检索结果-内蒙古大学图书馆

COVID-19 detection from Chest X-ray images using a novel lightweight hybrid CNN architecture

Multimedia Tools and Applications 2025年第13期84卷 11295-11317页

作者： Pradeep Dalvi, Pooja Reddy Edla, Damodar Purushothama, B.R. Dharavath, Ramesh Department of Computer Science and Engineering National Institute of Technology Farmagudi Goa Ponda403401 India Dhanbad India

The pandemic of COVID-19 has affected worldwide population. Diagnosing this highly contagious disease at an initial stage is essential for controlling its spread. In this paper, we propose a novel lightweight hybrid convolutional neural network to identify COVID-19-positive patients from Chest X-ray images using building blocks of MobileNet and ResNet50. This model shows improvement over the base models when applied to the same dataset independently. The proposed model is trained and tested on same configuration and dataset as that of its base models, for better performance analysis and comparison. It also has fewer parameters as compared to MobileNet and ResNet50, which makes it lightweight, adding to its better performance. It achieves an accuracy of 98.47 % with very few false negative predicted samples, which is extremely beneficial in medical diagnosis. Concerning the parameters, this model is built with only 200,034 trainable parameters, which marks a parameter reduction percentage of 99.15 % concerning ResNet50 and 93.82% compared to MobileNet. The generalisation performance of our proposed model is evaluated using stratified 5-fold cross validation technique. Our model achieved very good accuracy with very less parameters as compared to various state-of-the-art models. The model also classified Pneumonia images from Normal ones with an accuracy of 98.9 %. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Using breast density for hybrid region and pixel-level loss function

引用

Machine Vision and Applications 2025年第4期36卷 1-13页

作者： Aliniya, Parvaneh Nicolescu, Mircea Nicolescu, Monica Bebis, George Department of Computer Science and Engineering University of Nevada Reno Virginia Street RenoNV89557 United States

Breast cancer continues to be one of the most lethal cancer types, mainly affecting women. However, thanks to the utilization of deep learning approaches, there has been a considerable boost in the performance of the methods for breast cancer detection. The loss function is a core element of any deep learning architecture with a significant influence on their performance. The loss function is particularly important for tasks such as breast mass segmentation. For this task, challenging properties of input images, such as pixel class imbalance, may result in instability of training or poor detection results due to the bias of the loss function toward correctly segmenting the majority class. We propose a hybrid loss function incorporating both pixel-level and region-level losses, where the breast tissue density is used as a sample-level weighting signal. We refer to the proposed loss as Density-based Adaptive Sample-Level Prioritizing (Density-ASP) loss. Our motivation stems from the observation that mass segmentation becomes more challenging as breast density increases. This observation makes density a viable option for controlling the effect of region-level losses. We also propose to evaluate the method using automated density estimation approaches. To demonstrate the effectiveness of the proposed Density-ASP, we conducted mass segmentation experiments using two publicly available datasets: INbreast and CBIS-DDSM. Our experimental results demonstrate that Density-ASP improves segmentation performance compared to the commonly used hybrid losses across multiple metrics. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

A Probabilistic Reasoning Framework to Detect Fake News on Social Media 19

A Probabilistic Reasoning Framework to Detect Fake News on S...

引用

19th International Conference on Ubiquitous Information Management and Communication, IMCOM 2025

作者： Rashid, Mehreen Anber, Fabliha Samiullah, Md Ahmed, Chowdhury Farhan Leung, Carson K. Pazdor, Adam G.M. Department of Computer Science and Engineering University of Dhaka Dhaka Bangladesh Department of Computer Science University of Manitoba WinnipegMB Canada

