检索结果-内蒙古大学图书馆

Classifying white blood cells using combining different convolutional neural networks

Multimedia Tools and Applications 2025年 1-24页

作者： Toptaş, Murat Toptaş, Buket Hanbay, Davut Software Eng. Dept. Engineering and Natural Science Faculty Bandırma Onyedi Eylül University Balıkesir Turkey Computer Eng. Dept. Engineering Faculty Inonu University Malatya44280 Turkey

White blood cells are warrior cells that protect the human body against external factors. Each of these warrior cells performs a distinct task, making every piece of information about them highly valuable in the medical field. In this article, a classification framework for the four known types of white blood cells is proposed. It is hoped that the classification of these types will contribute to the prediction of diseases such as AIDS, malaria, leukemia, and many others. In the proposed method, images of white blood cells from the Blood Cell Classification and Detection dataset were used as input to Convolutional Neural Networks. The feature vectors extracted using these Convolutional Neural Network architectures were combined into a single vector. A Minimum Redundancy Maximum Relevance algorithm was then employed to identify the most effective features within the feature vector. Experiments were conducted using these selected features, and the analysis of each experiment was reported in detail. The Support Vector Machines classifier achieved an accuracy of 98.63% in classifying white blood cell types by combining features from multiple deep learning architectures. The experimental results demonstrated that the features obtained from different layers of the Convolutional Neural Networks had varying impacts on the classification performance. © The Author(s) 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Feature Selection-driven Bias Deduction in Histopathology Images: Tackling Site-Specific Influences

Feature Selection-driven Bias Deduction in Histopathology Im...

引用

Congress on Evolutionary Computation

作者： Farnaz Kheiri Azam Asilian Bidgoli Masoud Makrehchi Shahryar Rahnamayan Dept. of Electrical Computer and Software Engineering Ontario Tech University Oshawa Canada Dept. of Computer Science Wilfrid Laurier University Waterloo Canada Dept. of Engineering Brock University St. Catharines Canada

ISBN: (数字)9798350308365

ISBN: (纸本)9798350308372

The emergence of bias in deep neural models represents a significant reliability concern, which may lead to overoptimistic results on seen data while compromising the model's ability to generalize effectively on unseen datasets. Recent studies conducted on The Cancer Genome Atlas (TCGA), which is a publicly available repository of histopathology images, reveal that the TCGA cancerous features extracted by deep neural networks surprisingly are able to discriminate slides based on their origin sites. This finding undoubtedly indicates the existence of site-specific patterns embedded in the extracted features learned by deep networks rather than focusing on histomorphologic patterns. Consequently, this biased behavior raises concerns about the reliability of these networks. This observation motivates us to conduct a series of experiments in which we present two distinct evolutionary feature selection strategies, each differentiated by its objective function evaluation. The primary goal is to select the features with a minimized foot-print of data source signatures, thereby ensuring a more accurate and site-independent cancer classification. We have conducted nine comprehensive independent experiments across nine cancer types, employing each evolutionary strategy. The comparison between results obtained through evolutionary strategies and the original feature sets highlights the substantial impact of feature selection methods on bias reduction while maintaining accuracy in cancer-type discrimination. Furthermore, the comparison of the two strategies with each other demonstrates the intricate nature of bias and its integration with cancerous features during the training process.

关键词： Training Accuracy Histopathology Training data Feature extraction Prediction algorithms Data models

来源：评论

学校读者我要写书评

暂无评论

A Double MCDM-GIS Analysis for Solar Power Plant Site Selections in Disaster-Prone Area: A case study of Hawkes Bay

A Double MCDM-GIS Analysis for Solar Power Plant Site Select...

