检索结果-内蒙古大学图书馆

XAI-PSSGAN: Perception-Enhanced Spectrum Shift Generative Adversarial Network with Explainable AI System for NIR-II Fluorescence Molecular Imaging 22

XAI-PSSGAN: Perception-Enhanced Spectrum Shift Generative Ad...

引用

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Fu, Lidan Lu, Binchun Li, Lingbing Shi, Xiaojing Tian, Jie Hu, Zhenhua Cas Key Laboratory of Molecular Imaging Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Department of Precision Instrument Tsinghua University Beijing China Interventional Radiology Department Chinese Pla General Hospital Beijing China Beijing Advanced Innovation Center for Big Data-based Precision Medicine School of Engineering Medicine Beihang University Beijing China Engineering Research Center of Molecular and Neuro Imaging of Ministry of Education School of Life Science and Technology Xidian University Xi'an China National Key Laboratory of Kidney Diseases Beijing China

ISBN: (纸本)9798331520526

Fluorescence imaging in the second near-infrared window (NIR-II) facilitates the real-time optical contrast for in vivo biomedical imaging. However, the detection noise is an inevitable byproduct of the real-time imaging process. Additionally, fluorescent probes currently approved for clinical use remain limited within the wavelength of NIR-IIa (1000-1300 nm). NIR-IIb (1500-1700 nm) imaging has the potential to deliver higher resolution than NIR-IIa, but relies on toxic fluorescent probes not approved for human use. To support clinical high-precision imaging, we propose a NIR-II fluorescence enhanced imaging method based on perception-enhanced spectrum shift generative adversarial network with explainable AI systems (xAI-PSSGAN). The proposed method tackles the problem of limited unpaired data training, and generates NIR-IIb resembling images with enhanced resolution and reduced noise. Experimental results demonstrate that proposed method effectively improves the imaging quality and surpasses competing alternatives, which is of great significance to promote the high-quality NIR-II imaging in clinical practice. © 2025 IEEE.

关键词： Molecular imaging

来源：评论

学校读者我要写书评

暂无评论

DT-GRU Approach for Developing Remotely Monitored Central Nervous System Using Wearable Sensors

DT-GRU Approach for Developing Remotely Monitored Central Ne...

引用

Self Sustainable artificial intelligence Systems (ICSSAS), International Conference on

作者： Sreejith L Das Aaftab Alam B. Anjanee Kumar K. K. Sunalini Vidhya P Sakthivel M ECE Department Sri Venkateswara College of Engineering and Technology Chittoor Andhra Pradesh India Department of CSE Integral University Lucknow India Department of ECE Dadi Institute of Engineering and Technology Anakapalle Visakhapatnam India Department of English Koneru Lakshmaiah Education Foundation Vaddeswaram Guntur Andhra Pradesh India Department of Artificial Intelligence and Data Science M. Kumarasamy College of Engineering Karur India Department of CSE-AIML Malla Reddy College Of Engineering Hyderabad Telangana India

The capacity to evaluate numerous bio- behavioral factors concurrently is a potential strength of RMT for patients with neurological diseases. Although the benefits of RMT have been demonstrated in a large body of research, there is still some debate over its usefulness among medical experts. Providing doctors with visual summaries of data in the patient record is one way that RMT could enhance the system of care for patients with illnesses of the central nervous system. It's important to think about the constraints on the benefits these technologies can provide and the difficulties of using them effectively. The proposed strategy makes use of pre-processing, feature selection, and model training. data is preprocessed in LMS. Feature selection employs PCA. The DT -GRU algorithm is used to train the model. Two additional models, GRU and DT, are contrasted with the suggested method. The proposed method is quite effective.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimizing Neural Network Image Classification with Fractional Order Gradient Methods

SSRN

引用

SSRN 2023年

作者： Zhao, Haiming Chen, Jiejie Jiang, Ping Zeng, Zhigang Fan, Rui Yang, Honggang School of Computer and Information Engineering Hubei Normal University Huangshi435000 China School of Automation Wuhan University of Technology Wuhan430000 China School of Computer Hubei PolyTechnic University Huangshi435000 China Faculty of Artificial Intelligence Education Central China Normal University Wuhan430000 China School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430000 China Key Laboratory of Image Information Processing and Intelligent Control Ministry of Education of China Wuhan430000 China

