检索结果-内蒙古大学图书馆

The impact of backbone selection in YOLOv8 models on brain tumor localization

Iran Journal of computer Science 2025年 1-23页

作者： Ranjbarzadeh, Ramin Crane, Martin Bendechache, Malika School of Computing Faculty of Engineering and Computing Dublin City University Dublin Ireland ADAPT Research Centre School of Computer Science University of Galway Galway Ireland

This study investigates the utilization of the You Only Look Once (YOLOv8) deep learning framework for accurately identifying the location of brain tumors in medical imaging. We investigate the effects of model size and pretraining on the accuracy and computational efficiency of tumor detection by utilizing different setups of the YOLOv8 model. These setups include various configurations, ranging from very small to large, and can be pretrained on the COCO dataset or not. The experimental results, carried out on Google Colaboratory using NVIDIA Tesla T4 GPUs, show that pretrained models often achieve better performance by utilizing the extensive feature representations learned from the COCO dataset, resulting in increased precision in tumor location. For instance, the YOLOv8-XS model pretrained on COCO achieves an IoU of 0.278 and a Dice coefficient of 0.435, whereas its non-pretrained counterpart attains only 0.241 IoU and 0.388 Dice, indicating a 15% improvement in tumor localization accuracy. Similarly, pretrained YOLOv8-L achieves 0.269 IoU, outperforming standard object detection models such as Mask R-CNN (IoU: 0.212) and Faster R-CNN (IoU: 0.228). These results highlight the impact of pretraining on model performance, particularly for lightweight architectures, while also revealing diminishing returns for larger models. The research uncovers a subtle connection between the complexity of the model, pretraining, and the time required for training. It emphasizes the possible advantages and constraints of pretraining for various sizes of models. © The Author(s) 2025.

关键词： Brain tumor localization COCO dataset Deep learning Object detection Transfer learning

来源：评论

学校读者我要写书评

暂无评论

AI -Driven Detection of Gastrointestinal Disease based on Advanced Deep Learning Techniques 4

AI -Driven Detection of Gastrointestinal Disease based on Ad...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Izankar, Sakshi V. Puri, Chetan Burle, Rushikesh Computer Science and Design Faculty of Engineering and Technology Datta Meghe Institute of Higher Education and Research Maharashtra Wardha India Artificial Intelligence and Machine Learning Faculty of Engineering and Technology Datta Meghe Institute of Higher Education and Research Maharashtra Wardha India

ISBN: (纸本)9798331523923

Esophageal, gastric, and colorectal cancers (CRC) are some of the gastrointestinal (GI) cancers threatening global health with CRC being particularly concerning due to its high fatality rate. While early diagnosis is crucial in improving treatment efficiency, barriers such as image noise and visual subjective assessments during diagnostics such as colonoscopies often render it inconceivable. For better diagnostic accuracy in detecting gastrointestinal disease via endoscopic images, the research explores whether highly advanced artificial intelligence (AI) models such as Vision Transformers (ViTs) and pre-trained DenseNet201 can be effective. We compare these models to popular deep learning techniques such as Support Vector Machines (SVM), ResNet50, VGG19, and GoogleNet. At a training accuracy of 99.9% and a validation accuracy of 98.8%, experiments on the KVASIR dataset indicate that Vision Transformer is superior to other models since it can learn long-range dependencies. The robustness of deep learning models for identifying GI disease is exemplified by the superior performance of DenseNet201, ResNet50, and InceptionV3, all of which possess high accuracy, precision, and recall values. The paper illustrates the potential of Vision Transformers, when combined with transfer learning and pre-trained CNN models, to revolutionize endoscopic diagnoses and provide a valuable resource to help physicians manage gastrointestinal conditions more effectively. © 2025 IEEE.

关键词： Endoscopy

来源：评论

学校读者我要写书评

暂无评论

DoCoFL: Downlink Compression for Cross-Device Federated Learning 40

DoCoFL: Downlink Compression for Cross-Device Federated Lear...

引用

40th International Conference on Machine Learning, ICML 2023

作者： Dorfman, Ron Vargaftik, Shay Ben-Itzhak, Yaniv Levy, Kfir Y. VMware Research United States Viterby Faculty of Electrical and Computer Engineering Technion Haifa Israel

