IEEE Access最新文献

{"title":"Integration of Deep Learning Architectures With GRU for Automated Leukemia Detection in Peripheral Blood Smear Images","authors":"Amit Kumar Bairwa;Anita Shrotriya;Priyanka Mathur;Sandeep Joshi;Sakshi Shringi;Ambika Kumari","doi":"10.1109/ACCESS.2025.3565546","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565546","url":null,"abstract":"This paper investigates the integration of deep learning (DL) architectures with Recurrent Neural Networks (RNNs) for automated leukemia detection in Peripheral Blood Smear (PBS) images. Models such as DenseNet201, EfficientNetB3, Inception v3, InceptionResNetV2, MobileNetV2, ResNet50, VGG16, and Xception, each enhanced with a Gated Recurrent Unit (GRU) layer, are employed to capture spatial and temporal dependencies in image data. Through extensive experimentation on a standard leukemia dataset, the performance of each model is evaluated based on accuracy and computational efficiency. Among these, the Xception + GRU model achieves the highest classification accuracy, attaining an impressive 99.69%. This exceptional result underscores the efficacy of combining Convolutional Neural Networks (CNNs) with RNNs, particularly GRUs, in accurately detecting Leukemia from PBS images. The findings offer valuable contributions to medical image analysis, demonstrating the potential of DL techniques to enhance automated disease diagnosis. By advancing the precision of leukemia detection, this study provides promising implications for improving patient care and treatment outcomes.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"84217-84239"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979923","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144117426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Blockchain-Assisted Cross-Domain Authentication and Key Negotiation Scheme in IIoT","authors":"Zhengnan Xu;Guofang Dong;Ruicheng Yang","doi":"10.1109/ACCESS.2025.3565393","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565393","url":null,"abstract":"Traditional cross-domain schemes for industrial Internet of Things (IIoT) are deficient in decentralization and lack a precise mechanism for updating session keys, resulting in high overhead from repeated authentication and key negotiation. This paper proposes a blockchain-assisted cross-domain authentication and key negotiation scheme, aiming at realizing authentication and key negotiation between different domains of IIoT. First, a session token mechanism is introduced to limit the validity period of the negotiated session key and assist in updating the session key. Second, blockchain is utilized to assist in updating the validity period of the session key, thus significantly reducing the overhead of repeated authentication. Finally, we analyze the security and performance of the scheme, and the results show that the scheme can efficiently verify data integrity and add the functions of key update, user dynamic update, and semi-trusted third-party dynamic update. The experimental results show that compared with other schemes, our scheme reduces 69.4%, 78.9%, 88.8%, and 38.3% of the time consumed in terms of computation overhead and reduces 63.5%, 18.8%, 15.1%, and 52.9% of the cost in terms of communication overhead, respectively. The proposed scheme has a low overhead.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"88156-88173"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979842","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144139880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LLM-MalDetect: A Large Language Model-Based Method for Android Malware Detection","authors":"Ruirui Feng;Hui Chen;Shuo Wang;Md Monjurul Karim;Qingshan Jiang","doi":"10.1109/ACCESS.2025.3565526","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565526","url":null,"abstract":"Android malware poses a significant cybersecurity threat, enabling unauthorized data access, financial fraud, and device compromise. Although deep learning methods are widely used for malware detection, they often struggle with stability and adaptability in the face of evolving threats. While large language models (LLMs) have shown promise in this area, their application to Android malware detection remains underexplored, particularly with regard to optimizing the semantic relationships within Android application packages (APKs). To address this gap, we introduce LLM-MalDetect, a novel framework that improves LLM-based APK analysis by explicitly modeling semantic dependencies and leveraging structured prompt engineering for optimized detection. Our approach formalizes LLM adaptation through a robust string-based feature extraction method and a tailored fine-tuning strategy to enhance precision. Evaluations on benchmark datasets demonstrate that LLM-MalDetect achieves up to 98.97% accuracy, outperforms existing methods in terms of robustness, and enables real-time analysis.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"81347-81364"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979936","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143943827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"HuberAIME: A Robust Approach to Explainable AI in the Presence of Outliers","authors":"Takafumi Nakanishi","doi":"10.1109/ACCESS.2025.3565279","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565279","url":null,"abstract":"With the increasing accuracy of machine-learning models in recent years, explainable artificial intelligence (XAI), which allows for an understanding of the internal decisions made by these models, has become essential. However, many explanation methods are vulnerable to outliers and noise, and the results may be distorted by extreme values. This study devised a new method named HuberAIME, which is a variant of approximate inverse model explanations (AIME) and is robust to the Huber loss. HuberAIME limits the impact of outliers by weighting with iterative reweighted least squares and prevents the feature importance estimation of AIME from being degraded by extreme data points. Comparative experiments were conducted using the Wine dataset, which has almost no outliers, the Adult dataset, which contains extreme values, and the Statlog (German Credit) dataset, which has moderate outliers, to demonstrate the effectiveness of the proposed method. SHapley Additive exPlanations, AIME, and HuberAIME were evaluated using six metrics (explanatory accuracy, sparsity, stability, computational efficiency, robustness, and completeness). HuberAIME was equivalent to AIME on the Wine dataset. However, it outperformed AIME on the Adult dataset, exhibiting high fidelity and stability. On the Germain Credit dataset, AIME itself showed a certain degree of robustness, and there was no significant difference between AIME and HuberAIME. Overall, HuberAIME is useful for data that include serious outliers and maintains the same explanatory performance as AIME in cases of few outliers. Thus, HuberAIME is expected to improve the reliability of actual operations as a robust XAI method.