检索结果-内蒙古大学图书馆

Learning to Localize Cross-Anatomy Landmarks in X-Ray Images with a Universal Model

Biomedical Engineering Frontiers 2022年第1期3卷 298-308页

作者： Heqin Zhu Qingsong Yao Li Xiao S.Kevin Zhou Key Lab of Intelligent Information Processing of Chinese Academy of Sciences(CAS) Institute of Computing TechnologyCASBeijing 100190China Center for Medical Imaging RoboticsAnalytic Computing&Learning(MIRACLE)School of Biomedical Engineering&Suzhou Institute for Advanced ResearchUniversity of Science and Technology of ChinaSuzhou 215123China

Objective and Impact *** this work,we develop a universal anatomical landmark detection model which learns once from multiple datasets corresponding to different anatomical *** with the conventional model trained on a single dataset,this universal model not only is more light weighted and easier to train but also improves the accuracy of the anatomical landmark *** accurate and automatic localization of anatomical landmarks plays an essential role in medical image ***,recent deep learning-based methods only utilize limited data from a single *** is promising and desirable to build a model learned from different regions which harnesses the power of big *** model consists of a local network and a global network,which capture local features and global features,*** local network is a fully convolutional network built up with depth-wise separable convolutions,and the global network uses dilated convolution to enlarge the receptive field to model global *** evaluate our model on four 2D X-ray image datasets totaling 1710 images and 72 landmarks in four anatomical *** experimental results show that our model improves the detection accuracy compared to the state-of-the-art *** model makes the first attempt to train a single network on multiple datasets for landmark *** results qualitatively and quantitatively show that our proposed model performs better than other models trained on multiple datasets and even better than models trained on a single dataset separately.

关键词： convolution utilize separable

来源：评论

学校读者我要写书评

暂无评论

Personalized Federated Learning with Collaborative Aggregation Networks for Multi-Site Brain Disorder Diagnosis 4

Personalized Federated Learning with Collaborative Aggregati...

引用

4th International Conference on Industrial Automation, Robotics and Control Engineering, IARCE 2024

作者： Si, Qian Li, Yang School of Cyber Science and Technology Beihang University Beijing China Department of Automation Science and Electrical Engineering The Beijing Advanced Innovation Center for Big Data and Brain Computing State Key Laboratory of Virtual Reality Technology and Systems Beijing China Advanced Institute of Information Technology Peking University Beijing China Beihang University Beijing China

ISBN: (纸本)9798350380323

In multi-site brain disease diagnosis studies, traditional centralized training methods necessitate sharing medical data, posing significant privacy risks. Federated learning (FL) offers a privacy-preserving solution by enabling global model training through aggregating locally trained models from multiple data centers without sharing raw data. However, current FL approaches rely on a server-based network topology, where central server failure disrupts training. Additionally, data heterogeneity across sites often slows convergence and reduces accuracy. To overcome these issues, we introduce a decentralized personalized federated learning collaborative aggregation network (pFedCAN). This framework has two core components: (1) separating local models into shared and personalized layers, and (2) forming a collaborative aggregation network via similarity detection in the shared layers. Specifically, each center trains its local model, then separates it into shared and personalized layers. The shared layer is exchanged with other centers, while the personalized layer remains local. Data centers analyze similarities in received shared layers to build a collaborative network, where shared layers from similar centers are aggregated to refine the model. This approach flexibly adapts to varying levels of data heterogeneity, enhancing model training efficiency. Validation on public datasets, ABIDE I and ADHD, shows that the proposed method outperforms current leading techniques. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Neural Abstractive Summarization: A Brief Survey 3

Neural Abstractive Summarization: A Brief Survey

引用

3rd IEEE International Conference on Computer Communication and Artificial Intelligence, CCAI 2023

作者： Qian, Lu Zhang, Haiyang Wang, Wei Liu, Dawei Huang, Xin Xian Jiaotong-Liverpool University School of Advanced Technology Department of Computing Suzhou China De Montfort University Cyber Technology Institute United Kingdom Taiyuan University of Technology College of Data Science Taiyuan China

