检索结果-内蒙古大学图书馆

Alter.a.ive Strategies to Generate Class Activation Maps Supporting AI-based Advice in Vertebral Fracture Detection in X-ray Images

引用

Methods of infor.a.ion in medicine 2024年第3-04期63卷 122-136页

作者： Samuele Pe Lorenzo Famiglini Enrico Gallazzi Chandra Bortolotto Luisa Carone Andr.a.Cisarri Alberto Salina Lorenzo Preda riccardo Bellazzi Federico Cabitza Enea Parimbelli Department of Electrical Computer and Biomedical Engineering University of Pavia Pavia Italy. Department of Informatics Systems and Communication University of Milano-Bicocca Milan Italy. ASST G. Pini - CTO Milan Italy. Unit of Radiology Department of Clinical Surgical Diagnostic and Pediatric Sciences University of Pavia Pavia Italy. Department of Radiology I.R.C.C.S. Policlinic San Matteo Foundation Pavia Italy. Department of Reconstructive Surgery and Osteo-articular Infections C.R.I.O. Unit I.R.C.C.S. Galeazzi Orthopaedic Institute Milan Italy. Telfer School of Management University of Ottawa Ottawa Ontario Canada.

Balancing artificial intelligence (AI) support with appropr.a.e human oversight is challenging, with associated risks such as and . research ar.a. like eXplainable AI (XAI) and Frictional AI aim to address these challenges. Studies have shown that presenting XAI explanations as "juxtaposed evidence" supporting contrasting classifications, rather than just providing predictions, can be *** study aimed to design and compare multiple pipelines for generating juxtaposed evidence in the form of class activation maps (CAMs) that highlight ar.a. of interest in a fracture detection task with X-ray *** designed three pipelines to generate such evidence. The pipelines are based on a fracture detection task from 630 thoraco-lumbar X-ray images (48% of which contained fractures). The first, a appr.a.h, uses an algorithm of the Grad-CAM family applied to a resNeXt-50 network trained through transfer learning. The second, a appr.a.h, employs two networks-one optimized for sensitivity and the other for specificity-providing targeted explanations for positive and negative cases. The third, a appr.a.h, leverages autoencoders to cr.a.e activation maps from feature tensors, extracted from the raw images. Each appr.a.h produced two versions of activation maps: AM3-as we termed it-which captures fine-grained, low-level features, and AM4, highlighting high-level, aggregated features. We conducted a validation study by comparing the generated maps with binary ground-truth masks derived from a consensus of four clinician annotators, identifying the actual locations of fractures in a subset of positive *** proved to be the best performing Grad-CAM var.a.t and was used in both the single- and dual-model strategies. The generative appr.a.h demonstrated the gr.a.est over.a. with the clinicians' assessments, indicating its ability to align with human *** results highlight the potential of Judicial AI to enhance diagnostic decision-making and foster.a.s

关键词： eXplainable artificial intelligence class activation map clinical decision support systems medical imaging radiology

来源：评论

学校读者我要写书评

暂无评论

Effects of robotic Expertise and Task Knowledge on Physical Ergonomics and Joint Efficiency in a Human-robot Collaboration Task

Effects of Robotic Expertise and Task Knowledge on Physical ...

引用

IEEE-rAS Inter.a.ional Conference on Humanoid robots

作者： Matteo Pantano Ar.a.na Curioni Daniel regulin Tobias Kamps Dongheui Lee Technology department Siemens Aktiengesellschaft Munich Germany Department of Electrical and Computer Engineering Human-Centered Assistive Robotics Technical University of Munich Munich Germany Autonomous Systems Institut für Computertechnik Technische Uni-versitat Wien Vienna Austria Institute of Robotics and Mechatronics German Aerospace Center Wessling Germany