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"76796-76810"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979913","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913302","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust Sliding Mode Control Based on Fractional Order Reaching Law for Rehabilitation Robots","authors":"Rim Jallouli-Khlif;Fatma Abdelhedi;Chadia Zayane;Ahmed Said Nouri;Nabil Derbel","doi":"10.1109/ACCESS.2025.3565291","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565291","url":null,"abstract":"Rehabilitation robots, particularly lower-limb exoskeletons, are transforming healthcare by assisting individuals with mobility impairments. This study introduces a novel Sliding Mode Control (SMC) system based on a fractional-order reaching law, designed to enhance control performance and robustness. The proposed approach effectively manages the exoskeleton’s dynamic behavior, particularly during the transient regime, by reducing initial torque energy demand during start-up, ensuring precise trajectory tracking, and prioritizing patient safety and comfort. The method’s effectiveness is validated through MATLAB simulations and supported by a rigorous dual stability analysis, demonstrating asymptotic and finite-time convergence of the system in the reaching and sliding phases. A Comparison study with traditional SMC techniques proves that the FO-RL-SMC significantly improves energy efficiency during the transient phase and the overall dynamical behavior of the system. These results highlight the potential of the proposed FO-RL-SMC system to advance the performance of rehabilitation robots, emphasizing its value in addressing complex control challenges and improving patient outcomes.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"75744-75753"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979986","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143918666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Comprehensive Comparative Analysis of Deep Learning Models for Student Performance Prediction in Virtual Learning Environments: Leveraging the OULA Dataset and Advanced Resampling Techniques","authors":"Bayan A. Alnasyan;Mohammed Basheri;Madini O. Alassafi","doi":"10.1109/ACCESS.2025.3564719","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3564719","url":null,"abstract":"Predicting student performance in Virtual Learning Environments (VLEs) has become increasingly important with the growth of online education. Early identification of at-risk students allows timely interventions to improve academic outcomes. This study evaluates the performance of several Deep Learning (DL) models for tabular data, including ResNet, NODE, AutoInt, TabNet, TabTransformer (TT), SAINT, and GatedTabTransformer (GTT). Moreover, it examines the role of resampling techniques, including SMOTE, ROS, ADASYN, RUS, and Tomek Links, in addressing class imbalance. Using the OULA dataset, eight experiments were conducted for binary and multi-class classification tasks, testing different feature combinations: 1) behavioral, 2) demographic and behavioral, 3) academic and behavioral, and 4) demographic, academic, and behavioral. The results indicate that incorporating a comprehensive set of characteristics can significantly enhance the model’s performance, with academic characteristics proving more predictive than demographic characteristics. The SAINT model achieved the highest performance in binary classification (94.33% accuracy), leveraging its ability to capture meaningful yet straightforward feature interactions. For multi-class classification, SAINT again outperformed other models, achieving an accuracy of 73.22% when using the Tomek Links method, excelling in managing complex feature interactions and underrepresented classes such as “Distinction.” Statistical analysis was done using the Friedman aligned ranks test and the Nemenyi post-test to compare how well the models performed based on F1-scores from several experiments. The non-parametric Friedman test revealed significant differences among the models (<inline-formula> <tex-math>$p = 0.00013$ </tex-math></inline-formula>). SAINT and AutoInt consistently outperformed the other approaches, while ResNet and TT demonstrated the weakest performance. Post-hoc analysis using the Nemenyi test did not show statistically significant differences among mid-tier models (TabNet, GTT, NODE). A critical difference (CD) further confirmed that SAINT and AutoInt are the most effective architectures for addressing complex, imbalanced educational data. These findings highlight the importance of aligning model selection and resampling techniques with the complexity of the task and the characteristics of the data.