检索结果-内蒙古大学图书馆

Robust kidney carcinoma prognosis and characterization using Swin-ViT and DeeplabV3+with multi-model transfer learning

引用

APPLIED SOFT COMPUTING 2025年 170卷

作者： Rehman, Amjad Mahmood, Tariq Saba, Tanzila CCIS Prince Sultan Univ Artificial Intelligence & Data Analyt AIDA Lab Riyadh 11586 Saudi Arabia Univ Educ Dept Informat Sci Vehari Campus Vehari 61100 Pakistan

Chronic kidney disease (CKD) remains a significant health epidemic despite surgical advancements and a global shortage of nephrologists. Medical data analysis and clinical examination can assist healthcare professionals in identifying CKD in its early stages. Computer-based techniques can enhance diagnostic efficiency and accuracy in medical imaging, but current deep-learning-based models have limitations in extracting global information and multi-label classification. The cutting-edge variants of the Vision transformers (ViT) models show promise in capturing information from multiple image regions but require large-scale dataset training. The study proposes a novel approach based on the Swin-ViT model and various modified DL-based pre-trained models (ConvNext, Swint-Unet, MedT, mResNet152V2, EfficientNet-B7, and mVGG19) for prognosis-assisted kidney image and multilabel grading, with changes made to the final layers using the transfer learning paradigm to improve performance and reduce dependence on data scales. Transfer learning can improve classification performance and model generalization in small-scale datasets while reducing training costs. data enhancement and batch size cutting can improve performance, but smaller batches lead to longer training times. The experiment evaluates the Swin-ViT model for multi-label classification of CKD images via transfer learning, demonstrating that it has a stronger multi-label classification performance and that its attention mechanism is beneficial for precision-focused lesions. The study also generates attention mechanism GradCAM based on the Swin-ViT model, improving interpretability. Ablation experiments demonstrate that incorporating CNN and Transformer models outperform single-structure models. Future research can focus on extracting complex diseases and high-level semantic data to boost the model's ability.

关键词： Health issue Chronic renal failure Tumor detection Swin-ViT ConvNext Transfer learning Multilabel grading

来源：评论

学校读者我要写书评

暂无评论

Microscope-Assisted Hypertensive Retinopathy Diagnosis Using Deep Learning Models

引用

MICROSCOPY RESEARCH AND TECHNIQUE 2025年第0期 2025 Mar 24页

作者： Akbar, Shahzad Shahzore, Usama Saba, Tanzila Alamri, Faten S. Khan, Sadaf S. Khan, Amjad R. Riphah Int Univ Riphah Coll Comp Riphah Artificial Intelligence Res RAIR Lab Faisalabad Pakistan Prince Sultan Univ Artificial Intelligence & Data Analyt Lab CCIS Riyadh Saudi Arabia Princess Nourah Bint Abdulrahman Univ Coll Sci Dept Math Sci Riyadh Saudi Arabia

The retina is the most crucial part of the human eye, and it can be affected due to hypertension. However, retinal abnormalities due to hypertension are termed hypertensive retinopathy (HR). A severe stage of HR can lead to complete blindness if not diagnosed and treated on time. Manually analyzing retinal images for HR diagnosis is time-consuming and prone to errors. This research article provides a novel technique based on U-Net and Dense-Net for automatic HR detection and grading through retinal images. The presented method consists of preprocessing, vessel segmentation, artery or vein (A/V) classification, and vessel width calculation to compute the arteriovenous ratio (AVR). In the preprocessing phase, the Gabor filter is applied to the retinal image to enhance the vascular network of the image. The preprocessed image is fed into the U-Net architecture to segment the vascular network image. The segmented vascular network image is fed into the Dense-Net architecture for A/V classification. The A/V classified vascular network is divided into several artery and vein segments at the bifurcation and crossover points. The A/V segments are labeled for width calculation to compute the AVR. The AVR is a standard parameter for HR detection and grading. The evaluation results show an average accuracy of 99.40% in HR classification and 99.77% in HR grading on the AVRDB dataset. The evaluated results are beneficial for the automatic HR detection and grading for clinical purposes.

