Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Artificial intelligence in vestibular disorder diagnosis
Get PDFVertigo is a prevalent symptom of vestibular disorders, with ocular nystagmus analysis serving as a key indicator for distinguishing between peripheral and central vestibular conditions. Videonystagmography (VNG) provides objective and reliable measurements, making it a valuable tool for clinical assessments. However, the complexity and variability of vestibular diseases pose challenges for conventional VNG methods, such as caloric, kinetic, and saccadic tests, in accurately identifying vertigo subtypes. Traditional diagnostic approaches often fail to fully utilize nystagmus characteristics in correlating with specific vestibular disorders, limiting their effectiveness. Recent advancements in artificial intelligence (AI), particularly deep learning and machine learning (ML), offer promising solutions for improving vertigo diagnosis. These technologies facilitate automated, rapid, and precise analysis by extracting relevant clinical features and classifying vestibular disorders with higher accuracy. ML-based models enhance diagnostic reliability, reducing human bias and subjectivity in assessment. This study reviews the latest research on feature extraction and ML applications in vertigo diagnosis, emphasizing their potential to revolutionize clinical decision-making. It aims to provide a comprehensive understanding of AI-driven approaches and their role in advancing vertigo analysis, paving the way for more effective diagnostic methodologies in the future
Multi-objective optimization for algorithmic trading in the Vietnamese stock market
Get PDFThis study aims to optimize algorithmic trading strategies using the relative strength index (RSI) and the moving average convergence divergence (MACD) indicators in the Vietnamese stock market. An automated trading system is constructed to optimize indicator parameters using multi-objective particle swarm optimization (PSO) over three objective functions: total return, win rate, and number of trades. The system employs simultaneous optimization of parameters and signal aggregation for developing the optimal selection strategy. Based on daily Vietnam index data from 2018 to 2024, the results show that the PSO method surpasses the differential evolution (DE) method in both returns and execution time. Additionally, the optimal selection strategy achieves superior performance compared to benchmark strategies. It also demonstrates the ability to adapt to the preferences of traders by selecting appropriate indicators. Traders can use the MACD indicator to seek higher profits, while the RSI indicator is more suitable for minimizing transaction costs in a volatile market
Optimizing fatigue life predictions for scraper rings: classical vs modern models
Get PDFThis study provides a comprehensive comparative evaluation of classical and modern predictive models for fatigue life in scraper rings of internal combustion engines, which operate under high thermo-mechanical stresses. Accurate fatigue life predictions are essential for optimizing engine component design, preventing both over- and under-engineering while ensuring long-term reliability. The effectiveness of both traditional models and newer advanced approaches was analyzed using loading profiles that replicate real-world engine operating conditions. Results indicate that stress-life models offer more reliable predictions for high-cycle fatigue scenarios, while strain-life models perform better under low-cycle fatigue conditions. Furthermore, fracture mechanics models show great promise in predicting crack propagation and identifying failure mechanisms. Detailed inspections and Légraud-Poirier (LP) tests confirmed fatigue-induced cracking at critical locations of the scraper rings, emphasizing the importance of incorporating multi-axial loading in fatigue assessments. The findings underscore the necessity for using comprehensive loading profiles and thorough inspections to enhance the accuracy and dependability of fatigue life predictions, which are critical for improving the performance and durability of engine components
Smart wheat agriculture: an in-depth framework for optimized crop agroanalytics utilizing internet of things
Get PDFPrecision agriculture can be revolutionized by incorporating internet of things (IoT) technologies to maximize crop yield, especially for key crops like wheat. The creation and application of an IoT-enabled monitoring system intended especially for wheat farming is presented in this study. The system provides real-time data on important agronomic characteristics, such as soil health, temperature, humidity, and crop growth stages, by integrating a network of soil moisture sensors, weather stations, and remote sensing devices. With the help of the monitoring system, field conditions can be continuously and remotely observed, giving farmers the ability to make data-driven decisions that improve crop output and resource efficiency. The system can reduce input waste and increase output by optimizing irrigation schedules, adjusting fertilizer applications, and detecting early signs of crop stress or disease through real-time data analysis. Significant gains have been shown in production results, farm sustainability overall, and water usage efficiency in field studies carried out in different wheat-growing locations. According to the research, IoT-based monitoring systems can be extremely helpful in modernizing wheat production by offering useful information that results in more exact and environmentally friendly farming methods
Residual pixel-wise semantic segmentation for assessing enlarged fetal heart: a preliminary study
Get PDFThe four-chamber view is a crucial scan plane routinely employed in both second-trimester perinatal screening and fetal echocardiographic examinations. Sonographers typically measure biometrics in this plane, such as the cardiothoracic ratio (CTR) and heart axis, to diagnose fetal heart anomalies. However, due to the echocardiographic artifacts, the assessment not only suffers from low efficiency but also inconsistent results depending on the operators’ skills. This study proposes a residual pixel-wise semantic segmentation, which segmented the fetal heart and thoracic contours in a 4-chamber view for assessing an enlarged fetal heart condition. The accuracy of intersection-over-union (IoU) and dice coefficient similarity (DCS) is used for model validation to further regulate the evaluation procedure. We use 1174 US images, comprising about 560 enlarged heart images, and about 614 normal heart images. Out of these data, 248 images are used for unseen data, and the remaining for training/validation processes. The performance of the proposed model, when tested on unseen data, achieved satisfactory results with 97.71% accuracy, 90.36% IoU, and 94.93% DCS. These metrics collectively demonstrate the satisfactory performance of the proposed model compared to existing segmentation models. The outcomes underscore that the proposed model establishes a state-of-the-art standard for enlarged fetal heart detection