引用

Power and Renewable Energy engineering (PREE), International Conference on

作者： Rizki Dian Rahayani Nirmal-Kumar C Nair Dept. Electrical Computer and Software Engineering University of Auckland Auckland New Zealand Dept. Electrical Computer and Software EngineeringUniversity of Auckland Auckland New Zealand

Selecting an appropriate site for solar power plants is a critical long-term investment decision to maximize solar energy generation while minimizing investment and operational costs and ensuring environmental sustainability. While solar irradiance has traditionally been the primary factor in site selection, accounting for disaster risk assessment is becoming increasingly crucial, particularly in countries prone to natural disasters. Hawkes Bay, New Zealand, holds great promise as a region with high solar potential. Yet, it is susceptible to natural disasters like earthquakes, landslides, and floods. This research aims to identify the most suitable solar farm location in Hawkes Bay by integrating estimated solar energy production with disaster risk assessment. The study employs a two-stages multi-criteria decision-making (MCDM) approach, combining data envelopment analysis (DEA) and the analytical hierarchy process (AHP), alongside the utilization of geographical information system (GIS) technology. The finding, among 15 candidate locations, the area in a 5 km radius of Te Haroto weather station has the most potential for solar PV farms. 82 % of the location has low disaster risk, and only 7% of the area is considered high risk.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Wireless Data Transmission: A GAN-based Approach for Time Series Data Restoration

Enhancing Wireless Data Transmission: A GAN-based Approach f...

引用

International Conference on Information Networking

作者： Daejin Han Woongsoo Na Dept. of Computer Science and Engineering Kongju National University Cheonan Korea Dept. of Software Kongju National University Cheonan Korea

ISBN: (数字)9798350330946

ISBN: (纸本)9798350330953

Recent technological advancements have led to a significant increase in the use of wireless communication for data transmission, particularly in sensor networks and device-to-device communications, necessitating not only rapid but also reliable data exchange. In conventional wireless communication, if a packet is lost in transmission due to retransmission, it is retransmitted. In this case, the subsequent packets will wait for them so that the order of the packets is not reversed, which is called Head of Line (HOL) Blocking. This problem can lead to overall transmission delays and losses, and is one of the reasons for inconsistent data transfer times. In this paper, we propose a technique to significantly reduce the retransmission rate of the sender and data loss by retransmission using Generative Adversarial Net (GAN). To evaluate the performance of this study, we applied our technique to a portion of SolarCube solar data assuming that it was corrupted, such as NaN processing, value change, and sign change, and the average recovery rate was 92.36%, demonstrating that it is possible to detect and recover losses in the data transmission process. This research is expected to have applications in data loss detection and recovery during transmission of time series data.

关键词： Wireless communication Wireless sensor networks Time series analysis Propagation losses Generative adversarial networks Data transfer Device-to-device communication

来源：评论

学校读者我要写书评

暂无评论

Enhanced Classification of Alzheimer's Disease Using Deep Learning and Machine Learning Models with MRI Scans

Enhanced Classification of Alzheimer's Disease Using Deep Le...

引用

Frontiers of Information Technology (FIT)

作者： Amna Salahudin Eeman Qadeer Ayesha Ahmed Labiba Gillani Fahad Dept. of Software Engineering FAST National University of Computer and Emerging Sciences Islamabad Pakistan Dept. of Computer Science FAST National University of Computer and Emerging Sciences Islamabad Pakistan

ISBN: (数字)9798331510503

ISBN: (纸本)9798331510510

Alzheimer's disease (AD) is a slowly progressive neurodegenerative disease, which necessitates early and accurate diagnosis for effective management. This research presents a comprehensive approach to the classification of Alzheimer's disease using Magnetic Resonance Imaging (MRI) scans, focusing on four distinct classes: nondemented, mild, very mild, and moderate demented. Our main goals are to combine diverse datasets to improve the robustness and adaptability of the model and to fully investigate Alzheimer's disease in all four stages. To achieve these goals, we employed a combination of machine learning and deep learning models. The models utilized include Histogram of Oriented Gradients (HOG) with Support Vector Machine (SVM), Gabor filters with SVM, Gabor-Gray Level Co-occurrence Matrix (GLCM) with SVM, ResNet50, AlexNet, and InceptionV3. Among these, the InceptionV3 model achieved the highest accuracy of 96.20% using 5-fold crossvalidation followed the Resnet50 model achieving accuracy of 95.43%. Our approach addresses key challenges such as class imbalance, high intra-class variation, inter-class similarities, and dataset variability, demonstrating significant potential in improving Alzheimer's disease classification and detection.

