检索结果-内蒙古大学图书馆

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

作者： Niloufar Eghbali Zarch Hassan Bagher-Ebadian Tuka Alhanai Mohammad M. Ghassemi Computer Science Department Michigan State University East Lansing USA Department of Radiation Oncology Henry Ford Health Detroit US Division of Engineering New York University Abu Dhabi UAE

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Vision Transformers (ViTs) have shown promise in medical image semantic segmentation (MISS) by capturing longrange correlations. However, ViTs often struggle to model local spatial information effectively, which is essential for accurately segmenting fine anatomical details, particularly when applied to small datasets without extensive pre-training. We introduce Gabor and Laplacian of Gaussian Convolutional Swin Network (GLoG-CSUnet), a novel architecture enhancing Transformer-based models by incorporating learnable radiomic features. This approach integrates dynamically adaptive Gabor and Laplacian of Gaussian (LoG) filters to capture texture, edge, and boundary information, enhancing the feature representation processed by the Transformer model. Our method uniquely combines the longrange dependency modeling of Transformers with the texture analysis capabilities of Gabor and LoG features. Evaluated on the Synapse multi-organ and ACDC cardiac segmentation datasets, GLoG-CSUnet demonstrates significant improvements over stateof-the-art models, achieving a 1.14% increase in Dice score for Synapse and 0.99% for ACDC, with minimal computational overhead (only 15 and 30 additional parameters, respectively). GLoG-CSUnet’s flexible design allows integration with various base models, offering a promising approach for incorporating radiomics-inspired feature extraction in Transformer architectures for medical image analysis. The code implementation is available on GitHub at: https://***/HAaiL/GLoGCSUnet.

关键词： Adaptation models Analytical models Laplace equations Image analysis Computational modeling Transformers Feature extraction Synapses Biomedical imaging Radiomics

来源：评论

学校读者我要写书评

暂无评论

Genetic Signal Processing for Categorizing Genomic Data using Convolutional Neural Networks

Genetic Signal Processing for Categorizing Genomic Data usin...

引用

Evolutionary Algorithms and Soft Computing Techniques (EASCT), International Conference on

作者： Padmaja Pulicherla Srinivas Aluvala Munimanda Premchander Poranki V L R N Sai Sudha Revanasiddappa Department of CSE Institute of Technology and Management Hyderabad India School of Computer Science & Artificial Intelligence SR University Warangal India IT Department Malla Reddy Engineering College for Women Hyderabad India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaadeswaram India Department of AI and ML Nitte Meenakshi Institute of Technology Bengaluru India

Breakthroughs in high-throughput sequencing technology have resulted in exponential growth in acquired genetic data. Classic machine learning systems have difficulties in dealing with complex genetic data, leading to investigation of deep learning methods. A unique technique for categorizing genomic data using convolutional neural networks (CNNs) is described. The genetic signal processing approach can extract unique characteristics from genomic data. CNNs' ability to comprehend subtle patterns and correlations in cleaned-up data results in improved classification accuracy. A publicly accessible dataset is used to compare the proposed technique across various criteria. The study of genetic information using digital noise processing technology is known as genetic signal processing (GSP). GSP are used to turn the nucleotide sequence into a line graph, allowing for appropriate pattern classification. From the results, it clearly shows that proposed CNN achieves better results in terms of accuracy (0.96), precision (0.96), sensitivity (0.95), specificity (0.97) and f1-score (0.95).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Structured Two-stream Attention Network for Video Question Answering

arXiv

引用

arXiv 2022年

作者： Gao, Lianli Zeng, Pengpeng Song, Jingkuan Li, Yuan-Fang Liu, Wu Mei, Tao Shen, Heng Tao Center for Future Media School of Computer Science and Engineering University of Electronic Science and Technology of China China Monash University Australia JD AI Research