ISBN: (纸本)9798331507817

The rapid growth of social media has made the Internet a critical platform for spreading misinformation, which shapes public opinion and harms society. Despite significant research in fake news detection, most probabilistic efforts rely heavily on Naive Bayes, with limited exploration of other probabilistic models. This paper introduces a Bayesian network (BN) modelling-based framework for fake news detection, offering a probabilistic estimate of the likelihood of news being false. Unlike binary classification, this approach reflects human decision-making by evaluating three key questions: "who", "what", and "when". Each module corresponds to a specific feature set, enabling nuanced reasoning about the news’s credibility. The framework is flexible, allowing adjustments through expert input, knowledge bases, or real-world data. We validate the approach using a semi-synthetic dataset containing features from news content, user behaviour, and social context. The results highlight the framework’s capacity to leverage expert knowledge, providing a more reliable and adaptive solution compared to traditional classifiers. The BN-based method demonstrates enhanced robustness, positioning it as a promising tool for tackling misinformation in an evolving digital landscape. © 2025 IEEE.

关键词： Bayesian network Credibility Fake news Information management Social media

来源：评论

学校读者我要写书评

暂无评论

Emotion detection in text using a legendre memory unit based deep learning framework

引用

Multimedia Tools and Applications 2025年 1-16页

作者： Khan, Abrar Kumar, Prabhat Department of Computer Science and Engineering National Institute of Technology Patna Ashok Rajpath Bihar Patna800005 India

Emotion detection stands as a pivotal approach for extracting unbiased insights from textual documents. This process empowers business experts to identify areas of improvement and bolster key facets of their operations. Given that every online customer post holds the potential to influence the market and consumer behaviour, the significance of emotion recognition has surged. This article introduces a cutting-edge deep learning (DL) approach for emotion detection, specifically focusing on analyzing tweet text sourced from the CARER dataset. The proposed method employs a two-step process, utilizing sentence-transformer for text embedding and Legendre Memory Unit (LMU) based DL framework for emotion classification. The LMU is a novel memory cell designed for recurrent neural networks (RNNs). The key innovation of LMU lies in its ability to achieve "orthogonalization" of its continuous-time history. This means that the memory unit effectively processes input data in a way that retains important patterns while reducing interference from irrelevant or noisy information. The study has undertaken an exhaustive comparison with various state-of-the-art classification models, including RNN, long short term memory (LSTM), bidirectional-long short term memory (Bi-LSTM), gated recurrent unit (GRU), and bidirectional gated recurrent unit (Bi-GRU). Notably, the proposed method distinguishes itself by demonstrating superior performance, surpassing established benchmarks in the field. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

A Two-Level Neural-RL-Based Approach for Hierarchical Multiplayer Systems Under Mismatched Uncertainties

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第3期6卷 759-772页

作者： Zhong, Xiangnan Ni, Zhen Florida Atlantic University Department of Electrical Engineering and Computer Science Boca RatonFL33431 United States

AI and reinforcement learning (RL) have attracted great attention in the study of multiplayer systems over the past decade. Despite the advances, most of the studies are focused on synchronized decision-making to attain Nash equilibrium, where all the players take actions simultaneously. On the other hand, however, in complex applications, certain players may have an advantage in making sequential decisions and this situation introduces a hierarchical structure and influences how other players respond. The control design for such system is challenging since it relies on solving the coupled Hamilton-Jacobi equation. The situation becomes more difficult when the learning process is exposed to complex uncertainties with unreliable data being exchanged. Therefore, in this article, we develop a new learning-based control approach for a class of nonlinear hierarchical multiplayer systems subject to mismatched uncertainties. Specifically, we first formulate this new problem as a multiplayer Stackelberg-Nash game in conjunction with the hierarchical robust-optimal transformation. Theoretical analysis confirms the equivalence of this transformation and ensures that the designed control policies can achieve stable equilibrium. Then, a two-level neural-RL-based approach is developed to automatically and adaptively learn the solutions. The stability of this online learning process is also provided. Finally, two numerical examples are presented to demonstrate the effectiveness of the developed learning-based robust control design. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

NetraAadhaar: A Deep Learning-Driven Aadhaar Verification Platform for the Aid of Visually Impaired