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"75953-75972"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979810","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143918686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on Digital Media Art for Image Caption Generation Based on Integrated Transformer Models in CLIP","authors":"Lu Gao;Xiaofei Pang","doi":"10.1109/ACCESS.2025.3565682","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565682","url":null,"abstract":"Digital media art has a wide application in the field of image caption generation. In digital media art exhibitions or online works displays, some complex image works may have multiple layers of meanings or abstract expressions, which can help viewers better understand the works. It can also serve as another auxiliary element besides sound, collaborating with visual elements to provide a richer experience for the audience. The purpose of picture captioning is to provide textual descriptions that correlate to input images. The CLIP paradigm is highly versatile to resolve vision-text difficulties. In the field of picture description, the standard Transformer architecture has also exhibited good effects, which uses an image encoder and a text decoder. Large parameter numbers and the demand for further data preprocessing are still significant difficulties. In order to replace the fundamental features of conventional multi-modal fusion models, we propose a New Multi-modal Fusion Attention module (NMFA), which efficiently decreases parameter sizes and computational complexity in half. Expanding upon this, we propose the Transformer Fusion CLIP (TFC) model, which minimizes parameter sizes and processing demands while getting remarkable assessment scores. Additionally, we strengthen the mechanism for cumulative points and reward sequence length to encourage the construction of larger sequences. Finally, we combine the enhanced beam search technique to further train the TFC model. Results from our testing on the MSCOCO dataset reveal that we have not only greatly improved the efficiency of the TFC model but also speeded up its runtime by eight times and reduced model parameters by over 50%.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"75894-75910"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979925","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143918689","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimality Versus Generality: Performance Assessment of Meta-Heuristics in Educational Timetabling","authors":"Sina Abdipoor;Razali Yaakob;Say Leng Goh;Salwani Abdullah;Hazlina Hamdan;Khairul Azhar Kasmiran","doi":"10.1109/ACCESS.2025.3565522","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565522","url":null,"abstract":"Educational timetabling, a principal branch of operations research, presents challenging combinatorial optimization problems widely encountered in educational institutions. Meta-heuristics have commonly been applied to these problems and managed to attain promising performance in terms of optimality. However, their general applicability has been overlooked, hindering their effectiveness as versatile solvers. The limited generalizability of current approaches is the primary hurdle between the literature and real-world applications. This paper addresses this gap by introducing a generality taxonomy and conducting comprehensive theoretical and empirical analyses. This study highlights the adverse impact of extreme parameter tuning on generality, emphasizing the need for more generalized approaches. Furthermore, it introduces a performance assessment framework, penalizing problem-tailored solutions. It also examines the optimality vs. generality performance of the state-of-the-art approaches of the latest university course timetabling benchmark to further reinforce our claim and validate the efficacy of our framework. Our findings indicate that the current literature prioritizes optimality over generality. We believe adopting the proposed assessment framework is crucial for bridging the gap between research and practical applications, enabling fairer comparisons, and encouraging more adaptable approaches.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"75679-75696"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979955","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Corrections to “Analysis and Modeling of Direct Ammonia Fuel Cells for Solar and Wind Power Leveling in Smart Grid Applications”","authors":"Miswar A. Syed;Osamah Siddiqui;Mehrdad Kazerani;Muhammad Khalid","doi":"10.1109/ACCESS.2025.3561984","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3561984","url":null,"abstract":"Presents corrections to the paper, (Corrections to “Analysis and Modeling of Direct Ammonia Fuel Cells for Solar and Wind Power Leveling in Smart Grid Applications”).","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"71081-71081"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10980055","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143888396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Head-Driven Algorithm for Estimating Upper and Lower Body Motion in Virtual Reality Environments","authors":"Jemin Lee;Jeonghyeon Kim;Youngwon Kim","doi":"10.1109/ACCESS.2025.3565232","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3565232","url":null,"abstract":"This study proposes a novel algorithm for effectively estimating partial upper and lower body movements in a Virtual Reality (VR) environment using only head movements and synchronized head rotation axes, without the need for additional hardware. The proposed algorithm calculates the angle between the avatar’s pelvis and the head rotation axis to naturally reproduce the user’s upper body inclination and lower body bending. Notably, it offers the advantage of efficiently utilizing limited computational resources in multiplayer environments. The experiment was conducted in two stages. In the first stage, the objective performance of the algorithm was evaluated by comparing it with ground truth inclination data. In the second stage, participants performed two types of games (e.g., a dodgeball game and a limbo game) to assess their sense of immersion and embodiment. The objective results demonstrated that the proposed algorithm accurately and naturally expressed upper and lower body movements. Additionally, post-experiment surveys indicated that participants reported a high level of immersion and a natural interaction experience. This study presents a cost-effective solution for tracking upper and lower body movements in VR environments without requiring additional hardware, significantly enhancing the immersion of the VR experience. Future research will explore the expansion of the method to include upper body rotation estimation and full-body motion tracking, incorporating user locomotion.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"76627-76637"},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979843","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}