关键词： arteriovenous ratio artery/vein classification health risks hypertensive retinopathy microscope-assisted imaging visions loss

来源：评论

学校读者我要写书评

暂无评论

Deep Ensemble for Central Serous Microscopic Retinopathy Detection in Retinal Optical Coherence Tomographic Images

引用

MICROSCOPY RESEARCH AND TECHNIQUE 2025年第0期 2025 Feb 27页

作者： Hassan, Syed Ale Akbar, Shahzad Shoukat, Ijaz Ali Khan, Amjad R. Alamri, Faten S. Saba, Tanzila Riphah Int Univ Riphah Artificial Intelligence Res RAIR Lab Faisalabad Campus Faisalabad Pakistan CCIS Prince Sultan Univ Riyadh Artificial Intelligence & Data Analyt Lab Riyadh Saudi Arabia Princess Nourah Bint Abdulrahman Univ Coll Sci Dept Math Sci Riyadh Saudi Arabia

The retina is an important part of the eye that aids in focusing light and visual recognition to the brain. Hence, its damage causes vision loss in the human eye. Central serous retinopathy is a common retinal disorder in which serous detachment occurs at the posterior pole of the retina. Therefore, detection of CSR at an early stage with good accuracy can decrease the rate of vision loss and recover the vision to normal conditions. In the past, numerous manual techniques have been devised for CSR detection;nevertheless, they have demonstrated imprecision and unreliability. Thus, the deep learning method can play an important role in automatically detecting CSR. This research presents a convolutional neural network-based framework combined with segmentation and post-ocessing for CSR classification. There are several challenges in the segmentation of retinal images, such as noise, size variation, location, and shape of the fluid in the retina. To address these limitations, Otsu's thresholding has been employed as a technique for segmenting optical coherence tomography (OCT) images. Pigments and fluids are present in epithelial detachment, and contrast adjustment and noise removal are required. After segmentation, post-processing is used, combining flood filling, dilation, and area thresholding. The segmented processed OCT scans were classified using the fusion of three networks: (i) ResNet-18, (ii) Google-Net, and (iii) VGG-19. After experimentation, the fusion of ResNet-18, GoogleNet, and VGG-19 achieved 99.6% accuracy, 99.46% sensitivity, 100% specificity, and 99.73% F1 score using the proposed framework for classifying normal and CSR-affected images. A publicly available dataset OCTID comprises 207 normal and 102 CSR-affected images was utilized for testing and training of the proposed method. The experimental findings conclusively demonstrate the inherent suitability and efficacy of the framework put forth. Through rigorous testing and analysis, the results unequ

关键词： central serous microscopic retinopathy (CSR) deep learning health risks optical coherence tomography images retinal surface

来源：评论

学校读者我要写书评

暂无评论

AI-driven IoT-fog analytics interactive smart system with data protection

引用

EXPERT SYSTEMS 2025年第1期42卷

作者： Haseeb, Khalid Saba, Tanzila Rehman, Amjad Abbas, Naveed Kim, Pyoung Won CCIS Prince Sultan Univ Artificial Intelligence & Data Analyt AIDA Lab Riyadh Saudi Arabia Islamia Coll Peshawar Dept Comp Sci Peshawar Pakistan Incheon Natl Univ Coll Educ Incheon South Korea

In recent decades, fog computing has contributed significantly to the expansion of smart cities. It generated numerous real-time data and coped with time-constraint applications. They use sensors, physical objects, and network standards to monitor health imaging, traffic surveillance, industrial management, and so forth. Interactive applications have been proposed for the Internet of Things (IoT) to control wireless channels and improve communication. However, most of the existing lack of handing network interference and a reliable monitoring process. Moreover, many solutions are vulnerable to external threats, resulting in inconsistent and untrustworthy information for end users. Thus, this article proposes a framework that considers possible shortest paths to provide the most reliable and low-latency healthcare decision system using Q-learning. In addition, fog devices offer a trusted transmission interference system and are kept secure. The proposed framework is specially designed for rapid real-time medical data processing while enforcing robust security throughout the IoT-based transmission process. To identify the health sensors in pairwise objects with the initial computing cost, the proposed framework applies graph theory. It also extracts the most effective and least loaded communication edges by examining the behaviour of devices. Moreover, the identities of devices are verified using lightweight timestamps and secret information, accordingly, it decreases the privacy threats.

关键词： healthcare image mobile intelligence reliable processing smart systems

来源：评论

学校读者我要写书评

暂无评论

Towards Ergonomic Typing: Real-Time Typing Posture Classification on WatchOS

Towards Ergonomic Typing: Real-Time Typing Posture Classific...