引用

IEEE Access 2025年 13卷 74229-74251页

作者： Patil, Aditya Khan, Tauseef Mollah, Ayatullah Faruk Andhra Pradesh Amaravati522237 India Aliah University Department of Computer Science and Engineering Kolkata700160 India

Visually impaired individuals face challenges in verifying Aadhaar cards due to the absence of braille, often relying on others for assistance. Despite the widespread use of Aadhaar cards across India, research on assistive technologies for such users remains limited, making this work uniquely significant. To address this, we introduce NetraAadhaar, a deep learning-based mobile application designed to assist visually impaired individuals in Aadhaar card verification. The framework consists of: (i) text region extraction from Aadhaar cards using YOLOv8, (ii) recognition of extracted text via Tesseract OCR engine, (iii) text-to-speech conversion for auditory verification, and (iv) the developed framework is fine-tuned and seamlessly integrated into an end-to-end mobile application for real-time use. NetraAadhaar achieves an mAP-50 score of 92.5% for text detection and an overall text recognition accuracy of 87. 79%, with a precision greater than 90% in key classes, demonstrating its robustness. Compared to traditional OCR-based assistive tools, NetraAadhaar offers higher accuracy, real-time performance, and an end-to-end automated pipeline making Aadhaar verification significantly more accessible and reliable for visually impaired users. NetraAadhaar empowers visually impaired users to verify their Aadhaar cards independently, reducing reliance on others and mitigating risks of identity fraud. © 2013 IEEE.

关键词： Assistive technology

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Review of Block Chain Based Multi Factor Authentication 1st

A Comprehensive Review of Block Chain Based Multi Factor Aut...

引用

1st International Conference on Intelligent Systems, Blockchain, and Communication Technologies, ISBCom 2024

作者： Khan, Andleeb Nand, Parma Bhushan, Bharat Alazzawi, Esraa Mohammad Jamil, Akhtar Department of Computer Science and Engineering Sharda University Greater Noida India Department of Computer Engineering University of Baghdad Al-Jadriya Baghdad10071 Iraq Department of Computer Science National University of Computer and Emerging Sciences Islamabad44000 Pakistan

ISBN: (纸本)9783031823763

The ease and high availability of public and personal data in the existing automated and digital environments have made our technological systems vulnerable to security threats. The conventional approach of passwords and usernames is hence insufficient to ensure the security of data from cyber threats. Multi Factor Authentication can thus ensure effective security thereby minimizing the challenges of Single factor authentication. Blockchain based multi factor authentication (MFA) involves using blockchain for authentication purpose to enhance the security of a system. This paper reviews the recent advancements in the field of multi-factor authentication with blockchain as a means. It analyses the effectiveness of Blockchain based MFA systems by evaluating their strengths and weaknesses. It further studies the gaps in the recent literatures and summarises the future scope for Blockchain and MFA. It discusses techniques such as lightweight cryptographic methods, Arbiter PUF (Public Unclonable function), Clustering, Smart contracts and Machine learning used for authentication purposes. These techniques can be extended to include behavioural factors in multi-factor authentication. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Authentication Protocol

来源：评论

学校读者我要写书评

暂无评论

Beamforming and Reflection Design for Short Packet ISAC With Non-Ideal RIS: An A3C-Based Approach

引用

IEEE Transactions on Vehicular Technology 2025年第5期74卷 8379-8384页

作者： Mahmoudi, Behrad Khonsari, Ahmad Zeinali, Farshad Mili, Mohammad Robat Mashhadi, Mahdi Boloursaz Xiao, Pei University of Tehran Department of Electrical and Computer Engineering Tehran Iran University of Tehran School of Electrical and Computer Engineering College of Engineering Tehran Iran School of Computer Science in the Institute for Research in Fundamental Sciences Tehran Iran United Kingdom Institute for Advanced Innovative Solutions Tehran Iran

