In this study, the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) was used to improve the diagnosis of Alzheimer’s disease using medical imaging and the Alzheimer’s disease image dataset across four diagnostic classes. The WGAN-GP was employed for data augmentation. The original dataset, the augmented dataset and the combined data were mapped using Uniform Manifold Approximation and Projection (UMAP) in both a 2D and 3D space. The same combined interaction network analysis was then performed on the test data. The results showed that, for the test accuracy, the score was 30.46% for the original dataset (unbalanced), whereas for the WGAN-GP augmented dataset (balanced), it improved to 56.84%, indicating that the WGAN-GP augmentation can effectively address the unbalanced problem.
{"title":"Comprehensive Data Augmentation Approach Using WGAN-GP and UMAP for Enhancing Alzheimer’s Disease Diagnosis","authors":"Emi Yuda, Tomoki Ando, Itaru Kaneko, Yutaka Yoshida, Daisuke Hirahara","doi":"10.3390/electronics13183671","DOIUrl":"https://doi.org/10.3390/electronics13183671","url":null,"abstract":"In this study, the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) was used to improve the diagnosis of Alzheimer’s disease using medical imaging and the Alzheimer’s disease image dataset across four diagnostic classes. The WGAN-GP was employed for data augmentation. The original dataset, the augmented dataset and the combined data were mapped using Uniform Manifold Approximation and Projection (UMAP) in both a 2D and 3D space. The same combined interaction network analysis was then performed on the test data. The results showed that, for the test accuracy, the score was 30.46% for the original dataset (unbalanced), whereas for the WGAN-GP augmented dataset (balanced), it improved to 56.84%, indicating that the WGAN-GP augmentation can effectively address the unbalanced problem.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"3 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259474","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.3390/electronics13183678
Aimin Wang, Yu Li, Wenxuan Yang, Guangxu Pan
Since rail-to-ground insulation decreases, large-level direct currents (DCs) leak from railways and form metro stray currents, corroding the buried metal. To locate the rail-to-ground insulation deterioration area, a location method is proposed based on parameter identification methods and the monitored information including the station rail potentials, currents at the traction power substations (TPSs), and train traction currents and train positions. According to the monitoring information of two adjacent TPSs, the section location model of the metro line is proposed, in which the rail-to-ground conductances of the test section are equivalent to the lumped parameters. Using the rail resistivity and traction currents as the known information, the rail-to-ground conductances are calculated with the least square method (LSM). The rail-to-ground insulation deterioration sections are identified by comparing the calculated conductances with thresholds determined by the standard requirements and section lengths. Then, according to the section location results, a detailed location model of the degradation section is proposed, considering the location distance accuracy. Using the genetic algorithm (GA) to calculate the rail-to-ground conductances, degradation positions are located by comparing the threshold calculated with the standard requirements and location distance accuracy. The location method is verified by comparing the calculation results under different degradation conditions. Moreover, the applications of the proposed method to different degradation lengths and different numbers of degradation sections are analyzed. The results show that the proposed method can locate rail-to-ground insulation deterioration areas.
