engineering and Technology Department of Electrical and Computer Engineering Department of Biomedical Engineering IndianapolisIN United States

ISBN: (纸本)9781665419048

Human Physical Activity Detection (PAD) is essential for advanced lifestyle management. Nowadays, many devices like smartwatches and phones have motion sensors. And the placement of devices, especially phones, is diverse. Understanding the optimal sensor location is important for robust PAD applications. However, the biomechanical dynamics are highly complex, making signal processing very challenging. In this study, we propose to leverage deep learning to perform intelligent PAD, and comprehensively compare seven different body locations for optimal sensor placement recommendation. Multi-stage deep learning has been proposed and developed, which firstly leverages convolutional layers to abstract spatial features from the motion data, and then generates the physical activity type prediction with fully connected layers. Evaluated on the real-world database, the proposed deep learning model is very effective on PAD tasks, and successfully determines the thigh location as the optimal sensor placement method. This study will greatly advance deep learning-driven biomechanical dynamics mining for advanced lifestyle management. © 2022 IEEE.

关键词： Wearable sensors

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Meets Sparse Regularization: A Signal Processing Perspective

引用

arXiv 2023年第6期40卷 63页

作者： Parhi, Rahul Nowak, Robert D. The Biomedical Imaging Group École Polytechnique Fédérale de Lausanne Lausanne Switzerland The Department of Electrical and Computer Engineering University of Wisconsin–Madison MadisonWI United States

Deep learning has been wildly successful in practice and most state-of-the-art machine learning methods are based on neural networks. Lacking, however, is a rigorous mathematical theory that adequately explains the amazing performance of deep neural networks. In this article, we present a relatively new mathematical framework that provides the beginning of a deeper understanding of deep learning. This framework precisely characterizes the functional properties of neural networks that are trained to fit to data. The key mathematical tools which support this framework include transform-domain sparse regularization, the Radon transform of computed tomography, and approximation theory, which are all techniques deeply rooted in signal processing. This framework explains the effect of weight decay regularization in neural network training, the use of skip connections and low-rank weight matrices in network architectures, the role of sparsity in neural networks, and explains why neural networks can perform well in high-dimensional problems. Copyright © 2023, The Authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Soft,full Wheatstone bridge 3D pressure sensors for cardiovascular monitoring

引用

npj Flexible Electronics 2024年第1期8卷 910-917页

作者： Yoonseok Park Haiwen Luan Kyeongha Kwon Ted S.Chung Seyong Oh Jae-Young Yoo Gooyoon Chung Junha Kim Suhyeon Kim Sung Soo Kwak Junhwan Choi Hoang-Phuong Phan Seonggwang Yoo Hyoyoung Jeong Jaeho Shin Sang Min Won Hong-Joon Yoon Yei Hwan Jung John A.Rogers Department of Advanced Materials Engineering for Information and Electronics Integrated Education Institute for Frontier Science&Technology(BK21 Four)Kyung Hee UniversityYongin-siGyeonggi-do 17104Republic of Korea Department of Mechanical and Aerospace Engineering University of California San DiegoLa JollaCA 92093USA School of Electrical Engineering Korea Advanced Institute of Science and TechnologyDaejeonRepublic of Korea Department of Mechanical Engineering Columbia UniversityNew YorkNYUSA Division of Electrical Engineering Hanyang University ERICAAnsan 15588Republic of Korea Department of Semiconductor Convergence Engineering Sungkyunkwan UniversitySuwonRepublic of Korea Biomedical Research Division Korea Institute of Science and TechnologySeoul 02792Republic of Korea Department of Chemical Engineering Dankook UniversityYonginGyeonggi-do 16890Republic of Korea School of Mechanical and Manufacturing Engineering Tyree Institute of Health EngineeringUniversity of New South WalesSydneyNSW 2052Australia Querrey Simpson Institute for Bioelectronics Northwestern University EvanstonEvanstonIL 60208USA Department of Electrical and Computer Engineering University of California DavisDavisCA 95616USA Department of Electrical and Computer Engineering Sungkyunkwan University2066Seobu-roJangan-guSuwon-siGyeonggi-do 16419Republic of Korea Department of Electronic Engineering Gachon UniversitySeongnamsiGyeonggi-do 13120Republic of Korea Department of Electronic Engineering Hanyang UniversitySeoulRepublic of Korea Department of Materials Science and Engineering Biomedical EngineeringNeurological SurgeryChemistryMechanical EngineeringElectrical Engineering and Computer Science Simpson Querrey Institute and Feinberg Medical SchoolCenter for Bio-Integrated ElectronicsNorthwestern University EvanstonEvanstonIL 60208USA