As deep learning technology increasingly permeates various fields, the significance of optimization algorithms in neural network training has become more prominent. Typically, the training of deep learning networks relies on gradient descent optimizers based on calculus computations. Traditional integer-order calculus optimizers often fall short when dealing with complex and subtle data. To overcome these limitations, we innovatively utilized the Grünwald-Letnikov and Caputo definitions of fractional calculus to develop two new optimization algorithms: Fractional Order Adagrad (FAdagrad) and Fractional Order Adam (FAdam). These algorithms have been proven to converge to extremal points.A key advantage of our methods is their capability in handling complex, high-dimensional tasks, such as artistic painting image classification, where they exhibit exceptional performance in recognizing and processing subtle differences. The robustness and adaptability of these algorithms were further validated through experiments on the CIFAR-100 dataset. The results demonstrate the efficacy of fractional order optimizers in enhancing training efficiency and accuracy. © 2023, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

LHQ-SVC: Lightweight and High Quality Singing Voice Conversion Modeling

LHQ-SVC: Lightweight and High Quality Singing Voice Conversi...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yubo Huang Xin Lai Muyang Ye Anran Zhu Zixi Wang Jingzehua Xu Shuai Zhang Zhiyuan Zhou Weijie Niu School of Civil Engineering Southwest Jiaotong University China Zhenguan AI Lab ZhenGuan Innovation (Shenzhen) Technology Co. Ltd China SWJTU-Leeds Joint School Southwest Jiaotong University China School of Computing and Artificial Intelligence Southwest Jiaotong University China Shenzhen International Graduate School Tsinghua University China Department of Data Science New Jersey Institute of Technology USA School of Economics and Management Southwest Jiaotong University China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Singing Voice Conversion (SVC) has emerged as a significant subfield of Voice Conversion (VC), enabling the transformation of one singer’s voice into another while preserving musical elements such as melody, rhythm, and timbre. Traditional SVC methods have limitations in terms of audio quality, data requirements, and computational complexity. In this paper, we propose LHQ-SVC, a lightweight, CPU-compatible model based on the SVC framework and diffusion model, designed to reduce model size and computational demand without sacrificing performance. We incorporate features to improve inference quality, and optimize for CPU execution by using performance tuning tools and parallel computing frameworks. Our experiments demonstrate that LHQ-SVC maintains competitive performance, with significant improvements in processing speed and efficiency across different devices. The results suggest that LHQ-SVC can meet real-time performance requirements, even in resource-constrained environments.

关键词： Performance evaluation Computational modeling Static VAr compensators Parallel processing Signal processing Rhythm Real-time systems Timbre Speech processing Tuning

来源：评论

学校读者我要写书评

暂无评论

Privacy Preservation in Gen AI Applications

arXiv

引用

arXiv 2025年

作者： Swetha, S. Shaju, Ram Sundhar K. Rakshana, M. Ganesh, R. Balavedhaa, S. Thiruvaazhi, U. Information Science and Engineering Kumaraguru College of Technology Tamil Nadu Coimbatore India Computer Science and Engineering Kumaraguru College Of Technology Tamil Nadu Coimbatore India Artificial Intelligence and Data Science Kumaraguru College Of Technology Tamil Nadu Coimbatore India