Many compression techniques have been proposed to reduce the communication overhead of Federated Learning training procedures. However, these are typically designed for compressing model updates, which are expected to decay throughout training. As a result, such methods are inapplicable to downlink (i.e., from the parameter server to clients) compression in the cross-device setting, where heterogeneous clients may appear only once during training and thus must download the model parameters. Accordingly, we propose DoCoFL - a new framework for downlink compression in the cross-device setting. Importantly, DoCoFL can be seamlessly combined with many uplink compression schemes, rendering it suitable for bi-directional compression. Through extensive evaluation, we show that DoCoFL offers significant bi-directional bandwidth reduction while achieving competitive accuracy to that of a baseline without any compression. © 2023 Proceedings of Machine Learning research. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Providing and evaluating a comprehensive model for detecting fraudulent electronic payment card transactions with a two-level filter based on flow processing in big data

引用

International Journal of Information Technology (Singapore) 2023年第8期15卷 4161-4166页

作者： Banirostam, Hamid Banirostam, Touraj Pedram, Mir Mohsen Rahmani, Amir Masoud Department of Computer Engineering Central Tehran Branch Islamic Azad University Tehran Iran Department of Electrical and Computer Engineering Faculty of Engineering Kharazmi University Tehran Iran Department of Computer Engineering Science and Research Branch Islamic Azad University Tehran Iran

Previous research on fraud detection modeling is often based on a single algorithm, optimizing categories and clusters to find fraudulent patterns that they have provided unsupervised or supervised methods alone and within the framework of Hadoop. The proposed model, a model based on big data analysis extracts important features of user behavior patterns such as time, device type, values, and type of transaction, and their behavioral modeling. By creating different profiles for users, threshold values will be set for each of them. The proposed model for real-time fraud detection of electronic cash payment cards includes two fast and explicit filters. Fast filtering is the combination of the Hidden Markov Model of the first order and Self Organizing Maps (SOM) for fast transaction processing and the Baum-Welch algorithm to find the local Maximum Likelihood. Explicit filter model training is a combination of multilayer Perceptron Neural Network algorithms and Logistic Regression to create a cardholder profile and measure the amount of new transaction deviation from the created profile with the reduction mapping approach or parallel execution of the model. Also, transactions performed for aggregation filters and prediction of the output results of the Schaffer Demister function with the parameter θ in order to detect the degree of deviation of the transaction behavior from the normal state according to the profile of the customer and the maximum difference between two consecutive sequences of observations is used. Finally, the fraudulent or non-fraudulent label on the transaction will be applied and added to the relevant database for storage for later use. According to the model simulation results, the proposed Accuracy, Precision, Recall, and F1-Score criteria are 0.999, 0.9834, 0.7906, and 0.9214, respectively. Simulation results show that the proposed model will perform better in each of the compared criteria against each of the single algorithms. © 2023, The Author

关键词： Accuracy Electronic card Fraud Supervised algorithm Transaction Unsupervised

来源：评论

学校读者我要写书评

暂无评论

Enhanced Neutrosophic Set and Machine Learning Approach for Kidney Disease Prediction

引用

Neutrosophic Sets and Systems 2025年 80卷 465-477页

作者： Al-Doori, Humam M. Alkellezli, Tareef S. Abdelhafeez, Ahmed Eassaa, Mohamed Sawah, Mohamed S. El-Douh, Ahmed A. Cybersecurity Sciences Department College of Science Ashur University Baghdad Iraq Cybersecurity Engineering Department College of Engineering Ashur University Baghdad Iraq Computer Science Department Faculty of Information System and Computer Science October 6 University Giza12585 Egypt Applied Science Research Center Applied Science Private University Amman Jordan Department of Computer Science Faculty of Information Technology Ajloun National University P.O.43 Ajloun26810 Jordan Cybersecurity Technology Engineering Department College of Engineering Technology Ashur University Baghdad Iraq

Kidney disease (KD) is a gradually increasing global health concern. It is a chronic illness linked to higher rates of morbidity and mortality, a higher risk of cardiovascular disease and numerous other illnesses, and expensive medical expenses. The machine learning (ML) models are applied for KD prediction with higher accuracy and precision. The KD dataset has uncertainty and vague information, so, we used the neutrosophic set (NS) to deal with vague and uncertainty information in the KD dataset. The KD dataset is converted into the N-KD dataset with three membership functions: truth, indeterminacy, and falsity. Three ML models are used in this study such as logistic regression (LR), support vector machine (SVM), and k-nearest neighbor (KNN). These ML models are applied to the N-KD dataset. The results show the LR has higher accuracy and precision on the N-KD dataset than the original KD dataset. © 2025, University of New Mexico. All rights reserved.