Variations in parameters associated with the ambient environment can introduce noise in soft,body-worn *** example,many piezoresistive pressure sensors exhibit a high degree of sensitivity to fluctuations in temperature,thereby requiring active compensation *** research presented here addresses this challenge with a multilayered 3D microsystem design that integrates four piezoresistive sensors in a full-Wheatstone bridge *** optimized layout of the sensors relative to the neutral mechanical plane leads to both an insensitivity to temperature and an increased sensitivity to pressure,relative to previously reported devices that rely on similar operating *** this 3D pressure sensor into a soft,flexible electronics platform yields a system capable of real-time,wireless measurements from the surface of the *** above the radial and carotid arteries yields high-quality waveforms associated with pulsatile blood flow,with quantitative correlations to blood *** results establish the materials and engineering aspects of a technology with broad potential in remote health monitoring.

关键词： Wheat bridge thereby

来源：评论

学校读者我要写书评

暂无评论

Composite Power System Reliability Assessment Considering Uncertainty of Electric Vehicle Charging and PV Power Generation

Composite Power System Reliability Assessment Considering Un...

引用

2023 IEEE Industry Applications Society Annual Meeting, IAS 2023

作者： Thapa, Jitendra Olowolaju, Joshua Livani, Hanif Benidris, Mohammed Michigan State University Department of Electrical and Computer Engineering East LansingMI48824 United States University of Nevada Department of Electrical & Biomedical Engineering RenoNV89557 United States

ISBN: (纸本)9798350320169

Modern power systems are facing several challenges with the increasing penetration of inverter-based resources (IBRs) and electric vehicles (EVs). The problem arises due to inter-mittency and uncertainties associated with IBR-based resources and electric vehicles. It is indispensable from the perspective of power system planning and operation to consider these factors in reliability assessment. The reliability assessment of modern power systems that are heavily penetrated with renewable generators should consider their uncertainties to quantify the reliability index and include EVs charging load in their framework as well. Therefore, this paper proposes an approach for composite power system reliability assessment combined with the probabilistic prediction interval of PV power along with the integration of electric vehicles. An electric vehicle load model developed considering the different locations of charging stations, types of EV, and drivers' behavior is integrated. Finally, a sequential Monte Carlo simulation is used to determine the range of the reliability index after an interval forecast of the PV and EV load model is developed. The impact of different levels of EV penetration on the reliability of power system integrated with PV is also investigated. The significance of this work is to provide an accurate reliability assessment framework that takes PV uncertainties and electric vehicles into consideration. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Organoid intelligence:Integration of organoid technology and artificial intelligence in the new era of in vitro models

引用

Medicine in Novel Technology and Devices 2024年第1期21卷 39-44页

作者： Huaiyu Shi Andrew Kowalczewski Danny Vu Xiyuan Liu Asif Salekin Huaxiao Yang Zhen Ma Department of Biomedical&Chemical Engineering Syracuse UniversitySyracuseNYUSA BioInspired Institute for Material and Living Systems Syracuse UniversitySyracuseNYUSA Department of Mechanical&Aerospace Engineering Syracuse UniversitySyracuseNYUSA Department of Electrical Engineering&Computer Science Syracuse UniversitySyracuseNYUSA Department of Biomedical Engineering University of North TexasDentonTXUSA

Organoid Intelligence ushers in a new era by seamlessly integrating cutting-edge organoid technology with the power of artificial ***,three-dimensional miniature organ-like structures cultivated from stem cells,offer an unparalleled opportunity to simulate complex human organ systems in *** the convergence of organoid technology and AI,researchers gain the means to accelerate discoveries and insights across various *** intelligence algorithms enable the comprehensive analysis of intricate organoid behaviors,intricate cellular interactions,and dynamic responses to *** synergy empowers the development of predictive models,precise disease simulations,and personalized medicine approaches,revolutionizing our understanding of human development,disease mechanisms,and therapeutic *** Intelligence holds the promise of reshaping how we perceive in vitro modeling,propelling us toward a future where these advanced systems play a pivotal role in biomedical research and drug development.