With the trend of low batch manufacturing, more and more small and medium enterprises are leaning towards adopting collaborative robots to incr.a.e productivity and improve operator well-being. However, robots are often programmed by robot experts rather than the operators effectively working with the machine. Therefore, operators may perceive low levels of task autonomy due to the unpredictability of robot motions. Empowering operators to make their own choices regarding robot motions can improve such feelings. However, research in cognitive science shows that allowing operators to decide on robot motions in a collaborative task could be influenced by how people consider their travel path and their partner's action. To better understand these relations, considering preliminary results from a previous study, we designed a user study where we tested operators' decisions in a collaborative task where groups of robot experts and novices were asked to choose their preferred task configuration among four possible options that differed in terms of operators' physical ergonomics and robot travel path. Our results show that robotic experts prioritize joint team work rather than their ergonomics. Contrarily, novices prioritize individual efforts and tend to reduce the robot travel path while keeping their travel path constant, maintaining good physical ergonomics. In conclusion, providing operators with task decision autonomy can be advantageous, but operator background must be considered to ensure optimal physical ergonomics and travel paths.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Machine Learning and Wireless Sensor Network based Library Automation System

Machine Learning and Wireless Sensor Network based Library A...

引用

Inter.a.ional Conference on Power, Energy, Control and Transmission Systems (ICPECTS)

作者： J Manikandan T M Sivanesan Gunasekaran P S ramapraba Department of Electronics and Communication Engineering Sri Sairam Engineering College Chennai Tamil Nadu India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai Tamil Nadu India Department of Electronics and Communication Engineering Sidhartha Institute of Science and Technology Puttur Andhra Pradesh India Department of Electrical and Electronics Engineering Panimalar Engineering College Chennai Tamil Nadu India

ISBN: (纸本)9781665462761

remote sensor network and radio frequency identification innovation are on the world's outskirts. In this age of data innovation, the library as a repository for infor.a.ion gathering and distribution is critical. Innovations such as the personal computer, the Internet, and univer.a. identification have sped the shift to the digital age. New improvements were integrated to configure radio Frequency Identification remote or.a.ization with Wireless sensor network comprising hubs to work on the intelligent administration of the library, relying on manual and data the board. The three modules that were basically researched were writing seek positioning framework, book full casing framework, and programmed getting and returning entry monitor warning framework. The intelligent library framework based on Wireless sensor or.a.ization and radio Frequency Identification incorporated sophisticated rFID and ZigBee technologies for the simplification and inclusion of writing assets, administration inter.a.e, and the library's administrative procedure. The significance of library computerization and smart collaboration was recognized, and the notion of individuals situated was fully explored. According to the findings, the smart library framework, which is based on wireless sensor or.a.ization and radio frequency identification, was able to fundamentally understand the smart library executives. To summarize, the framework cr.a.es an unpredictable library administration and board framework at any time and in any location.

关键词： Wireless communication Wireless sensor networks Technological innovation Zigbee Or.a.izations Writing Libraries

来源：评论

学校读者我要写书评

暂无评论

A Novel Appr.a.h for Crop Yield Prediction based on Hybrid Deep Learning Appr.a.h

A Novel Approach for Crop Yield Prediction based on Hybrid D...

引用

Inter.a.ional Conference on Communication and Electronics Systems (ICCES)

作者： B. Srinivasa rao Priya Priya Seemantini Nadiger Sandeep rout Khushal N. Pathade Kamlesh Singh Department of Computer Science and Engineering Hyderabad Telangana India School of Agriculture Graphic Era Hill University Dehradun India Department Of Agriculture Division Of Agronomy School Of Agricultural Science Karunya Institute Of Technology And Sciences Coimbatore Tamilnadu India Faculty of Agriculture Sri Sri University Cuttack Odisha India P.G. Department of Botany Dr. R. G. Bhoyar Arts Commerce and Science College Wardha Maharashtra India Department of Computer Science and Engineering Graphic Era Hill University Dehradun India

