检索结果-内蒙古大学图书馆

Diabetic Retinopathy: An Exploration of Retinal Blood Vessel Segmentation Using Multilayered Thresholding

Annals of data science 2025年 1-15页

作者： Babu, K Mahesh Chowdappa, K. Bala Mayee, M. Kiran Rao, Adapa Srinivasa Mounika, Rudrapati Department of Computer Science and Engineering Ravindra College of Engineering for Women Kurnool India Kurnool India Department of Artificial Intelligence & Data Science Koneru Lakshmaiah Education Foundation AP Vaddeswaram India Telangana Maisammaguda India

Diabetic retinopathy is the leading cause of blindness worldwide;it is a consequence of diabetes that affects the retina’s blood vessels. Consequently, correct segmentation of the retinal arteries is essential for accurate diagnosis of such changes in disease progression, which is vital for adequate therapy. A novel approach to segmenting retinal blood vessels is introduced in this research. Starting with the unprocessed retinal picture Utilizing the wavelet transform, a method that incorporates many layers of the threshold approach, to improve samples. New way of brightening the selected vessels is associated with the Wavelet transform, which is effective in representing the multi-scale objects efficiently, and more accurate multilayered thresholding is used to segment the vessels. This approach is specifically aimed at enhancing segmentation of vessels which is exceptionally indispensable for the finding of Diabetic Retinopathy at its beginning phase. To test the proficiency of the proposed technique, different investigations on the openly accessible DRIVE and Gaze information bases. Concerning awareness, particularity, and exactness, that’s what the outcomes show ours outperforms other existing methods. In particular, the proposed technique has been tested and obtained the accuracy of 96%, sensitivity of 97. moderate sensitivity of 86% and specificity of 97% such method can be used for screening of Diabetic Retinopathy in clinical settings. The current review reveals insight into the improvement of strategies utilized in the conclusion of diabetic retinopathy and presents a potentially useful solution for segmenting the retinal blood vessels. © 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

Leaving None Behind: data-Free Domain Incremental Learning for Major Depressive Disorder Detection

引用

IEEE Transactions on Affective Computing 2025年第2期16卷 758-770页

作者： Chen, Tao Guo, Yanrong Hao, Shijie Hong, Richang Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Ministry of Education School of Computer Science and Information Engineering Hefei230009 China

While deep learning techniques have shown promising performance in the Major Depressive Disorder (MDD) detection task, they still face limitations in real-world scenarios. Specifically, given the data scarcity, some efforts have resorted to aggregating data from different domains to expand the data volume. However, their effectiveness is currently limited by the domain gap and data privacy. Additionally, the class imbalance issue is particularly severe in our application, leading to biased classifying performance accordingly. To address these challenges, we propose data-Free Domain Incremental Learning for the MDD detection (DIL-MDD) task, accommodating multiple feature distributions by only accessing well-trained models from previous domains and the data in the current domain. Specifically, DIL-MDD consists of two key modules: Adaptive Class-tailored Threshold Learning (ACTL) and data-Free Domain Alignment (DFDA). The first module measures the discrepancy between the outputs of two sequential domains, based on which we learn a class-tailored threshold adaptively. Building on this, we differentiate between samples that either exhibit similarities or dissimilarities with the previous domain, where this similar sample set is identified to investigate the feature distribution of the historical data. The second module imposes an alignment constraint to narrow the gap between these two sample sets, thereby exploring the expertise of the previous domain. To validate the effectiveness of the proposed method, we conduct extensive experiments on the public MDD datasets, i.e., DAIC-WOZ, MODMA, and CMDC. We also apply our method to another mental health condition, Autism Spectrum Disorder (ASD), to further demonstrate its applicability. Finally, the ablation studies validate the superiority of the proposed modules. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Predictive Modeling of Cardiovascular Disease: Optimizing Neural Networks 8th

Deep Learning for Predictive Modeling of Cardiovascular Dise...

