IEEE Access最新文献_第10页

Multiuser FDSS-Based DCT-Spread Massive MIMO OFDM System for Secure RIS-Assisted UAV-Enabled Networks

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544477

Md. Najmul Hossain;Atia Kaniz;Sk. Tamanna Kamal;Shaikh Enayet Ullah;Tetsuya Shimamura

{"title":"Multiuser FDSS-Based DCT-Spread Massive MIMO OFDM System for Secure RIS-Assisted UAV-Enabled Networks","authors":"Md. Najmul Hossain;Atia Kaniz;Sk. Tamanna Kamal;Shaikh Enayet Ullah;Tetsuya Shimamura","doi":"10.1109/ACCESS.2025.3544477","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544477","url":null,"abstract":"Enhancing spectral efficiency (SE), reducing out-of-band (OOB) power emission, minimizing multiuser interference (MUI), and ensuring secure data communication are key challenges in the development of future-generation unmanned aerial vehicle (UAV)-enabled wireless communication systems. To address these challenges and emphasize effective solutions, we propose a novel and secure reconfigurable intelligent surface (RIS)-assisted UAV-enabled multiuser frequency-domain spectrum shaping (FDSS)-based discrete cosine transform (DCT)-Spread massive multiple-input multiple-output (mMIMO) orthogonal frequency division multiplexing (OFDM) system. The system incorporates physical layer security (PLS) based on two-dimensional cosine-sine interleaved chaotic map (2D-CSIM), block diagonalization (BD) precoding, FDSS with null carriers, turbo and repeat-and-accumulate (RA) channel coding, Cholesky decomposition-based zero-forcing (CD-ZF), and lattice reduction-based minimum mean square error (LR-MMSE) signal detection schemes. The 2D-CSIM enhances confidentiality in audio signal transmission, while the BD precoding reduces MUI. In addition, the FDSS with null carriers reduces OOB emissions and enhances the system’s SE. The turbo and RA channel coding with CD-ZF and LR-MMSE signal detection schemes improve the bit error rate (BER) performance. Simulation results show the effectiveness of the proposed system. It achieves significant PLS for audio transmission, with low correlation coefficients of <2%> <tex-math>$1 times 10^{-3}$ </tex-math></inline-formula>, which are comparable to state-of-the-art systems. At signal-to-noise ratios (SNRs) of 27–28.5 dB, the four users achieved a BER of <inline-formula> <tex-math>$1times 10^{-4}$ </tex-math></inline-formula> with RA coding, LR-MMSE detection, and lower-order quadrature amplitude modulation (QAM) constellation.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"35254-35269"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10897968","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143526688","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}

引用次数: 0

Analysis of the Influence of CT Imaging Conditions on the CT Image Features of Pulmonary Nodule Phantoms

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544439

Jing Li;Daidi Zhong;Shuo Wang;Chunjing Tao;Hao Wang

{"title":"Analysis of the Influence of CT Imaging Conditions on the CT Image Features of Pulmonary Nodule Phantoms","authors":"Jing Li;Daidi Zhong;Shuo Wang;Chunjing Tao;Hao Wang","doi":"10.1109/ACCESS.2025.3544439","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544439","url":null,"abstract":"By studying the impact of CT imaging conditions on image features of pulmonary nodule phantoms, this study explores possibility to optimize the diversity of imaging conditions when building a test set for AI-enabled medical device software for pulmonary nodule analysis. In this study, 15 phantoms mimicking three types of pulmonary nodules were placed in different positions in a whole lung phantom. CT images were collected under combination of different parameters, including tube voltage, tube current-time product, collimator width and reconstruction operator (14 sets in total). The texture features (GLCM, GLRLM, GLSZM, GLDM, NGTDM) of CT images of pulmonary nodules were extracted using the PyRadiomics package in Python, and the five main texture features were selected using the LASSO method. To analyze the data, correlation analysis and non-parametric tests were applied, with a correction method used to control errors in multiple comparisons. The results showed that changes in the reconstruction kernel affected the stability of texture features in CT images of pulmonary nodules more significantly than other imaging parameters in this study. Different reconstruction kernels influenced the representation of image details, leading to variations in image sharpness and noise levels, which in turn affected the stability of the extracted feature values. Meanwhile, no significant changes were induced by variation of tube voltage, tube current-time product and collimator width during image acquisition. These findings suggest that when building the test set for AI-enabled medical device software for pulmonary nodule analysis, it is essential to carefully consider the diversity of reconstruction kernels and their impact on data quality, thereby enhancing the representativeness of the test set.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"35101-35112"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10897997","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143527608","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}

引用次数: 0

Positive Anchor Area Merge Algorithm: A Knowledge Distillation Algorithm for Fruit Detection Tasks Based on Yolov8