引用

Women in data Science at Prince Sultan University (WiDS PSU), International Conference of

作者： Fetoun AlZahrani Arwa Ghouti Tanzila Saba Artificial Intelligence and Data Analytics (AIDA) Lab CCIS Prince Sultan University Riyadh Kingdom of Saudi Arabia

ISBN: (数字)9798331520922

ISBN: (纸本)9798331520939

With the increasing prevalence of computer usage, there has been a rise in cases of repetitive strain injuries (RSIs) like Carpal Tunnel Syndrome (CTS) among individuals who spend extended hours typing. Hand pose estimation plays a vital role in improving ergonomic and safe human-computer interaction (HCI). To address this challenge, we propose ErgoType, a real-time hand pose estimation application developed for Apple WatchOS. The application delivers posture classification results directly on the Apple Watch and integrates with a companion iPhone application for detailed feedback. The system employs a Support Vector Machine (SVM) model trained on smartwatch emulator data to analyze key hand-position indicators, such as wrist and elbow angles, achieving a classification accuracy of 98%. Immediate feedback is provided via on-screen notifications, alerting users to poor typing habits such as excessive wrist flexion or prolonged typing without rest. This feedback encourages users to adopt healthier typing practices. ErgoType's combination of wearable technology and real-time feedback offers a proactive solution for reducing typing-related injuries and promoting workplace wellness. It is designed to support both professionals and frequent computer users, enhancing productivity and well-being through improved ergonomics.

关键词： Support vector machines Wrist Hands Carpal tunnel syndrome Wearable Health Monitoring Systems Ergonomics Pose estimation Real-time systems Injuries Strain

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Review and Analysis on Facial Emotion Recognition:Performance Insights into Deep and Traditional Learning with Current Updates and Challenges

引用

Computers, Materials & Continua 2025年第1期82卷 41-72页

作者： Amjad Rehman Muhammad Mujahid Alex Elyassih Bayan AlGhofaily Saeed Ali Omer Bahaj Artificial Intelligence&Data Analytics Lab College of Computer&Information SciencesPrince Sultan UniversityRiyadh11586Saudi Arabia MIS Department College of Business Administration Prince Sattam Bin Abdulaziz UniversityAlKharj11942Saudi Arabia

In computer vision and artificial intelligence,automatic facial expression-based emotion identification of humans has become a popular research and industry *** demonstrations and applications in several fields,including computer games,smart homes,expression analysis,gesture recognition,surveillance films,depression therapy,patientmonitoring,anxiety,and others,have brought attention to its significant academic and commercial *** study emphasizes research that has only employed facial images for face expression recognition(FER),because facial expressions are a basic way that people communicate meaning to each *** immense achievement of deep learning has resulted in a growing use of its much architecture to enhance *** review is on machine learning,deep learning,and hybrid methods’use of preprocessing,augmentation techniques,and feature extraction for temporal properties of successive frames of *** following section gives a brief summary of assessment criteria that are accessible to the public and then compares them with benchmark results the most trustworthy way to assess FER-related research topics *** this review,a brief synopsis of the subject matter may be beneficial for novices in the field of FER as well as seasoned scholars seeking fruitful avenues for further *** information conveys fundamental knowledge and provides a comprehensive understanding of the most recent state-of-the-art research.

关键词： Face emotion recognition deep learning hybrid learning CK+ facial images machine learning technological development

来源：评论

学校读者我要写书评

暂无评论

Early-Stage Melanoma Cancer Diagnosis Framework for Imbalanced data From Dermoscopic Images

引用

MICROSCOPY RESEARCH AND TECHNIQUE 2025年第3期88卷 797-809页

作者： Khan, Amjad Rehman Mujahid, Muhammad Alamri, Faten S. Saba, Tanzila Ayesha, Noor Prince Sultan Univ Artificial Intelligence & Data Analyt Lab CCIS Riyadh Saudi Arabia Princess Nourah Bint Abdulrahman Univ Coll Sci Dept Math Sci Riyadh Saudi Arabia Prince Sultan Univ Ctr Excellence Cyber Secur CYBEX Riyadh Saudi Arabia