To date, visual question answering (VQA) (i.e., image QA and video QA) is still a holy grail in vision and language understanding, especially for video QA. Compared with image QA that focuses primarily on understanding the associations between image region-level details and corresponding questions, video QA requires a model to jointly reason across both spatial and long-range temporal structures of a video as well as text to provide an accurate answer. In this paper, we specifically tackle the problem of video QA by proposing a Structured Two-stream Attention network, namely STA, to answer a free-form or open-ended natural language question about the content of a given video. First, we infer rich long-range temporal structures in videos using our structured segment component and encode text features. Then, our structured two-stream attention component simultaneously localizes important visual instance, reduces the influence of background video and focuses on the relevant text. Finally, the structured two-stream fusion component incorporates different segments of query and video aware context representation and infers the answers. Experiments on the large-scale video QA dataset TGIF-QA show that our proposed method significantly surpasses the best counterpart (i.e., with one representation for the video input) by 13.0%, 13.5%, 11.0% and 0.3 for Action, Trans., TrameQA and Count tasks. It also outperforms the best competitor (i.e., with two representations) on the Action, Trans., TrameQA tasks by 4.1%, 4.7%, and 5.1%. Copyright © 2022, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

COVID-19 Patient Care: A Content-Based Collaborative Filtering Using Intelligent Recommendation System 6th

COVID-19 Patient Care: A Content-Based Collaborative Filteri...

引用

6th Eai International Conference on science and Technologies for Smart Cities, SmartCity 2020

作者： Deebak, B.D. Al-Turjman, Fadi Schoool of Computer Science and Engineering Vellore Institute of Technology Vellore632014 India Department of Artificial Intelligence Engineering Research Center for AI and IoT Near East University Nicosia Mersin 10 Turkey

ISBN: (纸本)9783030760625

COVID-19 is a more transferable illness caused by a new novel coronavirus. It is highly emerging with efficient biosensors such as sensitive and selective that afford the diagnostic tools to infer the disease early. It can maintain a personalized healthcare system to evaluate the growth of disease under proper patient care. To discover as a personalized technology, the healthcare system prefers collaborative filtering. It can effectively deal with cold-start and sparse-data to conduct useful extensions. Due to the continuous expansion of scaling data in a medical scenario, content-based, collaborative filtering, and similarity metrics are preferred. It relies on the most similar social users or threats when the information is large. Many neighbors gain importance to obtain a set of users with whom a target user is likely to match. Forming communities reveal vulnerable users and also reduce the challenges of collaborative filtering like data-sparsity and cold-start problems. Thus, this framework proposes content-based collaborative filtering using intelligent recommendation systems (CCF-IRS) based on high correlation and shortest neighbor in the social community. The result is shown that the proposed CCF-IRS achieves better accuracy than the existing algorithms. © 2021, ICST Institute for computer sciences, Social Informatics and Telecommunications engineering.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Federated Learning via Active RIS Assisted Over-the-air Computation

arXiv

引用

arXiv 2023年

作者： Zhang, Deyou Xiao, Ming Skoglund, Mikael Poor, H. Vincent Division of Information Science and Engineering KTH Royal Institute of Technology Stockholm10044 Sweden Department of Electrical and Computer Engineering Princeton University PrincetonNJ08544 United States

In this paper, we propose leveraging the active reconfigurable intelligence surface (RIS) to support reliable gradient aggregation for over-the-air computation (airComp) enabled federated learning (FL) systems. An analysis of the FL convergence property reveals that minimizing gradient aggregation errors in each training round is crucial for narrowing the convergence gap. As such, we formulate an optimization problem, aiming to minimize these errors by jointly optimizing the transceiver design and RIS configuration. To handle the formulated highly non-convex problem, we devise a two-layer alternative optimization framework to decompose it into several convex subproblems, each solvable optimally. Simulation results demonstrate the superiority of the active RIS in reducing gradient aggregation errors compared to its passive counterpart. © 2023, CC BY-NC-ND.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Ecuador Banana Production & Democratization of Climate Change Machine Learning Models to Mobile Edge Devices!