The agricultural sector is crucial to the economic development of our country. Civilization's birth was facilitated by agricultural practices. The agricultural sector is vital to India's economy because of the country's status as an agrar.a. nation. So, agriculture has the potential to serve as the economic foundation of our nation. In agriculture planning, crop selection is crucial. Our Indian economy desperately needs widespr.a. reforms in the agricultural sector. In this proposed appr.a.h to use several machine learning methods to forecast future agricultural yields. After receiving the input image, the null values can be filtered out using the preprocessing appr.a.h. The relief method is then used to choose features. In order to extract features, a linear discriminant analysis appr.a.h is used. Finally, the CNN-BiLSTM-ECA model, which combines a CNN, a Bidirectional Long Short-Term Memory network, and an Attention Mechanism, is presented for use in training (AM). To reduce the impact of excessive noise and nonlinearity, CNN has been used to extract deep aspects of agricultural productivity. Crop yield is predicted using a BiLSTM network trained on the recovered deep characteristics. This proposed also implement an unique Efficient Channel Attention (ECA) module to incr.a.e the network model's sensitivity to key features and inputs. The average error made by each method is compared to one another. Farmers will be able to use the CNN-BiLSTM-ECA forecast to guide their planting decisions by taking into account var.a.les like expected temperatures, precipitation, available land, and more.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning Inter-Lingual Document representations via Concept Compression 22nd

Learning Inter-Lingual Document Representations via Concept ...

引用

22nd Inter.a.ional Conference on Intelligent Data engineering and Automated Learning, IDEAL 2021

作者： Lenz, Marc Tashu, Tsegaye Misikir Horváth, Tomáš Faculty of Informatics Department of Data Science and Engineering Telekom Innovation Laboratories ELTE - Eötvös Loránd University Pázmány Péter Sétány Budapest1117 Hungary College of Informatics Wollo University Kombolcha Institute of Technology Kombolcha208 Ethiopia Faculty of Science Institute of Computer Science Pavol Jozef Šafárik University Jesenná 5 Košice040 01 Slovakia

ISBN: (纸本)9783030916077

In this work, we proposed a novel appr.a.h to derive inter-lingual document representations. The introduced methods aim to enhance the quality of content-based Multilingual Document recommendation and infor.a.ion retrieval Systems. The main idea centers around cr.a.ing inter-lingual representations by using mappings to align monolingual representation spaces. According to the experimental results carried out on JrC-Acquis and EU bookshop multilingual corpora, the proposed concept compression appr.a.h has outperformed the traditional cross-lingual retrieval and recommendations methods. © 2021, Springer Nature Switzer.a.d AG.

关键词： Infor.a.ion retrieval systems

来源：评论

学校读者我要写书评

暂无评论

S-COAP: Semantic Enrichment of COAP for resource Discovery

引用

SN computer Science 2020年第2期1卷 1-7页

作者： Vandana, C.P. Chikkamannur, Ajeet A. Visvesvaraya Technological University Research Scholar Department of Information Science Engineering New Horizon College of Engineering Bangalore India Department of Computer Science and Engineering R. L. Jalappa Institute of Technology Doddaballapur Bangalore India

Internet of Things (IoT) paradigm is progressing at an enormous rate cr.a.ing an expansive ecosystem of interconnected smart devices which needs to be controlled and managed by end applications or users. So, the solution is to discover the best device matching the user requirement in timely manner. Ensuring efficient and context aware resource discovery is a challenge in seamless operation of IoT applications. Semantic Intelligence can be built for resource discovery. In this paper, we propose semantic enrichment of COAP protocol to discover the resources. resources are semantically modelled, and COAP protocol COrE Link for.a. is modified to incorporate semantic infor.a.ion. Also, we have proposed a framework which is aware of the social relationship between the devices to accommodate distributed topologies. We conducted the experiments in COOJA simulator on CONTIKI OS with COAP protocol semantically enriched. COAP COrE Link for.a. and resource Directory were semantically enriched, and its effectiveness was measured. © 2020, Springer Nature Singapore Pte Ltd.