Skin problems are a serious condition that affects people all over the world. Prolonged exposure to ultraviolet rays' damages melanocyte cells, leading to the uncontrolled proliferation of melanoma, a form of skin cancer. However, the dearth of qualified expertise increases the processing time and cost of diagnosis. Early detection of melanoma in dermoscopy images significantly enhances its chance of survival. Pathologists benefit substantially from the precise and efficient melanoma cancer diagnosis using automated methods. Nevertheless, the diagnosis of melanoma has consistently been a challenging procedure due to the imbalance images and limited data. Our objective was to employ a novel deep method to diagnose melanoma from dermoscopic images automatically. The research has proposed a novel framework for detecting skin malignancies. The proposed plan, which includes CNN, DenseNet, a batch normalization layer, maxpooling, and a ReLU layer activation function, solves the overfitting problem well. Furthermore, we used a large number of samples for testing and effectively employed data augmentation to prevent any issues related to class imbalance. The Adam optimizer is the most efficient deep learning optimizer for addressing challenges associated with large datasets, such as lengthy processing times. This is due to its specifically designed algorithm. Experiments ensure that the proposed framework achieved 95.70% micro average accuracy on the ISIC-2019 dataset and 93.24% accuracy on the HAM-10000 dataset. Comprehensive evaluation and analysis were used to evaluate our framework's performance. The results show that the proposed approach performs better with cross-validation by 94.8% accuracy than the most sophisticated deep learning-based technique. During studies, medical professionals will employ the proposed model to identify skin cancer in its early stages.

关键词： augmentation deep framework dermoscopy healthcare imbalance data melanoma cancer

来源：评论

学校读者我要写书评

暂无评论

UEF-HOCUrdu: Unified Embeddings Ensemble Framework for Hate and Offensive Text Classification in Urdu

引用

IEEE ACCESS 2025年 13卷 21853-21869页

作者： Ullah, Kifayat Aslam, Muhammad Khan, Muhammad Usman Ghani Alamri, Faten S. Khan, Amjad Rehman Univ Engn & Technol Lahore Dept Comp Sci Lahore 54890 Pakistan UET Al Khawarizmi Inst Comp Sci Natl Ctr Artificial Intelligence Lahore 54890 Pakistan Princess Nourah Bint Abdulrahman Univ Coll Sci Dept Math Sci Riyadh 11671 Saudi Arabia Prince Sultan Univ Artificial Intelligence & Data Analyt Lab AIDA CCIS Riyadh 11586 Saudi Arabia

Hate speech and other forms of hostile communication on social media have several implications such as;fostering violence, promoting social divide, and negative psychological effects. Since such toxic language is becoming more and more common, it is imperative to have a proper way of identifying it, especially in low resource language like Urdu. To meet this challenge, this research proposed a new ensemble based multi-classification model and generated new dataset of 36,000 Urdu tweets categorized as 'Hate', 'Offensive' and 'Neither'. This study sought to create a model that not only achieves a high classification accuracy but also overcome key challenges inherent in natural language processing, namely, high dimensionality, sparsity, overfitting, OOV words and dialectal variations. For this purpose, an extensive comparison of different learning algorithms were conducted. As a result, the most efficient models, namely FastText, XLM-RoBERTa, ULMFiT, and XGBoost were incorporated in the proposed ensemble approach to achieve the best results in both classification and mitigation of NLP issues. To further enhance the confidence in proposed model, a stratified 5-fold cross-validation technique has been utilized. The ensemble model performed the best and achieved macro F1 score of 0.94, complemented by comprehensive labeled dataset focusing on hate and offensive speech in Urdu. By addressing key research gaps, this research provides a valuable foundation for future work and benchmarking in Urdu hate speech multi-classification tasks.