Ecuador Banana Production & Democratization of Climate Chang...

引用

computer science, computer engineering, & Applied Computing (CSCE), Congress in

作者： Chandrasekar Vuppalapati Anitha Ilapakurti Sandhya Vissapragada Sharat Kedari Vanaja Mamidi Raja Vuppalapati Santosh Kedari Jaya Shankar Computer Engineering (Software Engineering) San Jose State University San Jose USA Data Science and AI Hanumayamma Innovations and Technologies Inc Fremont USA Data Science and Artificial Intelligence Hanumayamma Innovations and Technologies PVT Limited Hyderabad India

Bananas are an important and nutritionally diverse component in diets around the world. Revenue generated from trade in bananas plays an important role with regards to the food import bill of producing countries. Bananas have a particular significance in some of the least developed and low income food-deficit countries, where they contribute not only to household food security as a staple but also to income generation as a cash crop. Research conducted in ten banana producing countries revealed that income from banana farming accounts for some 75 percent of total monthly household income for smallholders. Nevertheless, climate change is impacting banana production in many countries. The impact of climate change not only increases lower yields in some countries but also render severe economic consequences to small holder farmers. As part of this innovative research paper, we provide deployment of complex banana production climate change models to farmers across the world to better plan adaptation strategies and thus reduce the overall negative impact and sustain financial returns.

关键词：

来源：评论

学校读者我要写书评

暂无评论

VizKG: A framework for visualizing SPARQL query results over knowledge graphs 6

VizKG: A framework for visualizing SPARQL query results over...

引用

6th International Workshop on the Visualization and Interaction for Ontologies and Linked Data, VOILA! 2021

作者： Raissya, Hana Darari, Fariz Ekaputra, Fajar J. Faculty of Computer Science Universitas Indonesia Depok Indonesia Tokopedia-UI AI Center of Excellence Jakarta Indonesia Institute of Information Systems Engineering TU Wien Vienna Austria

Despite the rise of the knowledge graph (KG) popularity, understanding SPARQL query results from a KG can be challenging for users. The use of data visualization tools, e.g., Wikidata Query Service and YASGUI, can help address this challenge. However, existing tools are either focused just on a specific KG or only provided as a web interface. This paper proposes VizKG, a framework that provides a wide range of visualizations for SPARQL query results over KGs. VizKG aims to assist users in extracting patterns and insights from data in KGs, and hence supporting further KG analysis. VizKG features a wrapper that links SPARQL query results and external visualization libraries by mapping query result variables to the required visualization components, currently allowing for 24 types of visualizations. Not only that, VizKG also includes visualization recommendations for arbitrary SPARQL query results as well as extension mechanisms for additional visualization types. In our evaluation, the visualization recommendation feature of VizKG achieves an accuracy of 87.8%. To demonstrate the usefulness of VizKG in practical settings, this paper also reports on use case evaluation over various domains and KGs. A Python-based, Jupyter Notebook friendly implementation of VizKG is openly available at https://***/project/VizKG/. Copyright © 2021 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Intelligent Deep Learning Framework for Breast Cancer Prediction using Feature Ensemble Learning 4

Intelligent Deep Learning Framework for Breast Cancer Predic...

引用

4th IEEE Global Conference for Advancement in Technology, GCAT 2023

作者： Rao, Chintala Nageswara Chatrapathy, K. Jainul Fathima, A. Sathish, Goski Mukherjee, Saptarshi Reddy, Pundru Chandra Shaker CSE-AI&ML Geethanjali College of Engineering and Technology Hyderabad India School of Computing&Information Technology REVA University Karnataka Bengaluru India Information Technology Francis Xavier Engineering College TamilNadu Tirunelveli India Information Technology St. Martins Engineering College Dhullapally Hyderabad India M.Tech-Bioinformatics Maulana Abul Kalam Azad University of Technology Uttarakhand India School of Computer Science and Engineering Lovely Professional University Punjab Phagwara India