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544361

Xiangqun Shi;Xian Zhang;Yifan Su;Xun Zhang

{"title":"Positive Anchor Area Merge Algorithm: A Knowledge Distillation Algorithm for Fruit Detection Tasks Based on Yolov8","authors":"Xiangqun Shi;Xian Zhang;Yifan Su;Xun Zhang","doi":"10.1109/ACCESS.2025.3544361","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544361","url":null,"abstract":"In the agricultural sector, employing machine vision technology for fruit target detection holds significant research importance and broad application prospects, such as enabling fruit growth monitoring, yield prediction, and fruit sorting. The Yolov8 model, as the latest model in the field of object detection, boasts advantages including high execution efficiency and detection accuracy. However, when it comes to fruit object detection, which means counting and locating target fruits in an image, the performance of the Yolov8 model shows a noticeable decline compared to its performance on the standard COCO dataset. To address this issue, knowledge distillation is a highly versatile method that uses a large teacher model to guide the training of a smaller student model, thereby improving the detection accuracy of the student model. This thesis proposes a Yolov8 knowledge distillation method tailored for fruit recognition tasks, which improves the network through knowledge distillation and implements a knowledge distillation method based on positive anchor area merging to enhance detection accuracy for fruit recognition tasks. On our self-constructed fruit dataset, which contains over 3,000 images for each category, we compared our model with other similar state-of-the-art models in terms of resource consumption and detection accuracy. While maintaining a low resource overhead, our model achieved an mAP(50) of 99.47%, which is higher than other models that range from 99.1% to 99.3%. In the ablation experiments, we also demonstrated the practical significance of dividing the positive sample area. Finally, we deployed the model on an embedded system for real-time detection of on-site images. These experiments illustrate the practicality of our method for recognizing fruits in real-world scenarios.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"34954-34968"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10897963","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521403","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}

引用次数: 0

A Study on the SLAM of Automotive Vehicles Using Bumper-Mounted Dual LiDAR

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544350

Jae-Hun Jang;Hyeong-Jun Kim;Kyung-Chang Lee

{"title":"A Study on the SLAM of Automotive Vehicles Using Bumper-Mounted Dual LiDAR","authors":"Jae-Hun Jang;Hyeong-Jun Kim;Kyung-Chang Lee","doi":"10.1109/ACCESS.2025.3544350","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544350","url":null,"abstract":"This study proposes a localization and three-dimensional (3D) mapping algorithm for autonomous vehicles using bumper-mounted light detection and ranging (LiDAR) sensors. Traditional roof-mounted LiDAR systems provide a wide field of view (FOV) but have drawbacks such as increased vehicle height, reduced aerodynamics, and greater exposure to environmental factors. The proposed approach addresses these challenges by installing two LiDAR sensors at the ends of the vehicle bumper, optimizing the bumper-level FOV while minimizing interference from ground-level obstacles. The algorithm integrates point cloud data from both sensors using iterative closest point (ICP) matching and feature extraction techniques, enabling accurate vehicle odometry and 3D mapping. Real-world experiments were conducted using a test vehicle equipped with the proposed system. The results demonstrate that the bumper-mounted configuration provides localization accuracy comparable to roof-mounted systems, especially in the XY plane, despite minor Z-axis discrepancies. Additionally, the proposed system expands the horizontal FOV through dual LiDAR placement, addressing the limitation of rear-view coverage. These findings suggest that the bumper-mounted LiDAR system offers a practical and reliable solution for autonomous driving, enhancing robustness against environmental challenges while maintaining mapping accuracy. This study confirms the potential of bumper-mounted LiDAR configurations for real-world applications in autonomous vehicle technology.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"35174-35182"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10897995","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143527584","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}

引用次数: 0

Leveraging ResNet50 With Swin Attention for Accurate Detection of OCT Biomarkers Using Fundus Images