关键词： Hate speech Text categorization Machine learning Deep learning Context modeling Transfer learning Multilingual data models Cyberbullying Cultural differences Urdu hate speech detection Urdu multi-class classification machine learning deep learning transfer learning ensemble learning model natural language processing (NLP)

来源：评论

学校读者我要写书评

暂无评论

Efficient Hybrid-Robust Approach for Cancer Biomarker Discovery Using Omics data

引用

IEEE ACCESS 2025年 13卷 51130-51149页

作者： Sid, Karima Zertal, Soumia Batouche, Mohamed Zerabi, Soumeya Oum El Bouaghi Univ Dept Math & Comp Sci Oum El Bouaghi 04000 Algeria Abdelhamid Mehri Constantine 2 Univ Lab Data Sci Comp & Artificial Intelligence LISIA Constantine 25016 Algeria Oum El Bouaghi Univ Dept Math & Comp Sci Lab Artificial Intelligence & Autonomous Things LI Oum El Bouaghi 04000 Algeria Princess Nourah bint Abdulrahman Univ Dept Informat Technol CCIS RC Riyadh 11671 Saudi Arabia Abdelhamid Mehri Constantine 2 Univ Dept Comp Sci & Its Applicat Lab Data Sci Comp & Artificial Intelligence LISIA Constantine 25016 Algeria

DNA Microarray datasets, also known as "omics" data, are important for the diagnosis of numerous diseases, including cancer and tumors. In the analysis of these data, feature selection techniques and classification algorithms are the workhorse for choosing candidate genes that serve as cancer biomarkers. However, microarray datasets present a challenge;they contain a greater number of features than the samples, which affects the performance of algorithms used in the analysis process. In order to extract precise information, it is necessary to employ a method that is both robust and performant. This paper emphasizes the importance of accurate and stable gene selection for the discovery of knowledge derived from high-dimensional data. A novel hybrid framework was put forth for consideration, comprising three distinct stages: Clustering, Parallel Filtering, and Hybrid-Parallel Optimization. In each step, a combination of techniques and algorithms is used to improve the results in terms of stability and/or accuracy. The proposal is evaluated and tested according to different scenarios;using thirteen gene expression datasets and two classifiers: artificial Neural Network (ANN) and Na & iuml;ve Bayes (NB). Comparison with related work demonstrates the efficacy of this approach, which enhances classification accuracy and stability while reducing the number of selected genes.

关键词： Feature extraction Stability criteria Classification algorithms Cancer Clustering algorithms Gene expression Accuracy Approximation algorithms Metaheuristics Training Biomarker discovery microarray data feature selection classification optimization algorithms Markov blanket approximation

来源：评论

学校读者我要写书评

暂无评论

TMRE:Novel Algorithm for Computing Daily Reference Evapotranspiration Using Transformer-Based Models

引用

Computers, Materials & Continua 2025年第5期83卷 2851-2864页

作者： Bushra Tayyaba Muhammad Usman Ghani Khan Talha Waheed Shaha Al-Otaibi Tanzila Saba National Center of Artificial Intelligence(NCAI) Al Khwarizmi Institute of Computer Sciences(KICS)Lahore54890Pakistan Department of Computer Science University of Engineering and Technology LahoreLahore54890Pakistan Artificial Intelligence&Data Analytics Lab Prince Sultan UniversityRiyadh11586Saudi Arabia Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia

Reference Evapotranspiration(ETo)iswidely used to assess totalwater loss between land and atmosphere due to its importance in maintaining the atmospheric water balance,especially in agricultural and environmental *** estimation of ETo is challenging due to its dependency onmultiple climatic variables,including temperature,humidity,and solar radiation,making it a complexmultivariate time-series *** machine learning and deep learning models have been applied to forecast ETo,achieving moderate ***,the introduction of transformer-based architectures in time-series forecasting has opened new possibilities formore precise ETo *** this study,a novel algorithm for ETo forecasting is proposed,focusing on four transformer-based models:Vanilla Transformer,Informer,Autoformer,and FEDformer(Frequency Enhanced Decomposed Transformer),applied to an ETo dataset from the Andalusian *** novelty of the proposed algorithm lies in determining optimized window sizes based on seasonal trends and variations,which were then used with each model to enhance prediction *** custom window-sizing method allows the models to capture ETo’s unique seasonal patterns more ***,results demonstrate that the Informer model outperformed other transformer-based models,achievingmean square error(MSE)values of 0.1404 and 0.1445 for forecast windows(15,7)and(30,15),*** Vanilla Transformer also showed strong performance,closely following the *** findings suggest that the proposed optimized window-sizing approach,combined with transformer-based architectures,is highly effective for ETo *** novel strategy has the potential to be adapted in othermultivariate time-series forecasting tasks that require seasonality-sensitive approaches.

关键词： Reference evapotranspiration ETo transformer informer autoformer FEDformer timeseries forecasting self-attention

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：