{"title":"Localization Method for Insulation Degradation Area of the Metro Rail-to-Ground Based on Monitor Information","authors":"Aimin Wang, Yu Li, Wenxuan Yang, Guangxu Pan","doi":"10.3390/electronics13183678","DOIUrl":"https://doi.org/10.3390/electronics13183678","url":null,"abstract":"Since rail-to-ground insulation decreases, large-level direct currents (DCs) leak from railways and form metro stray currents, corroding the buried metal. To locate the rail-to-ground insulation deterioration area, a location method is proposed based on parameter identification methods and the monitored information including the station rail potentials, currents at the traction power substations (TPSs), and train traction currents and train positions. According to the monitoring information of two adjacent TPSs, the section location model of the metro line is proposed, in which the rail-to-ground conductances of the test section are equivalent to the lumped parameters. Using the rail resistivity and traction currents as the known information, the rail-to-ground conductances are calculated with the least square method (LSM). The rail-to-ground insulation deterioration sections are identified by comparing the calculated conductances with thresholds determined by the standard requirements and section lengths. Then, according to the section location results, a detailed location model of the degradation section is proposed, considering the location distance accuracy. Using the genetic algorithm (GA) to calculate the rail-to-ground conductances, degradation positions are located by comparing the threshold calculated with the standard requirements and location distance accuracy. The location method is verified by comparing the calculation results under different degradation conditions. Moreover, the applications of the proposed method to different degradation lengths and different numbers of degradation sections are analyzed. The results show that the proposed method can locate rail-to-ground insulation deterioration areas.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"101 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259478","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.3390/electronics13183672
Daniel Djolev, Milena Lazarova, Ognyan Nakov
In recent years, rapid technological advancements have propelled blockchain and artificial intelligence (AI) into prominent roles within the digital industry, each having unique applications. Blockchain, recognized for its secure and transparent data storage, and AI, a powerful tool for data analysis and decision making, exhibit common features that render them complementary. At the same time, machine learning has become a robust and influential technology, adopted by many companies to address non-trivial technical problems. This adoption is fueled by the vast amounts of data generated and utilized in daily operations. An intriguing intersection of blockchain and AI occurs in the realm of federated learning, a distributed approach allowing multiple parties to collaboratively train a shared model without centralizing data. This paper presents a decentralized platform FBLearn for the implementation of federated learning in blockchain, which enables us to harness the benefits of federated learning without the necessity of exchanging sensitive customer or product data, thereby fostering trustless collaboration. As the decentralized blockchain network is introduced in the distributed model training to replace the centralized server, global model aggregation approaches have to be utilized. This paper investigates several techniques for model aggregation based on the local model average and ensemble using either local or globally distributed validation data for model evaluation. The suggested aggregation approaches are experimentally evaluated based on two use cases of the FBLearn platform: credit risk scoring using a random forest classifier and credit card fraud detection using a logistic regression. The experimental results confirm that the suggested adaptive weight calculation and ensemble techniques based on the quality of local training data enhance the robustness of the global model. The performance evaluation metrics and ROC curves prove that the aggregation strategies successfully isolate the influence of the low-quality models on the final model. The proposed system’s ability to outperform models created with separate datasets underscores its potential to enhance collaborative efforts and to improve the accuracy of the final global model compared to each of the local models. Integrating blockchain and federated learning presents a forward-looking approach to data collaboration while addressing privacy concerns.
{"title":"FBLearn: Decentralized Platform for Federated Learning on Blockchain","authors":"Daniel Djolev, Milena Lazarova, Ognyan Nakov","doi":"10.3390/electronics13183672","DOIUrl":"https://doi.org/10.3390/electronics13183672","url":null,"abstract":"In recent years, rapid technological advancements have propelled blockchain and artificial intelligence (AI) into prominent roles within the digital industry, each having unique applications. Blockchain, recognized for its secure and transparent data storage, and AI, a powerful tool for data analysis and decision making, exhibit common features that render them complementary. At the same time, machine learning has become a robust and influential technology, adopted by many companies to address non-trivial technical problems. This adoption is fueled by the vast amounts of data generated and utilized in daily operations. An intriguing intersection of blockchain and AI occurs in the realm of federated learning, a distributed approach allowing multiple parties to collaboratively train a shared model without centralizing data. This paper presents a decentralized platform FBLearn for the implementation of federated learning in blockchain, which enables us to harness the benefits of federated learning without the necessity of exchanging sensitive customer or product data, thereby fostering trustless collaboration. As the decentralized blockchain network is introduced in the distributed model training to replace the centralized server, global model aggregation approaches have to be utilized. This paper investigates several techniques for model aggregation based on the local model average and ensemble using either local or globally distributed validation data for model evaluation. The suggested aggregation approaches are experimentally evaluated based on two use cases of the FBLearn platform: credit risk scoring using a random forest classifier and credit card fraud detection using a logistic regression. The experimental results confirm that the suggested adaptive weight calculation and ensemble techniques based on the quality of local training data enhance the robustness of the global model. The performance evaluation metrics and ROC curves prove that the aggregation strategies successfully isolate the influence of the low-quality models on the final model. The proposed system’s ability to outperform models created with separate datasets underscores its potential to enhance collaborative efforts and to improve the accuracy of the final global model compared to each of the local models. Integrating blockchain and federated learning presents a forward-looking approach to data collaboration while addressing privacy concerns.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"118 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.3390/electronics13183673
Sofia Kouah, Asma Saighi, Maryem Ammi, Aymen Naït Si Mohand, Marwa Ines Kouah, David Megías
The Internet of Things refers to a network of interconnected devices, objects, and systems, that can interact with one another without human intervention. The adoption of IoT technology has expanded rapidly, significantly impacting various fields, including smart healthcare, intelligent transportation, agriculture, and smart homes. This paper focuses on smart street lighting, which represents the core piece of the smart city and the key public service for citizens’ safety. Nevertheless, it poses substantial challenges related to energy consumption, especially during energy crises. This work aims to provide an advanced solution that enables intelligent control of street lighting, enhances human safety, reduces CO2 emissions and light pollution, and optimizes energy consumption, as well as facilitates maintenance of the lighting network. The solution is twofold: First, it introduces IoT-based smart street lighting referential models; second, it presents a framework for controlling smart street lighting based on the referential models. The proposal uses an IoT-based fuzzy multi-agent systems approach to address the challenges of smart street lighting. The approach leverages the strengths and properties of fuzzy logic and multi-agent systems to address the system requirements. This is illustrated through a testbed case study conducted on a concrete IoT prototype.