关键词： COAP COrE Link IoT Ontology rD Semantic Social

来源：评论

学校读者我要写书评

暂无评论

Enhanced Electrochemical Sensing of Dopamine and Uric Acid Using a Sol-Gel Fabricated MWCNT Ar.a.-Modified Carbon Electrode

引用

Jour.a. of The Electrochemical Society 2025年第6期172卷 067523-067523页

作者： Abdulrahman Saad Alqahtani Hashim Elshafie Alavudeen Basha A Azath Mubarakali Parthasarathy P Venkatesh M College of Computing and Information Technology University of Bisha Bisha - 67714 Saudi Arabia Department of Computer Engineering College of Computer Science King Khalid University Abha - 61421 Saudi Arabia Adjunct Faculty Department of Biomedical Engineering Excel Engineering College Namakkal Tamil Nadu India Department of Informatics and Computer Systems College of Computer Science King Khalid University Abha - 61421 Saudi Arabia Department of ECE CMR Institute of Technology Bengaluru Karnataka - 560037 India

In this study, a glassy carbon electrode (GCE) modified with a vertically aligned ar.a. of multiwalled carbon nanotubes (MWCNTs) was fabricated via a temperature-controlled sol-gel method. The vertically aligned MWCNT structure significantly enhances the electrode's sur.a.e modification efficiency, offering superior electrochemical perfor.a.ce compared to conventional methods. The resulting MWCNT ar.a.-modified GCE demonstrates excellent selectivity by clearly differentiating the anodic oxidation potentials of dopamine (DA) and uric acid (UA), even in complex biological matrices. To further improve the electr.a.tive properties, the modified electrode was pre-tr.a.ed with ethanol and acetone prior to electrochemical measurements. remar.a.ly, this pre-tr.a.ment enabled the detection of an additional two-electron oxidation process during dopamine oxidation, associated with the conversion of 5,6-dihydroxyindole to indole-5,6-quinone—a feature not typically observed in other CNT-based electrodes. The optimized sensor exhibited high sensitivity and selectivity for the simultaneous detection of DA and UA in human serum and urine samples, achieving recovery rates ranging from 95% to 110%. These results highlight the potential of the sol-gel fabricated MWCNT ar.a.-GCE as an efficient and reliable platform for.a.vanced biosensing applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Quadratic stabilisation of switched affine systems

引用

Jour.a. of Control and Decision 2020年第1期7卷 1-23页

作者： Minqing Xiao Guisheng Zhai Chi Huang College of Mathematics and Informatics Fujian Key Laboratory of Mathematical Analysis and ApplicationFujian Normal UniversityFuzhouPeople’s Republic of China Department of Mathematical Sciences College of Systems Engineering and ScienceShibaura Institute of TechnologyTokyoJapan Department of Computer Science and Technology School of Economic Information EngineeringSouthwestern University of Finance and EconomicsChengduPeople’s Republic of China

Minqing Xiao received the *** from Chongqing University,Chongqing,China,in *** is currently a Professor with the College of Mathematics and informatics,Fujian Nor.a. University,Fuzhou,*** current research interests include robust control/filter theory,delta operator systems,networked control systems,and switched systems;Guisheng Zhai received his *** from Fudan University,China,in 1988,and he received his *** ***,both in system science,from Kobe University,Japan,in 1993 and 1996,*** two years of industr.a. experience,*** moved to Wakayama University,Japan,in 1998,and then to Osaka Prefecture University,Japan,in *** held visiting professor positions at University of Notre Dame from August 2001 to July 2002,and at Purdue University from March 2016 through Febr.a.y *** April 2010,he joined the faculty board of Shibaura institute of technology,Japan,where he currently is a full Professor of Mathematical *** research interests include large scale and decentralised control systems,robust control,switched systems and switching control,networked control and multi-agent systems,neural networks and signal processing,*** is on the editor.a. board of several academic jour.a.s including IET Control Theory&Applications,Inter.a.ional Jour.a. of Applied Mathematics and computer Science,Jour.a. of Control and Decision,and Frontiers of Mechanical *** is a Senior Member of IEEE,a member of ISCIE,SICE,JSST and JSME.

关键词： Switched affine systems(SAS) switched uncer.a.n affine systems(SUAS) quadratic stabilisation convex combination state-dependent switching output-dependent switching convergence set LMIs norm bounded uncer.a.nties

来源：评论

学校读者我要写书评

暂无评论

Design and Comparison of Collaborative Filtering technology for Product Suggestions in E-Commerce

Design and Comparison of Collaborative Filtering Technology ...