关键词： Organoids Machinelearning Deep learning Artificial intelligence Stem cells

来源：评论

学校读者我要写书评

暂无评论

Adaptive learning approach to improve generalization performance of a domain-aware CNN-based ideal model observer

Adaptive learning approach to improve generalization perform...

引用

Medical Imaging 2024: Image Perception, Observer Performance, and Technology Assessment

作者： Bogdanovic, Nebojsa Abbey, Craig K. Yang, Yongyi Brankov, Jovan G. Department of Electrical and Computer Engineering Illinois Institute of Technology United States Department of Psychological and Brain Sciences University of California Santa Barbara United States Department of Biomedical Engineering Illinois Institute of Technology United States

ISBN: (纸本)9781510671621

Convolutional neural networks (CNNs) have been previously used as model observers (MO) for the purpose of defect detection in medical images. Due to their limited generalizability, such CNN observers do not possess the ability to recognize whether an input image comes from the same distribution as the data it was trained on, i.e., the ability of having domain awareness. In this paper we propose an adaptive learning approach for training a domain-aware CNN ideal observer (IO). In our approach we use a variant of U-Net CNN which is trained simultaneously for defect localization prediction and for reconstruction of the input image. We demonstrate that the reconstruction mean-squared-error (MSE) by the network can serve as an indicator of how well the observer performs in the defect localization task, which is an important step towards developing a domain-aware MO. Furthermore we propose an adaptive learning approach by automatically selecting datasets on which the model in training has poor reconstruction MSE. Our results show that this adaptive training approach can improve the model performance both in generalization and defect localization compared to a non-adaptive approach, particularly for out-of-distribution images - images that were not seen during the training of the algorithm. © 2024 SPIE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral Image Classification Using Custom Spectral Convolutional Neural Networks (CSCNNs) 14

Hyperspectral Image Classification Using Custom Spectral Con...

引用

14th International Conference on Information Science and Technology, ICIST 2024

作者： Rouamba, Stephanie Zhang, Nian Mahmoud, Wagdy Thompson, Lara Denis, Max Deksissa, Tolessa University of the District of Columbia Department of Electrical and Computer Engineering WashingtonDC United States University of the District of Columbia Department of Mechanical Engineering Biomedical Engineering Program WashingtonDC United States University of the District of Columbia Department of Mechanical Engineering WashingtonDC United States Water Resources Research Institute University of the District of Columbia WashingtonDC United States

ISBN: (纸本)9798350353334

Custom spectral convolutional neural networks (CSCNNs) combine the strengths of convolutional neural networks with specialized spectral processing, resulting in improved classification accuracy by effectively capturing subtle variations in hyperspectral data. This paper proposes a CSCNNs based approach to classify hyperspectral images. The proposed method leverages the high-dimensional spectral data inherent in hyperspectral images, employing convolutional layers specifically designed to capture spectral-spatial features. By reducing dimensionality through principal component analysis and creating image patches, the model is trained to recognize complex patterns across different spectral bands. In addition, a comprehensive analysis of CSCNN performance is carried out, focusing on its architecture, key features, and benefits in computational efficiency and spectral representation. Experimental results on datasets such as Salinas-A, Pavia University (Pavia-U), and Indian Pines demonstrate that the CSCNN model surpasses traditional methods, achieving higher classification accuracy and more robust performance metrics like overall accuracy (OA), average accuracy (AA), and Kappa coefficient. The proposed model achieved an overall accuracy (OA) of 98.2%, 96.7%, and 92.5% for the Salinas-A, Pavia-U, and Indian Pines datasets, respectively. The model's average accuracy (AA) was 97.5%, 95.4%, and 89.8% for the same datasets. Furthermore, the model attained Kappa coefficients of 97.7%, 95.2%, and 89.1% for Salinas-A, Pavia-U, and Indian Pines, respectively. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：