关键词： Support vector machines Deep learning Adaptation models Accuracy Magnetic resonance imaging Focusing Vectors Robustness Alzheimer's disease Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Federated Anomaly Detection

Federated Anomaly Detection

引用

IEEE/IFIP International Conference on Dependable Systems and Networks â€“ Supplemental Volume (DSN-S)

作者： Chunjiong Zhang Byeong-Hee Roh Gaoyang Shan Dept. of AI Convergence Network Ajou University Suwon Korea Dept. of Software and Computer Engineering Ajou University Suwon Korea

ISBN: (数字)9798350395709

ISBN: (纸本)9798350395716

More and more portable intelligent devices are connected to the Internet in recent years. A way to effectively use the isolated cyber data without involving privacy and realize the cyber intrusion anomaly detection on the portable intelligent devices with relatively limited hardware storage resources and computing power is worth exploring. In this paper, we propose a framework of federated anomaly detection, which enables the device effectively detect the anomaly by sharing the parameters of the federated model in a fully distributed fashion. We formulate the model training problem as a distributed robust optimization problem and subsequently devise an efficient algorithm for it. Experimental studies have also been carried out to reveal the superior performance of the proposed framework and underscore the significant benefits of federated anomaly detection.

关键词： Training Performance evaluation Data privacy Federated learning Computational modeling Hardware Internet

来源：评论

学校读者我要写书评

暂无评论

Development of AI-Based Diagnostic Systems for Wound Images

Development of AI-Based Diagnostic Systems for Wound Images

引用

International Conference on Information and Communication Technology Convergence (ICTC)

作者： Seoyeon Jeong Jaeho Kim Kyusung Choi Young-Rae Cho Dept. of Computer and Telecommunications Engineering Yonsei University Wonju Republic of Korea Dept. of Software Yonsei University Wonju Republic of Korea

ISBN: (数字)9798350364637

ISBN: (纸本)9798350364644

Recent advancements in AI technology are revolutionizing the medical field by enhancing diagnostic accuracy, optimizing treatment plans, and providing personalized medical services. AI-based algorithms are particularly effective in analyzing medical images to detect diseases early. Telemedicine, which allows remote medical consultations, benefits from these techniques by improving access to healthcare and enabling accurate diagnoses through efficient data exchange. This work developed an AI-based diagnostic system using a Convolutional Neural Network (CNN) model to analyze wound images. The wound dataset from Kaggle, consisting of 377 images across seven categories, was augmented 25 times to improve model performance. The app allows users to take or upload wound images, which are then classified by the model, providing treatment and aftercare information. This app achieved a high accuracy of 95.716% in classifying wounds and provides basic treatment suggestions. Despite challenges like limited dataset size, the project successfully demonstrated the potential for real-time wound analysis in telemedicine and healthcare. Future improvements should include a more diverse dataset and advanced treatment recommendations to enhance the app's utility.

关键词： Analytical models Accuracy Telemedicine Medical services Wounds Real-time systems Information and communication technology Convolutional neural networks Artificial intelligence Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Improved Residual U-Net for COVID-19 Lung Infection Multi-Class Segmentation in CT Image

Improved Residual U-Net for COVID-19 Lung Infection Multi-Cl...