关键词： Logistic regression

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Optimization Methods for High-Dimensional Expensive Problems:A Survey

引用

IEEE/CAA Journal of Automatica Sinica 2024年第5期11卷 1092-1105页

作者： MengChu Zhou Meiji Cui Dian Xu Shuwei Zhu Ziyan Zhao Abdullah Abusorrah Department of Electrical and Computer Engineering New Jersey Institute of TechnologyNewarkNJ 07102 USA School of Information and Electronic Engineering Zhejiang Gongshang UniversityHangzhou 310018China School of Intelligent Manufacturing Nanjing University of Science and TechnologyNanjing 210094China Institute of Systems Engineering Macao University of Science and TechnologyMacao 999078China School of Artificial Intelligence and Computer Jiangnan UniversityWuxi 214122China School of Information Science and Engineering Northeastern UniversityShenyang 110819China Center of Research Excellence in Renewable Energy and Power Systems Department of Electrical and Computer EngineeringFaculty of EngineeringKing Abdulaziz UniversityJeddah 21589Saudi Arabia K.A.CARE Energy Research and Innovation Center King Abdulaziz UniversityJeddah 21589Saudi Arabia

Evolutionary computation is a rapidly evolving field and the related algorithms have been successfully used to solve various real-world optimization *** past decade has also witnessed their fast progress to solve a class of challenging optimization problems called high-dimensional expensive problems(HEPs).The evaluation of their objective fitness requires expensive resource due to their use of time-consuming physical experiments or computer ***,it is hard to traverse the huge search space within reasonable resource as problem dimension *** evolutionary algorithms(EAs)tend to fail to solve HEPs competently because they need to conduct many such expensive evaluations before achieving satisfactory *** reduce such evaluations,many novel surrogate-assisted algorithms emerge to cope with HEPs in recent *** there lacks a thorough review of the state of the art in this specific and important *** paper provides a comprehensive survey of these evolutionary algorithms for *** start with a brief introduction to the research status and the basic concepts of ***,we present surrogate-assisted evolutionary algorithms for HEPs from four main *** also give comparative results of some representative algorithms and application ***,we indicate open challenges and several promising directions to advance the progress in evolutionary optimization algorithms for HEPs.

关键词： computer optimization evolutionary

来源：评论

学校读者我要写书评

暂无评论

CQFFA:A Chaotic Quasi-oppositional Farmland Fertility Algorithm for Solving engineering Optimization Problems

引用

Journal of Bionic engineering 2023年第1期20卷 158-183页

作者： Farhad Soleimanian Gharehchopogh Mohammad H.Nadimi-Shahraki Saeid Barshandeh Benyamin Abdollahzadeh Hoda Zamani Department of Computer Engineering Urmia BranchIslamic Azad UniversityUrmia969Iran Faculty of Computer Engineering Najafabad BranchIslamic Azad UniversityNajafabad517Iran Big Data Research Center Najafabad BranchIslamic Azad UniversityNajafabad517Iran Afagh Higher Education Institute Urmia969Iran

Farmland Fertility Algorithm(FFA)is a recent nature-inspired metaheuristic algorithm for solving optimization ***,FFA has some drawbacks:slow convergence and imbalance of diversification(exploration)and intensification(exploitation).An adaptive mechanism in every algorithm can achieve a proper balance between exploration and *** literature shows that chaotic maps are incorporated into metaheuristic algorithms to eliminate these ***,in this paper,twelve chaotic maps have been embedded into FFA to find the best numbers of prospectors to increase the exploitation of the best promising ***,the Quasi-Oppositional-Based Learning(QOBL)mechanism enhances the exploration speed and convergence rate;we name a CQFFA *** improvements have been made in line with the weaknesses of the FFA algorithm because the FFA algorithm has fallen into the optimal local trap in solving some complex problems or does not have sufficient ability in the intensification *** results obtained show that the proposed CQFFA model has been significantly *** is applied to twenty-three widely-used test functions and compared with similar state-of-the-art algorithms statistically and ***,the CQFFA algorithm has evaluated six real-world engineering *** experimental results showed that the CQFFA algorithm outperforms other competitor algorithms.

关键词： Nature-inspired algorithm Farmland fertility algorithm Chaotic maps Quasi engineering optimization problems

来源：评论

学校读者我要写书评

暂无评论

Guest Editorial Introduction to the Special Section on Open Radio Access Networks: Architecture, Challenges, Opportunities, and Use Cases in Vehicular Networks

引用

IEEE Transactions on Vehicular Technology 2024年第7期73卷 9221-9225页

作者： Dev, Kapal Chih-Lin, I. Marojevic, Vuk Khowaja, Sunder Ali Lien, Shao-Yu Wang, Yue Munster Technological University Department of Computer Science China Mobile Research Institute Mississippi State University Electrical and Computer Engineering University of Sindh Faculty of Engineering and Technology National Chung Cheng University Samsung Electronics

