检索结果-内蒙古大学图书馆

Annual International Conference of the IEEE engineering in Medicine and Biology Society (EMBC)

作者： Rui Jin Seung-Jun Kim Dept. of Computer Science & Electrical Engineering University of Maryland Baltimore MD

ISBN: (数字)9798350371499

ISBN: (纸本)9798350371505

A novel data-driven functional magnetic resonance imaging (fMRI) data analysis method is proposed using a deep object-centric learning paradigm. The method can faithfully estimate the variabilities in the spatial neural activation maps, which capture functional interconnections in the brain, over fMRI volumes. The key idea is to treat the component maps composing individual fMRI volumes as "objects," whose latent representations are separately learned by a set of autoencoders. Numerical tests using synthetic and real data sets verify the advantages of the proposed method compared to existing matrix factorization-based approaches.

关键词： data analysis Functional magnetic resonance imaging engineering in medicine and biology Synthetic data

来源：评论

学校读者我要写书评

暂无评论

An Improved U-Shaped Network for ROI Segmentation of Skin Lesion Images

An Improved U-Shaped Network for ROI Segmentation of Skin Le...

引用

Communication & Computing (WCONF), World Conference on

作者： Rasika Gupta Rajeev Kumar Dept. of Computer Science and Engineering Delhi Technological University Delhi India

ISBN: (数字)9798350395327

ISBN: (纸本)9798350395334

Skin cancer, a serious public health concern across the world, is a deadly disease if not diagnosed at an early stage. It is often diagnosed through dermoscopic images that usually have low contrast, irregular boundaries and contains irrelevant artifacts. Due to these factors, the segmentation of Region of Interest (ROI) for effective diagnosis of this disease becomes a challenging task. However, UNet, a U-Shaped Network has proven to be an efficient segmentation model for medical image segmentation, but the result mostly suffers discrepancies with the ground truth segmented mask. Therefore, this paper introduces an improved UNet 3+ architecture with residual block at both encoder and decoder to handle gradient degradation in this deep network, incorporates attention module to focus on relevant features and reduces redundant skip connections. The improved model has been evaluated on publicly available PH2 dataset and achieved an accuracy of 96.62%, dice score of 93.84% and jaccard index of 96.79% on test data. The model surpasses other state-of-the-art UNet models for skin lesion segmentation.

关键词： Image segmentation Accuracy Computational modeling computer architecture Skin Loss measurement Lesions Public healthcare Diseases Skin cancer

来源：评论

学校读者我要写书评

暂无评论

Blockchain Based Credit System to Increase the Creditability through Carbon Offset Level

Blockchain Based Credit System to Increase the Creditability...

引用

Computing, Power and Communication Technologies (IC2PCT), IEEE International Conference on

作者： A Padmavathi G Charan Reddy Gauri Santhosh Dept. of Computer Science and Engineering Amrita Vishwa Vidyapeetham Chennai India

Our country has relied significantly on agriculture as its main source of income for several decades. The demand for food and crops is growing due to the growing population, which offers the agriculture sector tremendous prospects for growth. However, despite the soaring demand and bright future, India’s agriculture industry has failed to demonstrate appreciable profitability. In the paper that follows, we provide a novel strategy meant to revolutionize agricultural marketing networks. Blockchain technology is included in our approach to enable the virtualization of products and carbon offset certificates. This system encourages agricultural producers to reduce carbon emissions and address environmental concerns. By utilizing blockchain in a particular way, carbon offset credits can be used to securely track, trade, and for other fostering sustainability practices. This integration of technologies assures to change how agricultural products are marketed currently. It also helps in promoting sustainability and provides consumers with all the valuable information about the environmental impact of their purchases.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MusicoNet: A Social Network for Musicians Based on the Internet of Musical Things and People Paradigm

MusicoNet: A Social Network for Musicians Based on the Inter...

引用

Internet of Sounds, International Symposium on the

作者： Luca Turchet Jacopo Tomelleri Alessandro Ruffo Dept. Information Engineering and Computer Science University of Trento Trento Italy

