检索结果-内蒙古大学图书馆

Optimizing Parkinson’s Disease Detection: Hybrid S-transform-EEG Feature Reduction Through Trajectory Analysis

SN computer science 2025年第2期6卷 1-14页

作者： Afonso, Melina Maria Edla, Damodar Reddy Reddy, R. Ravinder Department of Computer Science and Engineering National Institute of Technology Goa Kottamoll Plateau Goa Cuncolim 403703 India Department of Computer Engineering Goa College of Engineering Farmagudi Goa Ponda 403401 India Department of Computer Science and Engineering Chaitanya Bharathi Institute of Technology Telangana Hyderabad 500075 India

With Parkinson’s disease (PD) becoming more common, there is an increasing demand for accurate technology-driven PD detection. This research attempts to address this requirement. Because of its accessibility, use, and non-invasiveness, electroencephalography (EEG) has shown to be a potential technique for detection. Extraction of a wide range of features from the intricate EEG signals of Parkinson’s disease patients has been made possible by utilizing deep learning algorithms and transformation methods, enabling comprehensive analysis. But this data is enormous and presents computational difficulties, particularly when developing accessible and detection systems. To address this, the research proposes a feature reduction model that uses a hybrid approach to cut down the computational complexity. This approach combines four different feature scoring methods-ANOVA F-value, Mutual Information, Chi-square, and linear regression F-value-to generate a unified ranking score for each feature extracted from S-transformed EEG images. This score is thus generated leveraging the strengths of the diverse mechanisms and used to sort the features. The model decreases the feature set depending on these scores and utilizes a deep neural network to detect PD. This methodically lowers the feature count while monitoring the impact on the accuracy of the neural network. Finding the ideal feature combination is the goal, but maintaining the capacity to detect between people who have and don’t have Parkinson’s disease is a must. By balancing diagnostic accuracy and computational efficiency, the suggested technique ultimately yields accuracy rates of 95.94% and 97.31% on two different datasets with the reduced features, which outperforms other state-of-the art optimization algorithms compared in this research. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Brain computer interface Deep learning Electroencephalography Feature reduction Parkinson’s disease Stockwell-transform

来源：评论

学校读者我要写书评

暂无评论

Detection of Acute Lymphoblastic Leukemia Using a Novel Bone Marrow Image Segmentation

引用

Tsinghua science and technology 2025年第2期30卷 610-623页

作者： M.Anline Rejula Ben M Jebin Ravi Selvakumar S.Amutha Eberlein George Department of Computer Applications-PG Scott Christian College(Autonomous)Nagercoil 629003India Department of Computer Science Malankara Catholic CollegeMariagiri 629153India School of Computer Science and Engineering Vellore Institute of TechnologyChennai 600127India Department of Pediatrics Government Medical College and HospitalGovernment of Tamil NaduNamakkal 637001India

In our study,we present a novel method for automating the segmentation and classification of bone marrow images to distinguish between normal and Acute Lymphoblastic Leukaemia(ALL).Built upon existing segmentation techniques,our approach enhances the dual threshold segmentation process,optimizing the isolation of nucleus and cytoplasm *** is achieved by adapting threshold values based on image characteristics,resulting in superior segmentation outcomes compared to previous *** address challenges,such as noise and incomplete white blood cells,we employ mathematical morphology and median filtering *** methods effectively denoise the images and remove incomplete cells,leading to cleaner and more precise ***,we propose a unique feature extraction method using a hybrid discrete wavelet transform,capturing both spatial and frequency *** allows for the extraction of highly discriminative features from segmented images,enhancing the reliability of *** classification purposes,we utilize an improved Adaptive Neuro-Fuzzy Inference System(ANFIS)that leverages the extracted *** enhanced classification algorithm surpasses traditional methods,ensuring accurate identification of acute lymphoblastic *** innovation lies in the comprehensive integration of segmentation techniques,advanced denoising methods,novel feature extraction,and improved *** extensive evaluation on bone marrow samples from the Acute Lymphoblastic Leukemia Image DataBase(ALL-IDB)for Image Processing database using MATLAB 10.0,our method demonstrates outstanding classification *** segmentation accuracy for various cell types,including Band cells(96%),Metamyelocyte(99%),Myeloblast(96%),***(97%),***(97%),and Neutrophil cells(98%),further underscores the potential of our approach as a high-quality tool for ALL diagnosis.