The ability of machines to comprehend and produce language that is similar to that of humans has revolutionized sectors like customer service, healthcare, and finance thanks to the quick advances in Natural Language Processing (NLP), which are fueled by Generative artificial intelligence (AI) and Large Language Models (LLMs). However, because LLMs trained on large datasets may unintentionally absorb and reveal Personally Identifiable Information (PII) from user interactions, these capabilities also raise serious privacy concerns. Deep neural networks' intricacy makes it difficult to track down or stop the inadvertent storing and release of private information, which raises serious concerns about the privacy and security of AI-driven data. This study tackles these issues by detecting Generative AI weaknesses through attacks such as data extraction, model inversion, and membership inference. A privacy-preserving Generative AI application that is resistant to these assaults is then developed. It ensures privacy without sacrificing functionality by using methods to identify, alter, or remove PII before to dealing with LLMs. In order to determine how well cloud platforms like Microsoft Azure, Google Cloud, and AWS provide privacy tools for protecting AI applications, the study also examines these technologies. In the end, this study offers a fundamental privacy paradigm for generative AI systems, focusing on data security and moral AI implementation, and opening the door to a more secure and conscientious use of these tools. © 2025, CC BY.

关键词： Privacy by design

来源：评论

学校读者我要写书评

暂无评论

Empirical Evaluation of Phishing Attack Identification Over Websites by Using Intelligent Deep Learning Principle

Empirical Evaluation of Phishing Attack Identification Over ...

引用

Emerging Trends in engineering and Medical sciences (ICETEMS), International Conference on

作者： Anitha G Mouleeswaran S K Dickson Irudayaraj V G B Santhi Rajkumar Chadge Hani Attar Department of ECE Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences (SIMATS) Saveetha University Chennai Tamil Nadu India Department of CSE Dayananda Sagar University Bangalore India Artificial Intelligence and Data Science Dhanalakshmi College of Engineering Chennai Tamil Nadu India Department of CSE New Prince Shri Bhavani College of Engineering and Technology Chennai Tamil Nadu India Department of Mechanical Engineering Yeshwantrao Chavan College of Engineering Nagpur India Department of Energy Engineer Zarqa University Zarqa Jordan College of Engineering University of Business and Technology Jeddah Saudi Arabia

ISBN: (数字)9798331529352

ISBN: (纸本)9798331529390

Phishing schemes keep on dominating the cyber threat landscape hence requiring enhancement of detection methods. This research sets out a new and novel idea for the detection of phishing website approach incorporating the XGBoost with ANN to boost the identification performance. The proposed model was compared with nine machine learning and deep learning models which includes Logistic Regression (LR), Decision Trees (DT), Random Forest (RF), SVM, KNN, Naïve Bayes, XG Boost, artificial Neural Networks (ANN), CNN. It is apparent that the Ensembled XGBoost-ANN model was able to attain a higher accuracy rate of 9,463% and ranks on top of all the other models. It also showed great accuracy to prove precision, recall and F1-score metrics of the model. The findings reported herein support the suitability of the ensemble technique in modeling the intricate relationships characteristic of phishing websites, thereby holding the promise of being a practical solution in real-world environments. The reason why this methodology is promising for deployment in cybersecurity frameworks targeting the mitigation of phishing attacks is its capability of achieving high detection rates without making many false-positive detections.

关键词： Deep learning Support vector machines Radio frequency Measurement Accuracy Phishing Prevention and mitigation artificial neural networks Nearest neighbor methods Random forests

来源：评论

学校读者我要写书评

暂无评论

Imbalanced Least Squares Regression with Adaptive Weight Learning

SSRN

引用

SSRN 2022年

作者： Li, Yanting Jin, Junwei Ma, Jiangtao Zhu, Fubao Jin, Baohua Liang, Jing Chen, C.L. Philip College of Computer and Communication Engineering Zhengzhou University of Light Industry Zhengzhou450001 China School of Artificial Intelligence and Big Data Henan University of Technology Zhengzhou450001 China School of Electrical and Information Engineering Zhengzhou University Zhengzhou450001 China School of Computer Science and Engineering South China University of Technology China