Cellular vehicle-to-everything (C-V2X) was introduced to support autonomous driving through 5G and beyond networks. C-V2X leverages cellular network infrastructure to integrate vehicle-to-network, vehicle-to-pedestrian, vehicle-to-infrastructure, and vehicle-to-vehicle communications. It has been suggested that Open RAN can be used to achieve the latency requirements essential to realize C-V2X as it achieves real-time optimization through the use of AI in Near real-time RAN Intelligence Controller (Near-RT RIC). The Open RAN will allow the access to historical traffic data or acquisition of data from vehicles. The data will then be transferred to Near-RT RIC for detecting network anomalies while maintaining reliable communication, which is essential for realizing autonomous driving. Open RAN also supports non-real-time RAN intelligent controller (Non-RT RIC) that allows more complex ML workflows such as policy-based feature extraction and optimization to guide vehicles when real-time acquisition is not available. Open RAN provides support for edge cloud, i.e. Open Cloud that helps to interface the Near-RT RIC with Open RAN central unit's user and control plane. Together, the Open RAN and C-V2X are considered to be the key-enabling technologies for achieving low-latency in autonomous vehicular communication networks. The issue attracted over 100 high-quality submissions from all over the world, among which 12 original contributions were eventually selected for publication. The novelty and key contributions of these articles are summarized as follows. © 1967-2012 IEEE.

关键词： Special issues and sections Vehicular ad hoc networks Cellular networks Autonomous driving 5G mobile communication Radio access networks Open RAN

来源：评论

学校读者我要写书评

暂无评论

The Role of AI and Machine Learning in Cancer Diagnosis and Treatment 4

The Role of AI and Machine Learning in Cancer Diagnosis and ...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Gawande, Tejas Verma, Prateek Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Computer Science and Engineering Maharashtra Wardha442001 India Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Artifical Intelligence and Machine Learning Maharashtra Wardha442001 India

ISBN: (纸本)9798331523923

AI together with ML technology now provides detailed rapid medical data evaluation to enhance cancer diagnosis and treatment methods. Advanced algorithms installed in these technologies help medical staff identify cancer indicators which human experts commonly miss leading to better diagnostic accuracy. AI-powered Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) scan analysis allows doctors to perform quicker and more efficient medical diagnosis. The usage of ML models enables medical professionals to detect prognostic trace elements thus they can enhance treatment choices by analyzing tumor cell biological classification groups. AI actively participates in protein and gene targeting together with clinical trial planning to move personal cancer treatment forward. AI applications face multiple obstacles that stop their effective adoption in the field of oncology. The majority of present-day barriers to AI implementation in oncology stem from problems with data privacy together with algorithmic biases and model validation methods while the high complexity of systems and decreasing model interpretability specifically hinder usage. The success of AI models depends on extensive and diverse datasets although the actual availability of relevant data sets combined with standardization procedures continues to be difficult. Medical diagnostic applications where AI performs its decision-making work present important transparency issues that create doubts about its effectiveness. The analysis of data privacy through federated learning and XAI for interpretability and transfer learning for efficiency and deep learning for accuracy improvement are recent solutions being researched to resolve these problems. The primary aim of this research explores how AI and ML influence cancer diagnosis together with treatment methods while studying present challenges alongside suggested methods to boost their operational performance. The combination of better data governance toget

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Classification of Multi-view mammogram images using a parallel pre-trained models system 8

Classification of Multi-view mammogram images using a parall...

引用

8th IEEE International Conference on Image and Signal Processing and their Applications, ISPA 2024

作者： Kacher, Abdelhafidh Merati, Medjeded Mahmoudi, Saïd University of Tiaret LIM Research Laboratory Computer Science Department Algeria University of Mons Faculty of Engineering ILIA Department Belgium

ISBN: (纸本)9798350309249

Breast cancer is the main lead of women's cancer-related mortalities. Therefore, early detection is imperative for preventing breast cancer from developing to advanced stages. Moreover, the emergence of computer-aided diagnosis systems combined with Deep Learning techniques improve breast cancer diagnosis at its early stages. This study proposes a system for classifying mammogram images scanned from different views based on Deep learning techniques. In this paper, we propose combining four pre-trained models in one system employing the Transfer Learning technique with different input images of the same lesion. This proposed system provides a binary classification of the lesion into malignant or benign classes. The Mini-DDSM dataset images were used in this study;the images were pre-processed and augmented to achieve the best results. Several well-known pre-trained models for image classification (ResNet, Xception, DenseNet, MobileNet, and Inception) were experimented with for the proposed model construction. The study results showed that our proposed model achieved an accuracy of 94.18%, a precision of 94.26%, a recall of 94.16%, and an F1-Score of 94.21%, outperforming the pre-trained model on classifying mammogram images of the same dataset. © 2024 IEEE.

关键词： computer aided diagnosis

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：