ISBN: (数字)9798350366525

ISBN: (纸本)9798350366532

To date, there are a number of online social networks dedicated to musicians. However, these do not truly leverage and capitalize on the musical diversity, i.e., the variability that exists across musicians, their instruments, musical activities and social relations. For instance, existing social networks are not designed to search and find for musicians with specific characteristics related to their profile and, above all, the actual particuliarities of their playing style. More importantly, their integration with the Internet of Musical Things (IoMusT) has been largely overlooked thus far. To bridge these gaps, in this paper we propose MusicoNet, a social network for musicians based on IoMusT technologies. The network leverages Semantic Web methods and is made accessible through an app for smartphones and tablet devices, which can wirelessly interact with smart musical instruments (or, through a laptop, with conventional instruments). MusicoNet was not conceived to exchange topic-oriented communication through textual and photographic posts as it occurs in popular social networks, but to support the search for and connectivity among musicians having given diversity factors. Such a search is not simply based on sole textual queries as it occurs in conventional social networks, but also on content-based queries which can be performed via musical instruments. We describe the technical implementation of MusicoNet, the IoMusT ecosystem it enables, and a preliminary technical validation. We then discuss the lessons learned and future avenues for the proposed technology, which represents the first instance of the recently proposed Internet of Musical Things and People paradigm.

关键词： Semantic Web Bridges Portable computers Social networking (online) Instruments Ecosystems Music Cultural differences Smart phones

来源：评论

学校读者我要写书评

暂无评论

Real-Time Gaze Tracking for Online Examination

Real-Time Gaze Tracking for Online Examination

引用

Pervasive Computational Technologies (ICPCT), International Conference on

作者： Preyanshu Bisht Suresh Kumar Dept. of Computer Science and Engineering Netaji Subhas University of Technology Delhi India

ISBN: (数字)9798331508685

ISBN: (纸本)9798331519476

The exponential growth in online education has increased the demand for automated systems to ensure academic integrity during online examinations. A real-time proctoring system addresses this need by monitoring a student's eye gaze and head movements during an exam. This system leverages Dlib's pre-trained frontal face detector and 68-point facial landmark predictor to detect facial features and track the direction of eye movement and head position. By analyzing these metrics, the system can flag behaviors associated with potential cheating, such as looking away from the screen for extended periods. This paper discusses the design and implementation of the proctoring system, details the detection algorithms, and evaluates its effectiveness for real-time monitoring.

关键词： Measurement Tracking Education Gaze tracking Detectors Real-time systems Detection algorithms Monitoring Facial features Faces

来源：评论

学校读者我要写书评

暂无评论

Enhancing Pneumonia Detection: CNN Interpretability with LIME and SHAP

Enhancing Pneumonia Detection: CNN Interpretability with LIM...

引用

International Conference on Electrical engineering and Information & Communication Technology (ICEEICT)

作者： Md. Hamid Hosen Arnob Saha Altaf Uddin Kahakashan Ashraf Sadia Nawar Dept. of Computer Science & Engineering East Delta University Chittagong Bangladesh

ISBN: (数字)9798350385779

ISBN: (纸本)9798350385786

A bacterial or viral infection of the lungs can cause pneumonia, one of the dangerous and potentially fatal illnesses that can have dire repercussions in a short amount of time. Therefore, a key component of a successful treatment plan is an early diagnosis. Therefore, a sophisticated and automated system that can diagnose chest X-rays and make the process of diagnosing pneumonia easier for both specialists and novices is required. This research aims to create a CNN model that will help with the accurate classification of pneumonia. In this work, we have presented our Deep Learning method for the classification challenge, which is taught using modified images through several pre-processing stages. With an overall accuracy of 93.30%, we were able to classify X-ray images of pneumonia using a custom CNN model. Our proposed model is able to precisely detect pneumonia from X-ray images with amazing accuracy and loss. Furthermore, we employed the LIME and SHAP tools of the XAI technique to generate a noteworthy conclusion to persuade medical practitioners.

关键词： Training Deep learning Pneumonia Computational modeling Predictive models Loss measurement Robustness

来源：评论

学校读者我要写书评

暂无评论

Beta-Sigma VAE: Separating beta and decoder variance in Gaussian variational autoencoder

arXiv

引用

arXiv 2024年

作者： Kim, Seunghwan Lee, Seungkyu Dept. of Computer Science and Engineering Kyung Hee University Korea Republic of

Variational autoencoder (VAE) is an established generative model but is notorious for its blurriness. In this work, we investigate the blurry output problem of VAE and resolve it, exploiting the variance of Gaussian decoder and β of beta-VAE [14]. Specifically, we reveal that the indistinguishability of decoder variance and β hinders appropriate analysis of the model by random likelihood value, and limits performance improvement by omitting the gain from β. To address the problem, we propose Beta-Sigma VAE (BS-VAE) that explicitly separates β and decoder variance σx2 in the model. Our method demonstrates not only superior performance in natural image synthesis but also controllable parameters and predictable analysis compared to conventional VAE. In our experimental evaluation, we employ the analysis of rate-distortion curve and proxy metrics on computer vision datasets. The code is available on https://***/overnap/BS-VAE. Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Work in Progress: Enhancing Students' Learning Experience in Formal Methods through Practical Assignments

Work in Progress: Enhancing Students' Learning Experience in...