引用

2022 International Conference of Advanced Technology in Electronic and Electrical engineering, ICATEEE 2022

作者： Khouloud, Sahib Ahlem, Melouah Fadel, Toure Amel, Slim LRI Badji-Mokhtar Annaba University Dept. Computer Science Annaba23000 Algeria Trois-Rivieres Laboratory of Software Engineering Dept. du Quebec Quebec Canada

ISBN: (纸本)9798350346749

COVID-19, the new coronavirus, is a threat to global public health. Today, there is an urgent need for automatic COVID-19 infection detection tools. This work proposes an automatic COVID-19 infection detection system based on CT image segmentation. A deep learning network developed from an improved Residual U-net architecture extracts infected areas from a CT lung image. We tested the system on COVID-19 public CT images. An evaluation using the F1 score, sensitivity, specificity and accuracy proved the effectiveness of the proposed network. Besides, experimental results showed that the proposed network performed well in extracting infection regions so, it can assist experts in COVID-19 infection detection. © 2022 IEEE.

关键词： Coronavirus

来源：评论

学校读者我要写书评

暂无评论

CADS: Cyber-Attack Detection System Using Decision Tree for Automotive Cyber-Physical Systems

CADS: Cyber-Attack Detection System Using Decision Tree for ...

引用

International Conference on Electrical, Communication, and computer engineering (ICECCE)

作者： Seo-Hee Park Youngmi Baek Dept. of Computer Software Engineering Changshin University Changwon-si Korea

Modern vehicles are equipped with a large number of electronic control units (ECUs) to provide more convenient and safe services for drivers. These ECUs are required to share the various control data for driving using the CAN (Controller Area Network) protocol that is one of the communication protocols for in-vehicle networks. Recently, as vehicles become capable of real-time, two-way communication with various elements on the road, modern vehicles faces a wide range of security issues. The in-vehicle networks are already suffered from the cyber-attacks exploiting newly exposed security vulnerabilities. However, it cannot guarantee the safety and security of the vehicles from as the CAN is designed without considering security. Therefore, this paper proposes a machine learning-based cyber-attack detection system (CADS) in CAN to ensure driving safety. This CADS learns the data pattern of the CAN frames generated while the driving and is able to distinguish between regular messages and cyberattack messages. The performance of the fully trained detection model of the CADS is evaluated using a vast number of testing dataset. The results show that our CADS in terms of accuracy, precision, recall, and F1 score has 100% of the detection rate for regular messages, replay attacks, brute-force attacks except for modification messages.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Textual Authenticity in the AI Era: Evaluating BERT and RoBERTa with Logistic Regression and Neural Networks for Text Classification

Textual Authenticity in the AI Era: Evaluating BERT and RoBE...

引用

International Symposium on Electronics and Telecommunications (ISETC)

作者： Layth Rafea Hazim Oguz Ata Dept. of Electrical and Computer Engineering Altinbas University Istanbul Turkey College of Computer Science and Mathematics Tikrit University Tikrit Iraq Dept. of Software Engineering Altinbas University Istanbul Turkey

ISBN: (数字)9798350390865

ISBN: (纸本)9798350390872

AI-generated content impersonating human writing is an issue that has gained attention as AI spreads its wings. This particular study serves as a comparison between the existing Logistic Regression and Feedforward Neural Networks (FNNs) by employing sentence-BERT-appended models and RoBERTa in the determination of authenticity in a given text. A dataset that was balanced between human-written and AI-generated texts was utilized. Techniques like tokenization and normalization were first used, followed by feature extraction using transformer-based models. Cross-validation and confusion matrix analysis that used measures such as accuracy, precision, recall, F1 score, and ROC AUC were included to guarantee the models’ robustness. The hybrid RoBERTa-FNN model that was deposed challengers was the most outstanding model in respect of precision and recall, and the highest accuracy (99.95%) was obtained as mentioned in the data. The improved performance serves as a proof of how effectively RoBERTa uses its embeddings to represent context on the fine-grained level required for this kind of text classification. This work is a stepping stone to the creation of strong AI text detection systems, besides our advancements to the knowledge of models and embedding performance with respect to text classification. The results lay emphasis on the selection of model configuration and the embedding technique, as they are the key factors in achieving the best results in practical applications.

关键词： Training Logistic regression Ethics Accuracy Text categorization Writing Transformers Data models Feedforward neural networks Artificial intelligence

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：