{"title":"Internet of Things-Based Multi-Agent System for the Control of Smart Street Lighting","authors":"Sofia Kouah, Asma Saighi, Maryem Ammi, Aymen Naït Si Mohand, Marwa Ines Kouah, David Megías","doi":"10.3390/electronics13183673","DOIUrl":"https://doi.org/10.3390/electronics13183673","url":null,"abstract":"The Internet of Things refers to a network of interconnected devices, objects, and systems, that can interact with one another without human intervention. The adoption of IoT technology has expanded rapidly, significantly impacting various fields, including smart healthcare, intelligent transportation, agriculture, and smart homes. This paper focuses on smart street lighting, which represents the core piece of the smart city and the key public service for citizens’ safety. Nevertheless, it poses substantial challenges related to energy consumption, especially during energy crises. This work aims to provide an advanced solution that enables intelligent control of street lighting, enhances human safety, reduces CO2 emissions and light pollution, and optimizes energy consumption, as well as facilitates maintenance of the lighting network. The solution is twofold: First, it introduces IoT-based smart street lighting referential models; second, it presents a framework for controlling smart street lighting based on the referential models. The proposal uses an IoT-based fuzzy multi-agent systems approach to address the challenges of smart street lighting. The approach leverages the strengths and properties of fuzzy logic and multi-agent systems to address the system requirements. This is illustrated through a testbed case study conducted on a concrete IoT prototype.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"6 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.3390/electronics13183680
Jack Grotke, Austin Alcantara, Joe Amitrano, Dhruv R. Seshadri
This study investigates the impact of a five-week functional exercise intervention designed to enhance the muscular endurance of the posterior shoulder musculature, aiming to mitigate shoulder fatigue and overuse injury. Twelve Division I collegiate butterfly swimmers were recruited and evenly randomized into exercise (EX) and control (CTRL) groups. Weekly 100-yard butterfly sprints were performed, with Muscle Oxygen Saturation (SmO2) continuously monitored using a wearable near-infrared spectroscopy (NIRS) device. This study is among the first to utilize wearable NIRS devices to monitor SmO2 underwater during swimming, demonstrating that a targeted 5-week exercise program significantly improves posterior shoulder endurance, as evidenced by increased Posterior Shoulder Endurance Test (PSET) scores and distinctive SmO2 adaptations in the EX-group compared to the CTRL group. These findings suggest that targeted dryland exercises can enhance posterior shoulder endurance with long-term implications for potentially reducing injury risk and improving performance.