ISBN: (纸本)9798350305258

The mortality and morbidity rates associated with breast cancer are extremely high. In the fight against breast cancer (BC), a correct diagnosis is crucial. The diagnostic process can benefit from the application of machine-learning (ML) and deep-learning (DL) strategies. In this paper, a deep-recurrent-neural-network (DRNN) technique optimized for breast cancer analysis using RNN and the Keras-Tuner optimization technique is proposed. The input layer, five hidden layers, five dropout layers, and the output layer make up the optimized DRNN. It fine-tuned the dropout layer's neurologist count and firing rate for each successive hidden layer. The most crucial data points have been elected utilizing three different feature-selection methods. The optimized DRNN was contrasted to five standard ML strategies: decision-tree (DT), support-vector-machine (SVM), random-forest (RF), Naïve-Bayes (NB), and the K-nearest neighbor algorithm (KNN). The elected features have been applied to both standard ML models and optimized DRNN. The outcomes showed that, compared to the other models, the cross-validation (CV) and testing outcomes were best achieved (99.16%) by the optimized DRNN with the features picked by univariate. © 2023 IEEE.

关键词： Breast cancer Carcinoma Deep Learning KNN Machine learning RNN SVM

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence-Based Polyp Detection And Measurement Add-On System For Endoscopy

Artificial Intelligence-Based Polyp Detection And Measuremen...

引用

IEEE International Conference on Consumer Electronics (ICCE)

作者： Wei-Cheng Kao Xiu-Zhi Chen Yen-Lin Chen Po-Wen Lu Dept. of Computer Science & Information Engineering National Taipei University of Technology Taipei Taiwan Dept. of Internal Medicine Division of Gastroenterology and Hepatology Shuang Ho Hospital New Taipei Taiwan

ISBN: (数字)9798331521165

ISBN: (纸本)9798331521172

The intelligent medical assistive system has emerged as a topic of considerable interest within the research community in recent years. The implementation of such a system has the potential to significantly reduce the workload of medical personnel while simultaneously enhancing the quality of medical treatment. The automated detection of endoscopic images through computer vision technology represents a significant aspect of the development of intelligent medical assistive systems. In this study, we propose the integration of artificial intelligence-based automated detection and measurement technology into a medical assistive system. The system is capable of detecting polyps in endoscopic images and measuring their sizes simultaneously, as well as calculating the distance to any region of interest in endoscopic images. The detection and measurement performance of the proposed system has been evaluated using self-collected real-world endoscopy images. To demonstrate the clinical unmet need that the proposed system addresses, it was deployed to an edge computing device and integrated with actual endoscopic equipment, which confirmed the effectiveness of the proposed system.

关键词： Performance evaluation Machine learning algorithms Endoscopes Instruments Medical treatment Size measurement Reliability Proposals Personnel Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Standardization of neurotechnology for brain-machine interfacing: State of the art and recommendations

IEEE Open Journal of Engineering in Medicine and Biology

引用

IEEE Open Journal of engineering in Medicine and Biology 2021年 2卷 71-73页

作者： Chavarriaga, Ricardo Carey, Carole Contreras-Vidal, Jose Luis McKinney, Zach Bianchi, Luigi Zurich University of Applied Sciences Zhaw Winterthur Switzerland Ieee Emb Standards Committee Engineering in Medicine and Biology Society Nsf Iucrc Brain University of Houston HoustonTX United States The BioRobotics Institute European Ctr of Excellence in Robotics and Ai Scuola Superiore Sant'Anna Pisa Italy Unified Terminology for Brain-Computer Interfaces Civil Engineering and Computer Science Engineering "tor Vergata" University of Rome Italy

Editorial template does not require an abstract and thus one is not included herein. I would like to thank Kara for her valuable assistance in the preparation of this Editorial and accompanying articles.

关键词： Standardization

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：