引用

8th International Conference on Information System Design and Intelligent Applications, ISDIA 2024

作者： Hossain, Mohammad Alamgir Amin, Sherif Tawfik Eishah, Hassan Abu Hamdan, Alfadil Ahmed Metwally, Abdelnasser Abdelwakil Limkar, Suresh Department of Computer Science College of Engineering and Computer Science Jazan University Jazan Saudi Arabia Department of Computer Science College of Engineering and Computer Science Jazan University Jazan Saudi Arabia Department of Artificial Intelligence and Data Science AISSMS Institute of Information Technology Maharashtra Pune India

ISBN: (纸本)9789819765805

Cardiovascular diseases (CVD) are a prominent contributor to illness and death on a global scale, underscoring the need for precise predictive models to facilitate timely intervention. The present study investigates the utilization of deep learning methodologies, specifically Convolutional Neural Networks (CNN) and Long Short-Term Memory networks (LSTM), in the context of predictive modeling of cardiovascular diseases. This study examines the efficacy of three well-known optimization techniques, namely Adam Optimization, RMSprop, and Stochastic Gradient Descent (SGD), within the framework of these neural network architectures. Among the various models based on Convolutional Neural Networks (CNNs), Stochastic Gradient Descent (SGD) has been identified as the optimizer that produces the most favorable outcomes for predicting CVD. The utilization of this optimization technique demonstrated exceptional efficacy in the training of the deep neural network, resulting in superior levels of accuracy, sensitivity, and specificity. On the other hand, it was observed that LSTM-based models exhibited the greatest improvement when utilizing RMSprop optimization. The utilization of RMSprop has been found to have a positive impact on the effectiveness of sequence modeling, resulting in enhanced predictive capabilities for assessing the risk of cardiovascular disease. The efficacy of this technique was demonstrated in its ability to capture temporal dependencies within the dataset, consequently enhancing the predictive capability of the model. The results of this study emphasize the importance of carefully choosing neural network architectures and optimization techniques when constructing predictive models for cardiovascular disease. Customizing the selection of neural network architecture and optimization algorithm according to the unique attributes of the dataset can substantially augment the precision and dependability of CVD risk evaluations. This, in turn, can ultimately lead t

关键词： Risk assessment

来源：评论

学校读者我要写书评

暂无评论

EEG time-frequency correlation analysis of different decision stages in the ultimatum game 16

EEG time-frequency correlation analysis of different decisio...

引用

16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2023

作者： Jin, Zhan Gao, Xiang Huang, Zhihua Fuzhou University College of Computer and Data Science China

ISBN: (纸本)9798350330755

The cognitive decision-making process in the ultimatum game (UG) paradigm consists of three stages: 1) option evaluation stage;2) action selection and execution stage;3) outcome evaluation stage. EEG correlation between different stages of UG is an important issue in exploring cognitive decision-making process. However, the ERP-based analysis is not sufficient to reveal EEG correlations at different stages. This study proposes an EEG time-frequency correlation analysis method. This method first calculates Event-Related Spectral Perturbations (ERSPs), and then uses non-parametric statistical tests with the Pearson correlation coefficient as the first-order statistical test quantity to calculate the EEG correlates of different decision stages in different decision states. It was found that there is a correlation between the first and second stages of the decision-making process when accepting a decision. These findings support the division of the decision-making process and indicate differences in different decision states, demonstrating that ERSP is a useful tool for decision analysis. © 2023 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

KubeSPT: Stateful Pod Teleportation for Service Resilience with Live Migration

引用

IEEE Transactions on Services Computing 2025年第3期18卷 1500-1514页

作者： Zhang, Hansheng Wu, Song Fan, Hao Huang, Zhuo Xue, Weibin Yu, Chen Ibrahim, Shadi Jin, Hai Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong China