{"title":"Functional Exercise Induces Adaptations in Muscle Oxygen Saturation in Division One Collegiate Butterfly Swimmers: A Randomized Controlled Trial","authors":"Jack Grotke, Austin Alcantara, Joe Amitrano, Dhruv R. Seshadri","doi":"10.3390/electronics13183680","DOIUrl":"https://doi.org/10.3390/electronics13183680","url":null,"abstract":"This study investigates the impact of a five-week functional exercise intervention designed to enhance the muscular endurance of the posterior shoulder musculature, aiming to mitigate shoulder fatigue and overuse injury. Twelve Division I collegiate butterfly swimmers were recruited and evenly randomized into exercise (EX) and control (CTRL) groups. Weekly 100-yard butterfly sprints were performed, with Muscle Oxygen Saturation (SmO2) continuously monitored using a wearable near-infrared spectroscopy (NIRS) device. This study is among the first to utilize wearable NIRS devices to monitor SmO2 underwater during swimming, demonstrating that a targeted 5-week exercise program significantly improves posterior shoulder endurance, as evidenced by increased Posterior Shoulder Endurance Test (PSET) scores and distinctive SmO2 adaptations in the EX-group compared to the CTRL group. These findings suggest that targeted dryland exercises can enhance posterior shoulder endurance with long-term implications for potentially reducing injury risk and improving performance.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"19 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper designs a safe, low-cost, and efficient permanent magnet synchronous motor (PMSM) booster pump system. The aim is to enhance the pump’s safety and reduce the incidence of electric shock accidents, while also achieving cost reduction and efficiency improvement. The pump components are made of a plastic material, and a safe voltage of 36 V is used as the operating voltage. Additionally, the PMSM is chosen to replace the induction motor (IM) as the pump’s driving device, utilizing sensorless control and field-weakening control strategies. The study results show that when the flow rate is 1.51 m3/h, the efficiency of the PMSM low-voltage pump can reach up to 20.86%. At the same flow rate of 1 m3/h, compared to other pumps, the PMSM low-voltage pump exhibits higher head, energy savings, and efficiency. The proposed PMSM low-voltage pump offers advantages such as high efficiency, energy savings, safety, and low cost. This study provides a reference for the domestic PMSM pump industry.
{"title":"Low-Voltage Water Pump System Based on Permanent Magnet Synchronous Motor","authors":"Xinrong Jin, Leifu Zhou, Tingting Lang, Yanbing Jiang","doi":"10.3390/electronics13183674","DOIUrl":"https://doi.org/10.3390/electronics13183674","url":null,"abstract":"This paper designs a safe, low-cost, and efficient permanent magnet synchronous motor (PMSM) booster pump system. The aim is to enhance the pump’s safety and reduce the incidence of electric shock accidents, while also achieving cost reduction and efficiency improvement. The pump components are made of a plastic material, and a safe voltage of 36 V is used as the operating voltage. Additionally, the PMSM is chosen to replace the induction motor (IM) as the pump’s driving device, utilizing sensorless control and field-weakening control strategies. The study results show that when the flow rate is 1.51 m3/h, the efficiency of the PMSM low-voltage pump can reach up to 20.86%. At the same flow rate of 1 m3/h, compared to other pumps, the PMSM low-voltage pump exhibits higher head, energy savings, and efficiency. The proposed PMSM low-voltage pump offers advantages such as high efficiency, energy savings, safety, and low cost. This study provides a reference for the domestic PMSM pump industry.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"14 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.3390/electronics13183675
Mohammad Moshawrab, Mehdi Adda, Abdenour Bouzouane, Hussein Ibrahim, Ali Raad
With the ability to analyze data, artificial intelligence technology and its offshoots have made difficult tasks easier. The tools of these technologies are now used in almost every aspect of life. For example, Machine Learning (ML), an offshoot of artificial intelligence, has become the focus of interest for researchers in industry, education, healthcare and other disciplines and has proven to be as efficient as, and in some cases better than, experts in answering various problems. However, the obstacles to ML’s progress are still being explored, and Federated Learning (FL) has been presented as a solution to the problems of privacy and confidentiality. In the FL approach, users do not disclose their data throughout the learning process, which improves privacy and security. In this article, we look at the security and privacy concepts of FL and the threats and attacks it faces. We also address the security measures used in FL aggregation procedures. In addition, we examine and discuss the use of homomorphic encryption to protect FL data exchange, as well as other security strategies. Finally, we discuss security and privacy concepts in FL and what additional improvements could be made in this context to increase the efficiency of FL algorithms.