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544332

S. Tamilselvi;M. Suchetha;Rajiv Raman

{"title":"Leveraging ResNet50 With Swin Attention for Accurate Detection of OCT Biomarkers Using Fundus Images","authors":"S. Tamilselvi;M. Suchetha;Rajiv Raman","doi":"10.1109/ACCESS.2025.3544332","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544332","url":null,"abstract":"Diabetes can impact the retina and cause a decline in vision for patients as a result of Diabetic Retinopathy (DR). Diabetic Macular Edema (DME) is a complication that results from the chronic damage to the tiny blood vessels of the retina that arises in the non-proliferative stage of DR (NPDR) but can also be present in proliferative DR (PDR) and potentially leading to vision loss. The duration of diabetes in patients affects both the prevalence and incidence of macular edema, as well as the progression of retinopathy. Therefore, regular screening of diabetes patients for the early detection of retinal abnormalities is essential to prevent the development and progression of DR and DME. The proposed model predicts DME-associated biomarkers, typically identified in Optical Coherence Tomography (OCT), using 2D fundus images. These biomarkers include center-involved diabetic macular edema (ci-DME), neurosensory detachment (NSD), Intraretinal fluid (IRF), disorganization of the retinal inner layers (DRIL), hyperreflective foci (HRF), and disruptions in the inner segment/outer segment (IS/OS) junction, utilizing 2D fundus images. The model integrates the feature extraction capability of ResNet50 with the spatial structural domain knowledge provided by the Swin attention augmentation layer. 2D fundus image datasets were collected to train and evaluate the model. In two distinct datasets, the model achieved a validation accuracy of 85.7% (95% CI: 81.6–90.6%) and 89.5% (95% CI: 85.6–93.4%), Cohen’s Kappa of 0.68 (95% CI: 0.61–0.77) and 0.75 (95% CI: 0.67–0.82), sensitivity of 88.6% (95% CI: 85.6–92.1%) and 79.6% (95% CI: 70.6–85.4%), specificity of 79.6% (95% CI: 70.3–88.9%) and 93.7% (95% CI: 90.7–96.8%), respectively, with an overall validation accuracy of 87%. The proposed model helps in identifying the DME-associated biomarkers, using 2D fundus images making it a promising tool for detecting and assessing DME-related features.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"35203-35218"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10897982","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143527591","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}

引用次数: 0

Induction Motor Under Unbalanced Voltage Subharmonics

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544458

Piotr Gnaciński;Marcin Pepliński;Adam Muc;Damian Hallmann

{"title":"Induction Motor Under Unbalanced Voltage Subharmonics","authors":"Piotr Gnaciński;Marcin Pepliński;Adam Muc;Damian Hallmann","doi":"10.1109/ACCESS.2025.3544458","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544458","url":null,"abstract":"In addition to fundamental harmonics, the voltage waveform may contain various undesirable components, including subharmonics (subsynchronous interharmonics) and interharmonics (SaIs), which are components with frequencies less than or not an integer multiple of the fundamental frequency. Voltage SaIs are considered especially detrimental power quality disturbances, exerting negative impacts on rotating machinery. They are generated by various electrical equipment, including single-phase equipment. This work addresses the effects of the occurrence of subharmonics in one phase of the supply voltage on three-phase cage induction motors. The results of 2D finite element computations and experimental investigations show that subharmonics occurring in one phase cause high torque pulsations and unacceptable vibrations. For the investigated motor, the minimal value of the voltage subharmonic causing excessive vibration is more than twice that in the case of balanced three-phase subharmonics. It is concluded that power quality standards should contain separate limits for SaIs averaged for three line voltages and for SaIs occurring in each phase voltage.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"35946-35957"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10898003","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143527601","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}

引用次数: 0

Yolov5s-PSG: Improved Yolov5s-Based Helmet Recognition in Complex Scenes

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544280

Yi Li;Xiaolin Shi;Xiaolong Xu;Han Zhang;Fan Yang

{"title":"Yolov5s-PSG: Improved Yolov5s-Based Helmet Recognition in Complex Scenes","authors":"Yi Li;Xiaolin Shi;Xiaolong Xu;Han Zhang;Fan Yang","doi":"10.1109/ACCESS.2025.3544280","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544280","url":null,"abstract":"In the field of industrial safety, due to the existence of color, distance and other reasons in complex industrial environments caused by the helmet small target detection methods have the problem of misdetection and omission, and the Yolov5s model for real-time detection of helmets is not ideal. for the purpose of Improvement of extraction capabilities multi-scale features, improve the network’s focus on key features, and achieve an excellent balanced effect for accuracy and speed, the model Yolov5s-PSG based on Yolov5s improvement is proposed. First, the C3_PPA module improves the detection accuracy through a multi-branch feature extraction strategy and an attention mechanism. Second, the SimAM attention mechanism enhances the inspection accuracy by adaptively weighting the feature maps and thus improving the detection efficiency. Finally, the GSconv convolutional layer combines the SC module and the DSC module in a channel shuffling manner to infiltrate the information of the SC module within some of the information of the DSC module, which improves the detection speed without changing the accuracy.Yolov5s-PSG model outperforms the conventional target inspection in terms of accuracy, recall, map value, and loss function, where the accuracy is improved to 93.1%, recall to 91.2%, and map value to 94.6%. These findings are important for helmet inspection to find the optimal solution in terms of accuracy and speed for better helmet inspection in shop-floor factory environments.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"34915-34924"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10897966","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143518845","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}

引用次数: 0

Multi-Graph Anomaly Detection in Business Processes With Scalable Neural Architectures