关键词： Adaptive Neuro-Fuzzy Inference System(ANFIS) blood cells segmentation leukemia classification

来源：评论

学校读者我要写书评

暂无评论

SARCASAM Analysis in Social Networks Using Deep Learning Algorithm 4

SARCASAM Analysis in Social Networks Using Deep Learning Alg...

引用

4th International Conference on Evolutionary Computing and Mobile Sustainable Networks, ICECMSN 2024

作者： Priya, M. Kumar, Vijaya K Vennila, P. Prasanna, M.A. Dept. of Computer Science and Engineering E.G.S. Pillay Engineering College Nagapattinam611002 India Dept. of Computer Science and Engineering Women's Engineering College Puducherry605008 India Dept. of Computer Science and Engineering K. Ramakrishnan College of Technology Trichy621112 India

Twitter is the one of the biggest social media sites, where users may share their thoughts, ideas, and opinions as well as discuss current events and live tweets. In the subject of opinion mining, creating reliable and effective algorithms for sarcasm detection on Twitter is an intriguing task. Sarcasm is the use of positive language to convey depressing emotions while speaking in opposition to one's own intentions. Sarcasm is frequently employed on social networking and micro blogging platforms, where users can offend others and find it difficult to express their true feelings. The deep learning technique utilised in the current algorithms to identify these sarcastic tweets has the limitation of not being able to predict continuous variables. A novel deep learning algorithm is proposed to identify both positive and negative terms as well as sarcasm in comments. Deep neural networks are used to classify the comments into positive and negative word categories. Customers' opinions are mined using sentiment analysis to find and extract information from the text. Sarcastically stated statements from social networking sites can be quickly categorised and recognised by using VADER (Valence Aware Dictionary and Sentiment Reasoner). © 2025 The Authors.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Tightening QC Relaxations of AC Optimal Power Flow Through Improved Linear Convex Envelopes

引用

IEEE Transactions on Power Systems 2025年第2期40卷 1465-1480页

作者： Narimani, Mohammad Rasoul Molzahn, Daniel K. Davis, Katherine R. Crow, Mariesa L. Department of Electrical and Computer Engineering NorthridgeCA91330 United States Georgia Institute of Technology School of Electrical and Computer Engineering AtlantaGA30332 United States Texas A&M University Electrical and Computer Engineering Department College StationTX77840 United States Missouri University of Science and Technology Electrical and Computer Engineering Department RollaMO65409 United States

AC optimal power flow (AC OPF) is a fundamental problem in power system operations. Accurately modeling the network physics via the AC power flow equations makes AC OPF a challenging nonconvex problem. To search for global optima, recent research has developed various convex relaxations that bound the optimal objective values of AC OPF problems. The QC relaxation convexifies the AC OPF problem by enclosing the non-convex terms within convex envelopes. The QC relaxation's accuracy strongly depends on the tightness of these envelopes. This paper proposes two improvements for tightening QC relaxations of OPF problems. We first consider a particular nonlinear function whose projections are the nonlinear expressions appearing in the polar representation of the power flow equations. We construct a polytope-shaped convex envelope around this nonlinear function and derive convex expressions for the nonlinear terms using its projections. Second, we use sine and cosine expression properties, along with changes in their curvature, to tighten this convex envelope. We also propose a coordinate transformation to tighten the envelope by rotating power flow equations based on individual bus-specific angles. We compare these enhancements to a state-of-the-art QC relaxation method using PGLib-OPF test cases, revealing improved optimality gaps in 68% of the cases. © 1969-2012 IEEE.

关键词： Load flow optimization

来源：评论

学校读者我要写书评

暂无评论

A Fine-Tuned DenseNet-Based Diagnosis Approach for Multi-Class Breast Cancer Classification 2

A Fine-Tuned DenseNet-Based Diagnosis Approach for Multi-Cla...