{"title":"Securing Federated Learning: Approaches, Mechanisms and Opportunities","authors":"Mohammad Moshawrab, Mehdi Adda, Abdenour Bouzouane, Hussein Ibrahim, Ali Raad","doi":"10.3390/electronics13183675","DOIUrl":"https://doi.org/10.3390/electronics13183675","url":null,"abstract":"With the ability to analyze data, artificial intelligence technology and its offshoots have made difficult tasks easier. The tools of these technologies are now used in almost every aspect of life. For example, Machine Learning (ML), an offshoot of artificial intelligence, has become the focus of interest for researchers in industry, education, healthcare and other disciplines and has proven to be as efficient as, and in some cases better than, experts in answering various problems. However, the obstacles to ML’s progress are still being explored, and Federated Learning (FL) has been presented as a solution to the problems of privacy and confidentiality. In the FL approach, users do not disclose their data throughout the learning process, which improves privacy and security. In this article, we look at the security and privacy concepts of FL and the threats and attacks it faces. We also address the security measures used in FL aggregation procedures. In addition, we examine and discuss the use of homomorphic encryption to protect FL data exchange, as well as other security strategies. Finally, we discuss security and privacy concepts in FL and what additional improvements could be made in this context to increase the efficiency of FL algorithms.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"34 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.3390/electronics13183681
Jiaqi Li, Seung-Hoon Hwang
Reformation of the 3.7–4.0 GHz band to expand 5G communication deployment poses a risk of 5G signals disrupting radar altimeter operation, leading to data loss or inaccuracies. Thus, this paper proposes a guard band protection method to facilitate the coexistence of 5G base stations and radar altimeters operating in the 4.2–4.4 GHz band. To enhance the adjacent channel leakage ratio (ACLR), we implemented spectral regrowth on an oversampled waveform using a high-power amplifier model, filtering out-of-band spectral emissions. The results demonstrated that a 150 MHz guard band enables coexistence, except in the case of the 16-by-16 antenna array in rural environments. Notably, for the 4-by-4 antenna array in urban environments, coexistence can be achieved using a 50 MHz guard band. The proposed mitigation techniques may also be extended to promote coexistence between non-terrestrial networks and 5G communication systems, including satellites, unmanned aerial vehicles, and hot air balloons.
{"title":"Guard Band Protection Scheme to Facilitate Coexistence of 5G Base Stations and Radar Altimeters","authors":"Jiaqi Li, Seung-Hoon Hwang","doi":"10.3390/electronics13183681","DOIUrl":"https://doi.org/10.3390/electronics13183681","url":null,"abstract":"Reformation of the 3.7–4.0 GHz band to expand 5G communication deployment poses a risk of 5G signals disrupting radar altimeter operation, leading to data loss or inaccuracies. Thus, this paper proposes a guard band protection method to facilitate the coexistence of 5G base stations and radar altimeters operating in the 4.2–4.4 GHz band. To enhance the adjacent channel leakage ratio (ACLR), we implemented spectral regrowth on an oversampled waveform using a high-power amplifier model, filtering out-of-band spectral emissions. The results demonstrated that a 150 MHz guard band enables coexistence, except in the case of the 16-by-16 antenna array in rural environments. Notably, for the 4-by-4 antenna array in urban environments, coexistence can be achieved using a 50 MHz guard band. The proposed mitigation techniques may also be extended to promote coexistence between non-terrestrial networks and 5G communication systems, including satellites, unmanned aerial vehicles, and hot air balloons.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"167 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Leveraging its high accuracy and stability, deep-learning-based coronary artery detection technology has been extensively utilized in diagnosing coronary artery diseases. However, traditional algorithms for localizing coronary stenosis often fall short when detecting stenosis in branch vessels, which can pose significant health risks due to factors like imaging angles and uneven contrast agent distribution. To tackle these challenges, we propose a preprocessing method that integrates Hessian-based vascular enhancement and image fusion as prerequisites for deep learning. This approach enhances fuzzy features in coronary angiography images, thereby increasing the neural network’s sensitivity to stenosis characteristics. We assessed the effectiveness of this method using the latest deep learning networks, such as YOLOv10, YOLOv9, and RT-DETR, across various evaluation metrics. Our results show that our method improves AP50 accuracy by 4.84% and 5.07% on RT-DETR R101 and YOLOv10-X, respectively, compared to images without special pre-processing. Furthermore, our analysis of different imaging angles on stenosis localization detection indicates that the left coronary artery zero is the most suitable for detecting stenosis with a AP50(%) value of 90.5. The experimental results have revealed that the proposed method is effective as a preprocessing technique for deep-learning-based coronary angiography image processing and enhances the model’s ability to identify stenosis in small blood vessels.