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3544268

Stanley Hsu;Ege Gülce;Teoman Berkay Ayaz;Alper Ozcan;Akhan Akbulut

{"title":"Multi-Graph Anomaly Detection in Business Processes With Scalable Neural Architectures","authors":"Stanley Hsu;Ege Gülce;Teoman Berkay Ayaz;Alper Ozcan;Akhan Akbulut","doi":"10.1109/ACCESS.2025.3544268","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3544268","url":null,"abstract":"Business Process Management (BPM) solutions are critical for organizational efficiency, but their potential remains limited by inadequate effectiveness in anomaly detection capabilities for real-world deployment. This study addresses key challenges in developing production-ready anomaly detection systems that are scalable, efficient, and adaptable across diverse business domains. We propose several enhancements to a state-of-the-art graph-based autoencoder model to overcome these barriers. This includes improved artificial anomaly injection methods that more accurately reflect real-world scenarios to overcome the scarcity of annotated datasets in real-world environments. A comprehensive study of multiple model architectures is conducted, incorporating Graph Attention v2 in the encoder and replacing Gated Recurrent Unit (GRU) decoders with Transformers, thereby achieving comparable or superior performance with half the computational cost. Introducing a denoising objective alongside reconstruction, we lay the foundation for targeted training on domain-specific anomalies without compromising general detection capabilities. We demonstrate the solution’s reliability and generalizability in varied business domains by conducting comprehensive evaluations on diverse public and private datasets. The results indicate significant improvements in scalability and real-world applicability while maintaining and enhancing detection accuracy, with results showing up to 22% increase in anomaly detection performance.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"34969-34984"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10898002","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521282","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}

引用次数: 0

Rethinking Programming Education: A Lecture-Free Approach

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-21 DOI: 10.1109/ACCESS.2025.3542389

Abdulhadi Shoufan

{"title":"Rethinking Programming Education: A Lecture-Free Approach","authors":"Abdulhadi Shoufan","doi":"10.1109/ACCESS.2025.3542389","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3542389","url":null,"abstract":"Just-in-time hands-on experience is essential for learning programming. It not only helps reinforce learned concepts immediately, but also encourages the exploration of new concepts and language constructs. Integrating hands-on activities into traditional programming classrooms, however, presents a fundamental challenge. Coordinating instructor explanations with student coding activities and providing timely feedback to all students before moving forward is impractical for large or medium-sized classes. This paper demonstrates how cognitive and constructive learning principles can transform a classroom into an active learning environment that blends conceptual learning and hands-on experience for an object-oriented programming course. Instead of attending traditional lectures, our students use the class time to participate in Moodle-based activities that feature diverse question types and immediate feedback. They work at their own pace, individually or in small teams, while the instructor guides them and provides support as needed. In a section of 41 students, 88% preferred this approach over lectures. These students scored 14.2% higher on the same exam compared to 111 peers enrolled in other sections that used traditional teaching in the same semester.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"34758-34767"},"PeriodicalIF":3.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10899859","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143512856","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}

引用次数: 0

Toward Efficient Cancer Detection on Mobile Devices

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-02-20 DOI: 10.1109/ACCESS.2025.3543838

Janghyeon Lee;Jongyoul Park;Yongkeun Lee

{"title":"Toward Efficient Cancer Detection on Mobile Devices","authors":"Janghyeon Lee;Jongyoul Park;Yongkeun Lee","doi":"10.1109/ACCESS.2025.3543838","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3543838","url":null,"abstract":"Recent advancements in deep learning for cancer detection have achieved impressive accuracy, yet high computational costs and latency remain significant barriers for practical deployment on resource-constrained devices, such as smartphones and IoT platforms. This study focuses on optimizing MobileNetV1 and MobileNetV2 models to achieve more efficient, real-time cancer type identification. Through optimization strategies including selective layer unfreezing, pruning, and quantization, we demonstrate significant improvements in model size, inference time, and efficiency. For MobileNetV1, model size was reduced from 13.1 MB to 3.23 MB, and inference time was cut from 23 ms to 14 ms, with an F1 score above 0.99. For MobileNetV2, the model size was reduced from 9.41 MB to 2.82 MB, with inference times reduced from 24 ms to 13 ms, while maintaining a high F1 score of 0.98. The efficiency scores for MobileNetV1 and MobileNetV2 were 0.984 and 0.994, respectively, significantly outperforming other state-of-the-art neural networks such as VGG16 (efficiency score: 0.458), ResNet50 (0.418), and DenseNet121 (0.731). These findings demonstrate that with appropriate optimizations, MobileNet models can be deployed on edge devices, achieving high accuracy (above 95%), fast inference times (under one second), and superior efficiency, making them ideal candidates for real-time cancer detection in resource-constrained environments like mobile and IoT devices.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"34613-34626"},"PeriodicalIF":3.4,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10896646","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143489197","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}

引用次数: 0