Integrated sensing and communication (ISAC) is a promising solution to mitigate the increasing congestion of the wireless spectrum. In this paper, we investigate the short packet communication regime within an ISAC system assisted by a reconfigurable intelligent surface (RIS) to meet the low latency ultra-reliable requirements in the next-generation wireless networks. We consider a non-ideal RIS model that captures effects of the phase-dependent amplitude variations in the reflection coefficients, and we have incorporated the near-field model into the channels between the RIS and the users or targets. In this setup, we jointly design the transmit beamforming and the RIS phase shifts to maximize the sum rate while satisfying the sensing signal-to-noise ratio (SNR) requirement. The system simultaneously carries out multitarget sensing and multi-user short packet communications with the help of the RIS. Considering the non-convex and dynamic nature of the resulting optimization problem, we propose an asynchronous advantage actor-critic (A3C) based method for beamforming and reflection design in this setup. Numerical results demonstrate the superiority of the proposed scheme over the benchmarks. © 1967-2012 IEEE.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

Advanced skin lesion detection via efficientNetB0 and vision transformer model with spatial-aware attention

引用

Multimedia Tools and Applications 2025年 1-19页

作者： Shaheen, Hera Singh, Maheshwari Prasad Singh, Amit Kumar Department of Computer Science and Engineering National Institute of Technology Patna Ashok Rajpath Bihar Patna800005 India

Due to the high incidence and possibly fatal nature of skin cancer, early identification is crucial for enhancing patient results. This paper presents a unique deep learning network, EfficientNetB0 ViT, to accurately classify skin lesions. The proposed method encompasses the scalability and efficiency of EfficientNetB0 with the global pattern recognition capabilities of Vision Transformers (ViT), strengthened by the Spatial-Aware Squeeze and Excitation (SASE) attention mechanism. This innovative technique improves classification performance by enabling the model to dynamically highlight the image’s most pertinent elements. SASE attention successfully recalibrates feature importance, reducing overfitting and improving generalization. This research highlights the effectiveness of SASE attention mechanism enhances skin lesion classification, providing a valuable tool for dermatologists in early diagnosis. The model outperformed numerous methods, achieving a high accuracy of 94.64% as well as average precision of 95%, average recall of 95%, and F1-scores of 95% on HAM10000 dataset. It achieves an accuracy of 93% © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Fake News Detection Using ML Algorithms 5th

Fake News Detection Using ML Algorithms

引用

5th EAI International Conference on Cognitive Computing and Cyber Physical Systems, IC4S 2024

作者： Pokkuluri, Kiran Sree Subha, Atluri Dhatri Sai Varma, Ch. Phaneendra Gurujukota, Ramesh Babu Usha Devi N, S.S.S.N. Raja Rao, P.B.V. Pamarthi, Nagaraju Shalem Raju, P.J.R. Department of Computer Science and Engineering Shri Vishnu Engineering College for Women Bhimavaram India Department of Computer Science and Engineering University College of Engineering JNTU Kakinada Kakinada India

ISBN: (纸本)9783031770807

In response to the growing threat of fake news in our digital era, this paper aims to propose a machine learning system to identify the fake news. Its primary aim is to accurately discern between authentic news articles and deceptive misinformation. The urgency for such a solution arises from the significant consequences of fake news, including the erosion of trust, societal polarization, and potential safety risks. This paper employs machine learning algorithms to analyses textual and contextual elements within news articles. This entails data collection, pre-processing, and model training, followed by a comprehensive evaluation of the model’s performance against established benchmarks. In conclusion, this project is dedicated to fortifying the ongoing struggle against fake news, safeguarding the reliability of information sources, and nurturing a society that thrives on accurate and verified information. The proposed ML system is compared with the baseline methods and results are plotted using various parameters precision, sensitivity, specificity, accuracy and f1score. The proposed model exhibits an accuracy of 96.65% which is promising when compared with existing literature. © ICST Institute for computer sciences, Social Informatics and Telecommunications engineering 2025.

关键词： Adversarial machine learning

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：