{"title":"A Hessian-Based Deep Learning Preprocessing Method for Coronary Angiography Image Analysis","authors":"Yanjun Li, Takaaki Yoshimura, Yuto Horima, Hiroyuki Sugimori","doi":"10.3390/electronics13183676","DOIUrl":"https://doi.org/10.3390/electronics13183676","url":null,"abstract":"Leveraging its high accuracy and stability, deep-learning-based coronary artery detection technology has been extensively utilized in diagnosing coronary artery diseases. However, traditional algorithms for localizing coronary stenosis often fall short when detecting stenosis in branch vessels, which can pose significant health risks due to factors like imaging angles and uneven contrast agent distribution. To tackle these challenges, we propose a preprocessing method that integrates Hessian-based vascular enhancement and image fusion as prerequisites for deep learning. This approach enhances fuzzy features in coronary angiography images, thereby increasing the neural network’s sensitivity to stenosis characteristics. We assessed the effectiveness of this method using the latest deep learning networks, such as YOLOv10, YOLOv9, and RT-DETR, across various evaluation metrics. Our results show that our method improves AP50 accuracy by 4.84% and 5.07% on RT-DETR R101 and YOLOv10-X, respectively, compared to images without special pre-processing. Furthermore, our analysis of different imaging angles on stenosis localization detection indicates that the left coronary artery zero is the most suitable for detecting stenosis with a AP50(%) value of 90.5. The experimental results have revealed that the proposed method is effective as a preprocessing technique for deep-learning-based coronary angiography image processing and enhances the model’s ability to identify stenosis in small blood vessels.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"19 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259544","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.3390/electronics13183677
Abdallah Al-Sabbagh, Khalil Hamze, Samiya Khan, Mahmoud Elkhodr
Phishing attacks continue to pose a significant threat to cybersecurity, employing increasingly sophisticated techniques to deceive victims into revealing sensitive information or downloading malware. This paper presents a comprehensive study on the application of Machine Learning (ML) techniques for identifying phishing websites, with a focus on enhancing detection accuracy and efficiency. We propose an approach that integrates the CfsSubsetEval attribute evaluator with the K-Means Clustering algorithm to improve phishing detection capabilities. Our method was evaluated using datasets of varying sizes (2000, 7000, and 10,000 samples) from a publicly available repository. Simulation results demonstrate that our approach achieves an accuracy of 89.2% on the 2000-sample dataset, outperforming the traditional kernel K-Means algorithm, which achieved an accuracy of 51.5%. Further analysis using precision, recall, and F1-score metrics corroborates the effectiveness of our method. We also discuss the scalability and real-world applicability of our approach, addressing limitations and proposing future research directions. This study contributes to the ongoing efforts to develop robust, efficient, and adaptable phishing detection systems in the face of evolving cyber threats.
{"title":"An Enhanced K-Means Clustering Algorithm for Phishing Attack Detections","authors":"Abdallah Al-Sabbagh, Khalil Hamze, Samiya Khan, Mahmoud Elkhodr","doi":"10.3390/electronics13183677","DOIUrl":"https://doi.org/10.3390/electronics13183677","url":null,"abstract":"Phishing attacks continue to pose a significant threat to cybersecurity, employing increasingly sophisticated techniques to deceive victims into revealing sensitive information or downloading malware. This paper presents a comprehensive study on the application of Machine Learning (ML) techniques for identifying phishing websites, with a focus on enhancing detection accuracy and efficiency. We propose an approach that integrates the CfsSubsetEval attribute evaluator with the K-Means Clustering algorithm to improve phishing detection capabilities. Our method was evaluated using datasets of varying sizes (2000, 7000, and 10,000 samples) from a publicly available repository. Simulation results demonstrate that our approach achieves an accuracy of 89.2% on the 2000-sample dataset, outperforming the traditional kernel K-Means algorithm, which achieved an accuracy of 51.5%. Further analysis using precision, recall, and F1-score metrics corroborates the effectiveness of our method. We also discuss the scalability and real-world applicability of our approach, addressing limitations and proposing future research directions. This study contributes to the ongoing efforts to develop robust, efficient, and adaptable phishing detection systems in the face of evolving cyber threats.","PeriodicalId":11646,"journal":{"name":"Electronics","volume":"374 1","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142259476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：