Least squares regression (LSR) has demonstrated promising performance in various classification tasks owing to the effectiveness and efficiency. However, there are some deficiencies that seriously plague its application in imbalanced data scenarios. The first is the success of LSR heavily relies on the balanced class distribution, more fundamentally, a severely imbalanced class distribution may damage the effectiveness of the algorithm seriously. Secondly, the utilized binary label matrix in conventional LSR model may be too strict to learn a discriminative transform matrix for imbalanced learning. To address the above issues, in this paper an adaptive weight learning mechanism and a novel label relaxation constraint are proposed and incorporated into the framework of LSR to tackle the imbalanced classification problem. Specifically, the weight of each sample can be adaptively obtained according to the original distribution information of the imbalanced data, in which the importance of the minority class samples can be better reflected with greater weights. Then a new label relaxation matrix is constructed to widen the margins between different classes. Further, we provide an iterative algorithm to solve the resulting optimization problem. Extensive experimental results on diverse imbalanced datasets show that the proposed method outperforms the majority of imbalanced learning approaches. © 2022, The Authors. All rights reserved.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Effect of spatial heterogeneity on level of rejuvenation in Ni80P20 metallic glass

引用

Chinese Physics B 2022年第9期31卷 435-443页

作者： Tzu-Chia Chen Mahyuddin KM Nasution Abdullah Hasan Jabbar Sarah Jawad Shoja Waluyo Adi Siswanto Sigiet Haryo Pranoto Dmitry Bokov Rustem Magizov Yasser Fakri Mustafa A.Surendar Rustem Zalilov Alexandr Sviderskiy Alla Vorobeva Dmitry Vorobyev Ahmed Alkhayyat Dhurakij Pundit University Bangkok 10210Thailand Data Science and Computational Intelligence Research Group Universitas Sumatera UtaraMedanIndonesia Optical Department College of Health and Medical TechnologySawa UniversityMinistry of Higher Education and Scientific ResearchAl-MuthanaaSamawahIraq College of Health&Medical Technology Al-Ayen UniversityIraq Faculty of Engineering Universitas Muhammadiyah SurakartaJawa Tengah 57102Indonesia Department of Mechanical Engineering Faculty of Science and TechnologyUniversitas Muhammadiyah Kalimantan TimurSamarinda 75124Indonesia Institute of Pharmacy Sechenov First Moscow State Medical University8 Trubetskaya St.Bldg.2Moscow 119991Russia Kazan Federal University KazanRussia Department of Pharmaceutical Chemistry College of PharmacyUniversity of MosulMosul-41001Iraq Department of Pharmacology Saveetha Dental College and HospitalSaveetha Institute of Medical and Technical SciencesChennaiIndia Nosov Magnitogorsk State Technical University MagnitogorskRussia Innovative University of Eurasia PavlodarRepublic of Kazakhstan K.G.Razumovsky Moscow State University of Technologies and Management(The First Cossack University) Moscow 109004Russia College of Technical Engineering The Islamic UniversityNajafIraq

Understanding the relation between spatial heterogeneity and structural rejuvenation is one of the hottest topics in the field of metallic glasses(MGs).In this work,molecular dynamics(MD)simulation is implemented to discover the effects of initial spatial heterogeneity on the level of rejuvenation in the Ni_(80)P_(20)*** this purpose,the samples are prepared with cooling rates of 10^(10) K/s-10^(12) K/s to make glassy alloys with different atomic ***,it is found that the increase in the cooling rate leads the Gaussian-type shear modulus distribution to widen,indicating the aggregations in both elastically soft and hard *** the primary evaluations,the elastostatic loading is also used to transform structural rejuvenation into the atomic *** results indicate that the sample with intermediate structural heterogeneity prepared with 10^(11) K/s exhibits the maximum structural rejuvenation which is due to the fact that the atomic configuration in an intermediate structure contains more potential sites for generating the maximum atomic rearrangement and loosely packed regions under an external *** features of atomic rearrangement and structural changes under the rejuvenation process are discussed in detail.