引用

IEEE World engineering Education Conference (EDUNINE)

作者： Ruth Schorr Dept. of Computer Science and Engineering Frankfurt University of Applied Sciences Frankfurt Germany

ISBN: (数字)9798331542788

ISBN: (纸本)9798331542795

Formal methods are crucial for ensuring higher integrity levels for safety-critical systems. However, teaching these methods can be quite challenging. Students often show low motivation and are primarily focused on passing formal methods courses with minimal effort. Performance in compulsory formal methods courses is usually below average, with students perceiving the subject as overly mathematical and lacking practical relevance. To address these challenges and enrich the learning experience, we have integrated mandatory group homework assignments into our teaching framework. Students are required to work collaboratively on case studies and present their solutions during class. This work-in-progress paper provides an experience report on enhancing the learning possibilities of master’s students in a model checking course at the Frankfurt University of Applied sciences (FRA-UAS).

关键词： Model checking engineering education

来源：评论

学校读者我要写书评

暂无评论

PassChain: A Blockchain-Based Decentralized Password Manager

PassChain: A Blockchain-Based Decentralized Password Manager

引用

Blockchain and Distributed Systems Security (ICBDS), IEEE International Conference on

作者： Nur Adnan Chowdhury Anik Md. Sabbir Hossain Habibur Rahman Ummay Khadiza Rumpa Joy Datta Rifat Nawaz Sagar Karmoker Md. Adib Khan Mahamudul Hasan Dept. of Computer Science and Engineering East West University Dhaka Bangladesh

ISBN: (数字)9798350354348

ISBN: (纸本)9798350354355

A Nobel approach to the password management system is introduced in this paper, which is through a decentralized system named blockchain. Our goal is to secure people’s passwords by providing them with a secure and distributed password management system named "PassChain." In today’s digital world, a secure password management system is crucial, but it is still challenging. The traditional system lacks security and is a centralized system, so trust issues also arise. Our proposed blockchain-based solution offers full control over users’ passwords. Because of the decentralized method, they are not stored on a single server. Decentralized servers are also more secure than databases. This approach will ensure user security and privacy and will also reduce dependency on third-party services that are prone to vulnerabilities.

关键词： Privacy databases Passwords User interfaces Blockchains Browsers Security Servers Best practices

来源：评论

学校读者我要写书评

暂无评论

Real Estate Price Prediction Using PySpark MLlib

Real Estate Price Prediction Using PySpark MLlib

引用

Smart Power Control and Renewable Energy (ICSPCRE), IEEE International Conference on

作者： Kadam Prajwal Dharmaraj Pradeep Kumar Gupta Keerthana Ajith Mahi Kolli Sangita Khare Niharika Panda Dept. of Computer Science and Engineering Amrita School of Computing Bengaluru India

ISBN: (数字)9798350377002

ISBN: (纸本)9798350377019

Prices of real estates in metropolitan landscapes harbour immense importance in steering the complexities of dynamic city environments. In India, the real estate sector contributes around 6–7% to India's GDP. Therefore, accurate forecasting of real estate prices is a factor in making informed decisions, affecting various stakeholders. Many techniques were used for this in past several years, like Hedonic models, Repeat-Sales models, Rule-Based Systems and Heuristics, etc. This study leverages the power of machine learning and big data to develop a robust framework for predicting housing prices in metropolitan landscapes. We employ PySpark, a powerful big data processing framework with built-in MLlib library (machine learning library), to analyse large-scale housing data encompassing various cities in India & to predict the prices for the same. By implementing a comparative analysis of prominent regression models - Random Forest, Linear Regression, Decision Tree, and Gradient-Boosted Tree - our approach identifies the most effective algorithms for real estate price prediction using MLlib library. Also, this study highlights the need for scalable solutions to manage an increasing number of data sources and emphasizes the PySpark library which will simplify big data handling and enable parallel computing. This study paves the way for utilizing advanced machine learning techniques and big data platforms to gain valuable insights into real estate markets. Our findings emphasize the critical role of combining these powerful tools to navigate the complex dynamics of urban housing and predict prices with greater accuracy.

关键词： Adaptation models Accuracy Machine learning algorithms Urban areas Predictive models Big data Libraries

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：