引用

2nd International Conference on Computational Intelligence, Communication technology and Networking, CICTN 2025

作者： Kansal, Kanika Kansal, Kajal Department of Computer Science and Engineering Data Science ABES Engineering College Uttar Pradesh Ghaziabad India School of Computer Science Engineering and Technology Bennett University Uttar Pradesh Greater Noida India

ISBN: (数字)9798331530389

ISBN: (纸本)9798331530389

Breast cancer is a prevalent tumor across women and is associated with a high mortality rate. Prompt diagnosis is one of the biggest challenges that needs to be addressed globally, as it can considerably improve survival rates. An automatic diagnosis system based on deep learning models is critical for improving detection, prediction accuracy, and survivability. This study used various DenseNet-based architectures for multi-class classification using the BUSI dataset. The experiments indicate that the fine-tuned DenseNet201 model outperforms with an accuracy of 98.28%. The results demonstrate the promising capability of the fine-tuned DenseNet201 model to have a high potential for further implementation in practice for achieving early diagnosis and highly individualized approaches for better treatment. The further development of extensive and diverse data sets will be essential for developing new diagnostic methods and improving the existing ones. © 2025 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Attributed Network Clustering through Mutual Affinity Propagated Non-negative Matrix Factorization

引用

SN computer science 2025年第5期6卷 1-17页

作者： Pahari, Saikat Dey, Paramita Department of Computer Science Engineering OmDayal College of Engineering and Architecture West Bengal Howrah 711316 India Department of Information Technology Government College of Engineering and Ceramic Technology West Bengal Kolkata 700010 India

Attributed Network Clustering (ANC) has garnered significant attention in research for identifying communities within a complex network like a social, biological, or information network. Generally, such networks are represented as a graph where the nodes denote entities and edges are relationships between two adjacent nodes. In real-world networks, nodes are associated with attributes to express their properties. In this context, a community is defined as a group of nodes that interact more frequently with each other than with nodes outside their group. ANC involves creating a mapping function that converts nodes and associated attributes into lower-dimensional vector representations. Once transformed into these vectors, various machine learning algorithms are used to cluster the nodes and reveal the underlying communities. However, most ANC methods confront two major challenges. First, they assimilate node attribute information from the local neighborhood, as a result, they can’t explore the mutual affinity between node and attribute. Second, they do not consider the values of the community membership of the nodes at the time of clustering. This leads to poor performance, as the embedding method and clustering methods are independent. To address this, this article proposes a method named Mutual Affinity Propagation-based Nonnegative Matrix Factorization (MAP-NMF). The proposed method works in two phases. In the first phase, an attribute-augmented transition probability matrix is constructed to capture the mutual affinity between the node and attribute and obtain node representation using singular value decomposition of the same. The second phase produces community membership from the obtained latent representation through an NMF-based joint optimization. Moreover, a novel updating rule is proposed for the joint optimization method. Empirical study on both real-world and artificial networks proves the superiority of our proposed algorithm. © The Author(s), under exc

关键词： Attribute augmented graph Attributed network embedding Joint optimization Non-negative matrix factorization

来源：评论

学校读者我要写书评

暂无评论

Data Aggregation Point Placement and Subnetwork Optimization for Smart Grids

引用

computers, Materials & Continua 2025年第4期83卷 407-434页

作者： Tien-Wen Sung Wei Li Chao-Yang Lee Yuzhen Chen Qingjun Fang Fujian Provincial Key Laboratory of Big Data Mining and Applications College of Computer Science and MathematicsFujian University of TechnologyFuzhou350118China Department of Computer Science and Information Engineering National Yunlin University of Science and TechnologyYunlin640301TaiwanChina

To transmit customer power data collected by smart meters(SMs)to utility companies,data must first be transmitted to the corresponding data aggregation point(DAP)of the *** number of DAPs installed and the installation location greatly impact the whole *** the traditional DAP placement algorithm,the number of DAPs must be set in advance,but determining the best number of DAPs is difficult,which undoubtedly reduces the overall performance of the ***,the excessive gap between the loads of different DAPs is also an important factor affecting the quality of the *** address the above problems,this paper proposes a DAP placement algorithm,APSSA,based on the improved affinity propagation(AP)algorithm and sparrow search(SSA)algorithm,which can select the appropriate number of DAPs to be installed and the corresponding installation locations according to the number of SMs and their distribution locations in different *** algorithm adds an allocation mechanism to optimize the subnetwork in the *** is evaluated under three different areas and compared with other DAP placement *** experimental results validated that the method in this paper can reduce the network cost,shorten the average transmission distance,and reduce the load gap.