关键词： metalic glasses mechanical properties molecular dynamics simulation disordered solids

来源：评论

学校读者我要写书评

暂无评论

Assisted Design of High-Performance Electrochromic Devices Using Multilayer Perceptron Neural Network with High R2

SSRN

引用

SSRN 2024年

作者： Li, Muyun Cai, Qingyue Lai, Huayi Ning, Honglong Kong, Sifan Guo, Chenxiao Jiang, Bocheng Su, Guoping Yao, Rihui Peng, Junbiao Guangdong Basic Research Center of Excellence for Energy & Information Polymer Materials State Key Laboratory of Luminescent Materials and Devices School of Materials Sciences and Engineering South China University of Technology Guangzhou510640 China Aberdeen Institute of Data Science and Artificial Intelligence South China Normal University Foshan528225 China School of Software South China Normal University Foshan528225 China

Electrochromic devices (ECDs), serving as thermal insulation and decorative components, have been extensively employed across various domains. To meet diverse application demands in different contexts, the product parameters of ECDs exhibit continual variations, notably within the electrochromic layer. In this work, a multilayer perceptron neural network was leveraged to comprehend the transition from material parameter alterations to device performance parameter changes. Specifically, 336 groups of electrochromic solutions containing different concentrations and voltages were prepared. Subsequently, a model comprising three hidden layers and three output neurons was developed, expanding the dataset simulation to 2000 groups, yielding an R2 of 0.975 on modulation amplitude. Through this model, accurate estimations of performance parameters of numerous unmanufactured liquid-based ECDs were achieved, with projected outcomes highlighting a peak modulation amplitude of 78.38%@620nm. It has been verified that the modulation amplitude prediction error was below 5% and demonstrated stable cyclic performance over 1000 times (97.93%@1.45V) , thereby underscoring the feasibility of our approach. This research serves as a pertinent instance of the synergy between multilayer perceptron neural networks and ECDs, thereby advancing the deployment of neural network in the innovation and application of chemical engineering, materials and other fields. © 2024, The Authors. All rights reserved.

关键词： Thermal insulation

来源：评论

学校读者我要写书评

暂无评论

Raman response, piezoelectricity, and transport properties of the two-dimensional Janus HfSiX3H (X=N/P/As) semiconductors: A first-principles study

引用

Physical Review B 2024年第23期110卷 235403-235403页

作者： Tuan V. Vu Nguyen T. Hiep Huynh V. Phuc Bui D. Hoi A. I. Kartamyshev Nguyen N. Hieu Laboratory for Computational Physics Institute for Computational Science and Artificial Intelligence Van Lang University Ho Chi Minh City 70000 Vietnam Faculty of Mechanical - Electrical and Computer Engineering School of Technology Van Lang University Ho Chi Minh City 70000 Vietnam Institute of Research and Development Duy Tan University Da Nang 550000 Vietnam Faculty of Natural Sciences Duy Tan University Da Nang 550000 Vietnam Division of Physics School of Education Dong Thap University Cao Lanh 870000 Vietnam Faculty of Physics University of Education Hue University Hue 530000 Vietnam

Exploring advanced materials with extraordinary properties is necessary for developing high-performance devices. In this work we construct two-dimensional Janus HfSiX3H (X=N/P/As) monolayers and systematically examine their physical properties by utilizing first-principles calculations. Specifically, we focus on the stabilities, Raman spectra, electronic structures, piezoelectricity, and transport properties of the monolayers. As a result, the studied HfSiN3H, HfSiP3H, and HfSiAs3H monolayers are found as semiconductors with good structural, mechanical, dynamic, and thermodynamic stabilities. The piezoelectric coefficients (e11 and e31) are calculated to examine the piezoelectricity of the HfSiX3H monolayers. It is noted that besides the in-plane piezoelectricity, the three monolayers show additional out-of-plane piezoelectric effects because of their noncentrosymmetric structures. We also include fully anisotropic acoustic deformation potential, ionized impurity, piezoelectric, and polar electron–phonon scattering to calculate the total carrier mobilities of Janus HfSiX3H. It is demonstrated that Janus HfSiX3H monolayers exhibit low carrier mobilities at room temperature (<100cm2V−1s−1). Scattering from polar optical phonons is found to dominate the total mobilities of both electrons and holes.

关键词： Electronic structure Piezoelectricity Transport phenomena Vibrational states Layered semiconductors First-principles calculations

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：