ISBN: (纸本)9798350335262

Due to the development of neural networks, abstractive summarization has received more attention than extractive one, and has gained significant progress in generating fluent and human-like summaries with novel expressions. Seq2seq has become the primary framework for abstractive summarization, employing encoder-decoder architecture based on RNNs or CNNs, and Transformers. In this paper, we focus on reviewing the neural models that are based on seq2seq framework for abstractive summarization. Moreover, we discuss some of the most effective techniques for improving seq2seq models and provide two challenging directions, i.e. generating query-based abstractive summaries and incorporating commonsense knowledge, for in-depth investigation. © 2023 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Frist-principles prediction of elastic, electronic,and thermodynamic properties of high entropy carbide ceramic (TiZrNbTa)C

引用

Rare Metals 2022年第3期41卷 1002-1014页

作者： Kai Xiong Bao-Wen Wang Ze-Peng Sun Wei Li Cheng-Chen Jin Shun-Meng Zhang Si-Yong Xu Lei Guo Yong Mao National Center for International Research on Photoelectric and Energy Materials School of Materials and EnergyMaterials Genome InstituteYunnan UniversityKunming 650091China Advanced Computing Center Information Technology CenterYunnan UniversityKunming 650091China

High-entropy carbide ceramics (HECs) have drawn increasing attention as their excellent mechanical and thermal properties. In this work, the crystal stability,mechanical behavior, electronic and thermodynamic properties of (TiZrNbTa)C HEC are investigated by the first-principles calculations. Obtained results reveal that the disordered transition-metal (TM) atoms result in serious local lattice distortion within the crystal. The lattice distortion plays a key role for the structural stabilization,mechanical anisotropy and thermodynamic behaviors of(TiZrNbTa)C. Increasing pressure leads to decrease the lattice parameter, volume and brittleness, meanwhile increase the elastic constants, elastic moduli, mechanical anisotropy, sound velocity, and Debye temperature. It is also discovered that charge delocalization occurs with the increase in pressure. The mechanical stability and anisotropy of (TiZrNbTa)C are attributed primarily to TM-C bonding.

关键词： High-entropy carbide ceramics Elastic constants Anisotropy Thermodynamic First-principles

来源：评论

学校读者我要写书评

暂无评论

Enhanced Prediction Of Rheumatoid Arthritis Flares Through Ensemble Learning Methods 2

Enhanced Prediction Of Rheumatoid Arthritis Flares Through E...

引用

2nd IEEE International Conference on Advances in Information technology, ICAIT 2024

作者： Priyadharshini, K. Vinitha, S. Sivaprakash, S. Kayalvili, S. Thilakavathy, P. Vasavi, M. Department of Computing Technologies School of Computing SRM Institute of Science and Technology Kattankulathur Campus Chennai India Department of Artificial Intelligence and Machine Learning Jerusalem College of Engineering Chennai India School of Computer Science and Engineering Vellore Institute of Technology Vellore India Department of Artificial Intelligence Kongu Engineering College Erode India Vels Institute of Science Technology and Advanced Studies Chennai India ET Department CVR College of Engineering Hyderabad India

ISBN: (纸本)9798350383867

RA is an inflammatory, autoimmune disease with an unpredictable course, during which patients experience exacerbations (increases in symptoms) and its associated complications, such as deterioration of the joint and decreased quality of life. Tipping the exact point when remission is reached or RA flares is most essential for devising optimal treatment and enhancing patient's well-being. In this study, we put forward the usage of various types of ensemble learning algorithms aimed at improved the predictive power of RA flares. We use diverse data sources, including clinical data, medical imagery, genetic profiles and patient-reported data, and ensemble methods like Random Forest, Gradient Boosting, AdaBoost and Stacking processors to help us identify the patterns automatically. Thus ensembles of models are shown to be leading in performance, reaching superior accuracy, accuracy precision, recall rate, F1-score and area under the receiver operating characteristic curve (AUC-ROC). Feature importance analysis highlights the significant impact of clinical features which include the levels of CRP, ESR, age of the patient, duration of the disease on the likelihood of RA flares among the individual variants. The sensitivity analysis shows how well the ensemble model behaves under the generation of different model parameters and input data. The presented study emphasizes the immense facilitative ability of the ensemble methods of learning as predicting RA flares and this can also provide great information about the disease process and personalized medical decision-making. The areas for future research could be improved ensembles models, identification of novel feature selection approaches and validation of model performance in a variety of clinical settings. This will give the researchers an opportunity to implement the models in practice and make clinics more effective. © 2024 IEEE.