Container orchestration systems, such as Kubernetes, streamline containerized application deployment. As more and more applications are being deployed in Kubernetes, there is an increasing need for rescheduling - relocating a running pod to different nodes - due to system upgrades, node failures, and load-balancing optimizations. Live migration, which transfers services from source nodes to target nodes with minimal downtime, is the ideal support for rescheduling. However, implementing live migration for pods that run stateful services is challenging, because Kubernetes manages pods as stateless. First, the current pod's network namespace initialization process causes a mismatch in the network state between the migrated pod and internal containers. Second, migrating the memory state results in extended downtime. Third, Kubernetes operations on pods do not consider preserving the state of the pods. Therefore, we propose KubeSPT to achieve live migration of stateful pods in rescheduling scenarios. Firstly, we synchronize the network state of pods and internal containers by controlling packet flow and implement fast service redirection. Secondly, we introduce a Hot data and Lazy-Restore method for memory restoration to reduce migration downtime. Finally, we decouple pod migration operations from other Kubernetes operations to ensure compatibility with live migration. Experimental results show that KubeSPT reduces downtime by 86%-93% compared to current rescheduling methods. © 2008-2012 IEEE.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning with Contrastive Unsupervised Representations for Traffic Engineering in Hybrid SDN 15

Reinforcement Learning with Contrastive Unsupervised Represe...

引用

15th International Conference on Communication Software and Networks, ICCSN 2023

作者： Tang, Qi Yang, Ruiyu Guo, Yingya College of Computer and Data Science Fuzhou University China

ISBN: (纸本)9798350329360

Due to economic costs and hardware limitations of full SDN deployment, partially deploying Software-defined networking (SDN) switches in traditional networks has become the most popular network architecture nowadays. Traffic engineering (TE) promotes the improvement of hybrid SDN performance by optimizing route selection and balancing network flows. Previous TE methods in hybrid SDN focus on centrally searching for the best link weight settings and leveraging heuristic algorithms to calculate the optimal splitting ratio of traffic through SDN switches. Due to the fluctuation of network traffic, previous algorithms fail to learn a routing policy that can adapt to the fast-changing network flows and network performance degrades as a result. Therefore, in this paper, to better learn the traffic features and improve the routing performance, we innovatively propose a TE approach combining contrastive learning and reinforcement learning to optimize routing of network traffic in hybrid SDN. Each agent trains an encoder that well represents the traffic features through contrastive learning and the traffic features are fed into the training of the actor neural network for learning the map between the traffic and routing policy through reinforcement learning. After offline training, the agent deployed on the SDN switch can quickly infer an effective traffic splitting policy that determines the splitting ratio of traffic on the SDN switch. Extensive experiments on three different network topologies show that our proposed algorithm provides significant improvements over state-of-arts. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Energy-Aware Edge Server Placement in Dynamic Scenario Using Reinforcement Learning 15

Energy-Aware Edge Server Placement in Dynamic Scenario Using...

引用

15th International Conference on Communication Software and Networks, ICCSN 2023

作者： Chen, Cen Guo, Yingya College of Computer and Data Science Fuzhou University China

ISBN: (纸本)9798350329360

Mobile edge computing (MEC) provides high-quality network service to mobile users. In a MEC network, the placement of edge servers not only affects the service experience of users, but also has a great influence on the energy consumption of the MEC system. In this paper, we study the edge server placement problem in dynamic MEC scenarios, to minimize the energy consumption of the MEC system in real time, so as to achieve the goal of energy saving at every time. We propose a reinforcement learning (RL) based approach called ESDR. Moreover, in order to make RL suitable for our proposed problem and enable the agent to explore better in MEC scenarios, we introduce Double Q-learning and delaying policy updates to improve the original RL framework. Furthermore, experiments are conducted using a real-world dataset. The results show that ESDR is outstanding in terms of energy saving with fewer edge servers than other approaches in dynamic MEC scenarios. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Customized Movie Recommendation for Enhanced Entertainment

Customized Movie Recommendation for Enhanced Entertainment

引用

2024 International Conference on Cybernation and Computation, CYBERCOM 2024

作者： Minu, M.S. Talapalli, Srikar Chevuri, Saahithi Medhasri, T.S. SRM Institute of Science and Technology Department of Computer Science Engineering Ramapuram Chennai India SRM Institute of Science and Technology Department of Computer Science Engineering with Specialization in Big Data Analytics Ramapuram Chennai India