关键词： Smart grid data aggregation point placement network cost average transmission distance load gap

来源：评论

学校读者我要写书评

暂无评论

Dual-Phase Regressive Deep Neural MapReduce Classifier for Scalable and Accurate Diabetic Prediction

引用

SN computer science 2025年第5期6卷 1-22页

作者： Dhinakaran, D. Edwin Raja, S. Thiyagarajan, M. Selvaraj, D. Ramani, R. Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R & D Institute of Science and Technology Chennai India Department of Computer Science and Engineering R.M.K. Engineering College Tiruvallur India Department of Electronics and Communication Engineering Panimalar Engineering College Tamil Nadu Chennai India Department of Information Technology P.S.R Engineering College Sivakasi India

In the domain of healthcare analytics, accurate and efficient prediction of chronic diseases such as diabetes remains a significant challenge due to the complexity of medical data and the necessity for high precision in classification. Traditional machine learning models struggle with scalability and feature selection, leading to suboptimal performance in large datasets. To address these challenges, we propose the Dual-Phase Regressive Deep Neural MapReduce Classifier (DPRDNMR), an advanced predictive framework that integrates the Enhanced Feature Discriminative Preprocessing Model (EFDPM) for optimal feature selection, bivariant regression for mapping, and a deep neural network for classification, all optimized through a distributed MapReduce framework. Our model demonstrates superior performance across key evaluation metrics compared to conventional classification approaches such as Support Vector Machines (SVM), Deep Neural Networks (DNN), Random Forest, Logistic Regression, and XGBoost. Specifically, using the PIMA Indian Diabetic Dataset, the proposed framework achieves an accuracy of 90.3% for 250 patients, surpassing DNN (88.5%), SVM (86.6%), and other baseline models. Additionally, it exhibits a precision of 88.9%, recall of 99.6%, specificity of 87.1%, and an F1-score of 89.2%, significantly outperforming competing methods by an average margin of 2–5%. These results affirm the robustness of our approach in handling complex medical data with enhanced predictive accuracy and scalability. By leveraging EFDPM for feature selection, distributed processing, and deep learning-based classification, our framework facilitates early and accurate diabetic prediction, thereby assisting healthcare professionals in informed decision-making and improving patient outcomes. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Classification Decision support system Deep neural networks Feature selection Mapreduce Regression analysis

来源：评论

学校读者我要写书评

暂无评论

Enhanced Predictive Modeling for Myocardial Infarction Risk Assessment through Multimodal Data Integration 3

Enhanced Predictive Modeling for Myocardial Infarction Risk ...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Abdulkadar, S. Dharun, S. Kirthivashan, V. Janani, B. K.S. Rangasamy College of Technology Department of Computer Science and Engineering Tiruchengode India

ISBN: (纸本)9798331509675

Since myocardial infarction (MI) continues to be a major cause of death worldwide, precise risk prediction is essential. In order to improve MI risk assessment, this work presents an improved predictive model that makes use of multimodal data, including clinical, genetic, imaging, and lifestyle information. By utilizing sophisticated machine learning methods, specifically the XGBClassifier, the model combines various datasets to reveal intricate patterns, attaining exceptional sensitivity and specificity. Robust reliability testing, a comparison with current approaches, and the use of interpretability tools such as SHAP guarantee that the suggested model provides useful information for clinical decision-making. Addressing dataset biases and extending feature integration to improve generalizability are two potential future steps. © 2025 IEEE.

关键词： AI Model and Accuracy Myocardial infarction streamlit metrics Train XGBClassifier

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Driven Brain Tumor Segmentation from MRI Scans 3

Deep Learning-Driven Brain Tumor Segmentation from MRI Scans

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Dharun, S. Bhuvaneshwaran, P.S. Dineshkumar, R. Kaviarasu, K. K.S.Rangasamy College of Technology Department of Computer Science and Engineering Tiruchengode India

ISBN: (纸本)9798331509675

Brain tumors present significant challenges for medical imaging and diagnostics. Brain tumors from Magnetic Resonance Imaging (MRI) pictures need to be precisely segregated for treatment planning and monitoring. This paper proposes a novel hybrid design that combines the VGG19 and U-Net models to increase the accuracy of brain tumor segmentation. VGG19 is employed for efficient feature extraction, whereas U-Net uses skip connections to help retain spatial information. The hybrid design aims to generate more accurate segmentation results by improving both geographical context and feature representation. © 2025 IEEE.

关键词： Deep 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：