关键词： Sensitivity analysis

来源：评论

学校读者我要写书评

暂无评论

Custom Transformer-Based Approach for Enhanced Bengali Automatic Speech Recognition 1

Custom Transformer-Based Approach for Enhanced Bengali Autom...

引用

1st International Conference on Innovations in Communications, Electrical and Computer Engineering, ICICEC 2024

作者： Deepa, R. Jayalakshmi, V. Thilakavathy, P. Manikandan, G. Surendran, R. SRM Institute of Science and Technology School of Computing Department of Computing Technologies Chennai India SRM Institute of Science and Technology Department of Computer Science and Applications Ramapuram Chennai India Vels Institute of Science and Engineering Vels Institute of Science Technology and Advanced Studies Chennai India Anna University St. Joseph's College of Engineering Department of Artificial Intelligence and Machine Learning Chennai India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798350376517

ISBN: (纸本)9798350376517

Automated speech recognition (ASR) systems struggle with Bengali, which is the fifth most spoken language. Bengali has a varied morphology, many dialects, and limited high-quality annotated voice data. Traditional voice recognition techniques work well however struggle with Bengali phonetics and syntax. This method eliminates the need for large amounts of tagged data and the difficulties of differentiating sounds in different languages, both issues are addressed by this procedure. The research in this paper offers an Intellectual Bengali Speech Recognition System using Deep Learning Techniques (IBSRS-DLT). The model in this system uses transformer architecture and attention processes to manage Bengali speech pattern long-range dependencies. The IBSRS-DLT outperforms standard models in recognition accuracy using a pre-trained transformer model modified for phonetic and lexical correctness and fine-tuned on Bengali datasets. After training on a vast Bengali speech corpus, real-world datasets and rigorous simulations confirm the system. IBSRS-DLT can be used for automatic transcription, voice-controlled devices, and Bengali virtual assistants, such applications are efficient and accessible. The simulation showed a considerable boost in recognition speed and a decrease in keyword mistakes compared to current methods. This paper shows that deep learning can solve linguistic issues in ASR systems, enabling scalable regional language detection. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Thermally robust electrically reversible spin selection in a quantum dot qubit with double reservoir spin accumulations

引用

Physical Review Applied 2024年第5期22卷 054049-054049页

作者： R. Jansen S. Yuasa Research Center for Emerging Computing Technologies National Institute of Advanced Industrial Science and Technology (AIST) Tsukuba Ibaraki 305-8568 Japan

Spin selection is explored for a quantum dot spin qubit with spin accumulations ΔμS and ΔμD in the source and drain reservoir, created by ferromagnetic contacts. If ΔμS and ΔμD have opposite signs, a single electron can be trapped on the quantum dot in two opposite spin states, depending on the source-drain voltage. The selected spin is then electrically reversible and with low error rates for temperatures up to above 4 K. If ΔμS and ΔμD have the same sign, the spin selectivity is enhanced and persists up to even higher temperature than for a spin qubit with spin accumulation in only one of the reservoirs. These results offer opportunities to improve and electrically control spin qubit initialization and readout at elevated temperature.

关键词： Electric field effects Quantum information with solid state qubits Spin accumulation Spintronics Quantum dots Qubits Semiconductors Spin

来源：评论

学校读者我要写书评

暂无评论

Detecting Brain Tumor Stages Using Convolutional AutoEncoder (CAE) with Hybrid Deep Learning Method

Detecting Brain Tumor Stages Using Convolutional AutoEncoder...