ISBN: (纸本)9798331542344

The surging popularity of online movie databases has created a challenge for viewers: choosing a film from a massive library can be overwhelming. In this paper, it proposes to design a new hybrid movie recommendation powered by machine learning techniques as a solution to improve user experience. Among those techniques, two are combined into this paper: content-based filtering, using the real features of a movie such as genre and a director, and collaborative filtering, which uses the rating provided by users. This hybrid approach thus improves the intrinsic weaknesses that reside in both these methods individually. On the other hand, content-based filtering tends to reduce exploration to movies of the exact type one is doing well at, whereas, inversely, collaborative filtering settings often mean that a user will be exposed only to films within his or her already-recognized taste profile. It is expected that the proposed hybrid system will solve both problems by balancing this line between exploring familiar cinematic territory and making new recommendations based on both user preferences and the entire collective intelligence of the user base. The approach would aim to enrich user satisfaction by facilitating the selection of movies and providing a richer and more diversified movie-viewing experience. © 2024 IEEE.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Thalaseemia Prediction Using Deep Neural Networks: A Step Towards Personalized Medicine 2

Thalaseemia Prediction Using Deep Neural Networks: A Step To...

引用

2nd International Conference on Machine Learning and Autonomous Systems, ICMLAS 2025

作者： Pream Sudha, V. Karpagavalli, S. Sathiyakumari, K. Sountharraj, A.C. Dr.N. G.P Arts and Science College Department of Computer Science with Data Analytics Coimbatore India PSGR Krishnammal College for Women Department of Computer Science Coimbatore India

ISBN: (纸本)9798331505745

Thalaseemia is a series of genetic hemolytic diseases brought on by defective hemoglobin synthesis. It is prevalent in many Asian, African, and Mediterranean nations. Four million Indians are Thalaseemia carriers, and more than 100,000 are actively combating the condition, according to the World Health Organization. It is specifically taken under investigation because cases of Thalassemia that have clinical, hematologic characteristics cannot be effectively separated from cases of sickle cell anemia and are misdiagnosed as such. The precise profiles of the genetic and pathophysiological mechanisms of patients with Thalassemia are not yet well defined, despite the fact that they are clinically treated identically. Even among people with the same genotype, there are differences in the severity of the clinical phenotype. The frequency of complications changes over time, both within and between individuals. Patients display a striking variation in severity, from a milder clinical course to severe transfusion dependency and progressive organ destruction. This study focuses on the creation of a discriminative model for Thalaseemia classification. In this study, the prediction of Thalassemia is modeled as a pattern classification issue, and the models are constructed using deep learning approaches. The hybrid CNN-Long Short Term Memory (LSTM) based Thalaseemia prediction model has demonstrated improved performance, according to the performance evaluation of these models. © 2025 IEEE.

关键词： Theranostics

来源：评论

学校读者我要写书评

暂无评论

Model-based framework for exploiting sensors of IoT devices using a botnet: a case study with android

引用

Cyber-Physical Systems 2025年第1期11卷 1-46页

作者： Khaliq, Zubair Khan, Dawood Ashraf Baba, Asif Iqbal Ali, Shahbaz Farooq, Sheikh Umar Department of Computer Science University of Kashmir Jammu and Kashmir Srinagar India AI Thndr Dubai United Arab Emirates Department of Computer Science Tuskegee University TuskegeeAL United States Data Engineering hyke.ai Dubai United Arab Emirates

Botnets have become a severe security threat not only to the Internet but also to the devices connected to it. Factors like the exponential growth of IoT, the COVID-19 pandemic, and the ever-larger number of cybercriminals are incentivising the growth of botnets in this domain. The recent outbreak of botnets like Dark Nexus (derived from Qbot and Mirai), Mukashi, LeetHozer, Hoaxcalls, etc. shows the alarming rate at which this threat is converging. The botnets have attributes that make them an excellent platform for malicious activities in IoT devices used by organisations to safeguard the personal and confidential data of their customers, employees, and business partners. The IoT devices have built-in sensors or actuators that can be exploited to monitor or control the physical environment of the entities connected to them, thereby violating the fundamental concept of privacy-by-design of these devices. In this paper, we design and describe a modular botnet framework for IoT. The framework uses an enhanced centralised architecture associated with a novel ‘Domain Fluxing Technique’. The proposed framework will provide insights into how privacy in IoT devices can be incorporated at design time to check the sensors and actuators in these devices against malicious exploitation, consequently preserving privacy. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Internet of things

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：