引用

Automation, Computing and renewable Systems (ICACrS), Inter.a.ional Conference on

作者： P. rajasekar B. Mohanraj S. N. Padhi N. Sivakumar Lavanya J Charles Prabu V Department of Data Science and Business Systems School of Computing SRM Institute of Science and Technology Chennai Tamil Nadu India Department of Information Technology Karpagam College of Engineering Coimbatore Tamil Nadu India Department of Mechanical Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram Andhra Pradesh India Department of Computer Science and Engineering Panimalar Engineering College Chennai City Campus Chennai Tamil Nadu India Department of Fashion Designing SRM Institute of Science and Technology CSH Kattankulathur Campus Chennai Tamil Nadu India Department of Computer Science Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai Tamil Nadu India

ISBN: (纸本)9781665460859

E-commerce websites and internet purchasing have never been more necessary. Product recommendations are a key tactic employed by online buying platforms. This study compares two different filtering algorithms that are used in product recommendationson websites like Amazon and Flipkart. reviews of electrical devices, user data, and timestamps were all gathered from Kaggle and put into a dataset. These specifics then undergo several preprocessing steps. The steps are data visualization, duplicate value analysis, missing value analysis, and datatype analysis. The exhibited data is then used to train and test the machine learning models. For this purpose, two machine learning models are developed using two different algorithms. The algorithms K-Nearest Neighbor(KNN) and Support Vector Decomposition (SVD) are employed. The models are then trained to acquire used in the prediction procedure after that. The best models for the prediction process are then determined by scrutinizing the results of both models. The root Mean Square Error (rMSE), Mean Square Error (MSE), and Mean Absolute Error (MAE) values are located and compared because the findings of the analysis weren't *** erroneous data are plotted into a graph and evaluated to further demonstrate the point. As the SVD method is discovered to be the best of the group, the algorithm's gr.a.est and worst predictions are discovered and examined. Even the poorest model constructed for the study's predictions turned out to be superior to several other prediction algorithms used for product suggestions. This algorithm can be deployed into the backend of any e-commerce website making it more effective.

关键词： Training Support vector machines Analytical models Machine learning algorithms Data visualization Machine learning Predictive models

来源：评论

学校读者我要写书评

暂无评论

Power silicon carbide schottky diodes as current mode ?-radiation detectors 32

Power silicon carbide schottky diodes as current mode ?-radi...

引用

32nd IEEE Inter.a.ional Conference on Microelectronics, MIEL 2021

作者： Ilic, S.D. Andjelkovic, M.S. Car.a.al, M.A. Lallena, A.M. Krstic, M. Stankovic, S. ristic, G.S. University of Belgrade Center of Microelectronic Technologies Technology and Metallurgy Serbia IHP-Leibniz-Institut für Innovative Mikroelektronik System Architectures Department Frankfurt Germany University of Granada Department of Electronics and Computer Technology Granada Spain Vincca Institute of Nuclear Sciences Department of Radiation and Environmental Belgrade Serbia University of Niš Faculty of Electronic Engineering Department of Microelectronics Niš Serbia

ISBN: (纸本)9781665445283

In this paper, the feasibility of using commercial power Silicon Carbide (SiC) Schottky diodes as a current mode ?-radiation detector have been examined. Diodes with almost identical electric characteristics are purchased from two different manufacturers, On Semiconductor.a.d roHM. They have been tested under gamma radiation exposure from a Co-60 source. The current response during ir.a.iation has been measured for various dose rates with reversed diode bias. Investigated range of dose rates was from 0.258 Gy/h to 26.312 Gy/h, and reverse diode bias values were 10 V, 20 V, 50 V, 100 V and 200 V. Tested Schottky diodes produce stable current response for the investigated dose rates. Although the manufacturers are different, the results show that the dosimetric characteristics of these diodes have an excellent match. Sensitivity was proportional to the applied reverse bias voltage. A simple power-law can very well describe the dependence of measured radiation-induced current on dose rate. © 2021 IEEE.

关键词： Silicon carbide

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：