引用

Sustainable computing and Smart Systems (ICSCSS), International Conference on

作者： D Agalya S. Kamalakkannan Department of Computer Science School of Computing Sciences Vels Institute of Science Technology & Advanced Studies Chennai India Department of Information Technology School of Computing Sciences Vels Institute of Science Technology & Advanced Studies Chennai India

ISBN: (数字)9798350379990

ISBN: (纸本)9798350391558

The Brain Tumor (BT) is the formation of abnormal brain cells, few of which can develop into cancer. Early and prompt detection of disease and development of treatment strategies improve patients' wellbeing and life span. Models are created using Deep Learning (DL) and Magnetic Resonance Imaging (MRI) to classify and diagnose brain tumours. This facilitates the effortless detection of brain tumours. The BT is usually caused through abnormal brain cell proliferation that damage the brain structure as well as eventually resulted in various stages of BT. Early BT discovery as well as timely treatment have reduced the mortality rate. Consideration of deep structure for analysis that have been used in both non-linear feature extraction and unsupervised learning, which relies significantly on the Autoencoder (AE). This transfer learning is utilized to obtain better accuracy. Applications for the AE and its variations have been effective in many domains, including recommender systems, data production, pattern recognition, and computer vision. Moreover, the Convolutional AE (CAE) toolkits are addressed for better performance in detection of the brain tumor. The goal of this research is determining whether DL methods can be utilized to automate the detection process. This work uses Convolutional Neural Network (CNN) with VGG19 as the hybrid model to provide the better or maximum accuracy of 92.39% than the traditional Convolutional Neural Network (CNN) with VGG19. This study objective is to use TL with hybrid CNN techniques through this assessment and analysis to direct researchers and medical experts towards effective BT detection systems.

关键词： Deep learning Accuracy Magnetic resonance imaging Brain cells Feature extraction Brain modeling Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Feedback voltage driven chaos in a three-terminal spin-torque oscillator

引用

Physical Review B 2024年第13期110卷 134403页

作者： Tomohiro Taniguchi National Institute of Advanced Industrial Science and Technology (AIST) Research Center for Emerging Computing Technologies Tsukuba Ibaraki 305-8568 Japan

In this work, we report an excitation of chaos and a nontrivial magnetization switching via transient chaos in a three-terminal spin-torque oscillator (STO). The driving force of the chaos is a voltage-controlled magnetic anisotropy (VCMA) effect generated by a feedback signal from the STO since the feedback effect is known to be effective in exciting chaos in a dynamical system. Solving the Landau-Lifshitz-Gilbert equation numerically and applying temporal and statistical analyses to its solution, the existence of the chaos driven by the feedback VCMA effect is identified. Simultaneously, however, transient chaos is also observed, where the magnetization initially shows chaotic behavior but finally switches its direction. This transient dynamics from chaos to magnetization switching was unexpected because the sign of the feedback VCMA effect was chosen so that the switching current increases and, as a result, the situation rather favors the condition for sustaining chaos. It is implied that this switching happens when narrowing a stable region of the magnetic potential energy by the feedback effect and magnetization precession pointing to a saddle point coincidentally occur simultaneously.

关键词： Chaos Spintronics Magnetic multilayers Landau-Lifschitz-Gilbert equation

来源：评论

学校读者我要写书评

暂无评论

Write-error reduction in voltage-driven magnetization switching using a recording layer with low magnetic damping

引用

Physical Review Applied 2024年第5期21卷 054008-054008页

作者： Tatsuya Yamamoto Tomohiro Ichinose Takayuki Nozaki Shingo Tamaru Kay Yakushiji Hitoshi Kubota Shinji Yuasa National Institute of Advanced Industrial Science and Technology (AIST) Research Center for Emerging Computing Technologies Tsukuba Ibaraki 305-8568 Japan

We develop perpendicularly magnetized magnetic tunnel junctions consisting of a recording layer exhibiting a low magnetic damping to investigate the influence of magnetic damping on the write-error rate of voltage-driven magnetization switching. The effective magnetic damping is reduced to about one-third that of a conventional Ta/Co−Fe−B/MgO recording layer by eliminating the spin pumping effect. The low magnetic damping contributes to a 3-orders-of-magnitude reduction in the write-error rate for the longer write pulses, which is explained by the suppression of thermal fluctuation during the switching process. The low magnetic damping also enables a long period of magnetization precession to control the dynamics via the voltage-controlled magnetic anisotropy effect.

关键词： Fluctuations & noise Magnetic anisotropy Magnetization switching Magnetoelectric effect Spin pumping Devices for digital logic, storage & processing Magnetic multilayers Magnetic tunnel junctions

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：