Effective crop categorization using wavelet transform based optimized long short-term memory technique
Get PDFEffective crop categorization is important for keeping track of how crops grow and how much they produce in the future. Gathering crop data on categories, regions, and space distribution in a timely and accurate way could give a scientifically sound reason for changes to the way crops are organized. Polarimetric synthetic aperture radar dataset provides sufficient information for accurate crop categorization. It is essential to classify crops in order to successfully. This article presents wavelet transform (WT) based optimizedlong short-term memory (LSTM) deep learning (DL) for effective crop categorization. Image denoising is performed by WT. Denoising algorithms for images attempt to find a middle ground between totally removing all of the image’s noise and preserving essential, signal-free components of the picture in their original state. After denoising of images, crop image classification is achieved by LSTM and support vector machine (SVM) algorithm. LSTM has achieved 99.5% accuracy
Technical-economic performance of fuel cell integration in autonomous hybrid systems
Get PDFIncreasing energy demand and greenhouse gas emissions reinforce the importance of renewable resources in energy systems. This study evaluates the technical and economic viability of integrating a fuel cell (FC) in autonomous hybrid systems to supply a community in Algeria, with an average power of 6.91 kW and a daily energy requirement of 165.6 kWh. Four hybrid system configurations were compared using HOMER software: i) photovoltaic (PV) and batteries (BAT), ii) PV, BAT, and diesel generator (DG), iii) PV, BAT, and FC, and iv) PV, BAT, DG, and FC. The PV/BAT/DG/FC system was identified as the optimal configuration, balancing energy efficiency, reducing energy surpluses, reducing reliance on DG, and reducing CO2 emissions while maintaining competitive energy costs. These results demonstrate that the integration of FC can improve the sustainability and stability of autonomous hybrid energy systems
Blockchain for future smart grid: a comprehensive survey
Get PDFDue to the unique features and characteristics of blockchain technology, its applications have expanded across various sectors, including finance, banking, supply chains, and smart grids (SGs). Blockchain ensures security and trust in transactions without requiring a third party, making it particularly valuable in decentralized systems. This paper explores the integration of blockchain technology into SG systems. It begins with a comprehensive review of conventional and smart power grids, identifying the key challenges modern SGs face, particularly issues related to trust and fraud. An in-depth analysis of blockchain technology follows, highlighting its potential, advantages, and defining characteristics. The study then examines several blockchain-based SG applications and provides a comparative analysis of prior research. The findings of this review illuminate the critical role of blockchain in enhancing SG performance by addressing trust and fraud prevention challenges. Furthermore, this research has significant implications for the energy sector, as it underscores the potential of blockchain to revolutionize SGs through increased security, transparency, and efficiency. By providing a foundation for future studies, this paper aims to guide the development of unified blockchain frameworks that address scalability, privacy, and energy management, paving the way for a more secure and efficient decentralized energy syste
Insights into peer-to-peer botnet dynamics: reviewing emulation testbeds and proposing a conceptual model
Get PDFPeer-to-peer (P2P) botnets have emerged as a resilient cybercrime tool, utilizing decentralized architectures to evade detection and complicate takedown efforts. Existing botnet emulation testbeds often fall short in replicating the dynamic and large-scale environments that these botnets operate in, limiting their effectiveness in research and defense strategy development. This paper addresses these gaps by proposing a scalable, flexible emulation testbed for P2P botnets that integrates advanced virtualization and automation technologies. Our framework enables the accurate emulation of real-world botnet behaviors without relying on reverse engineering, offering researchers a secure and adaptable environment to test and validate botnet detection and mitigation strategies. The testbed’s dynamic scalability and robust configuration management streamline experimentation across diverse network topologies and botnet types. Our results show that this approach significantly enhances the ability to study P2P botnets in a controlled, reproducible setting, providing valuable insights for advancing cybersecurity defenses
Development of classification model for thoracic diseases with chest X-ray images using deep convolutional neural network
Get PDFThoracic disease is a medical condition in the chest wall region. Accurate thoracic disease diagnosis in patients is critical for effective treatment. Atelectasis, mass, pneumonia, and pneumothorax are thoracic diseases that can lead to life-threatening conditions if not detected and treated early enough. When diagnosing these diseases, human expertise can also be susceptible to errors due to fatigue or emotional factors. This research proposes developing a real-time deep learning-based classification model for thoracic diseases. Three deep convolutional neural network (CNN) models - MobileNetV3Large, ResNet-50, and EfficientNetB7 - were evaluated for classifying thoracic diseases from chest X-ray images. The models were tested in 5-class (atelectasis, mass, pneumothorax, pneumonia, and normal), 4-class (atelectasis, pneumothorax, pneumonia, and normal), and 3-class (atelectasis, pneumonia, and normal) modes to assess the impact of high interclass similarity. Retrained MobileNetV3Large achieved the highest classification accuracy: 75.72% next to ResNet-50 (75.2%) and last EfficientNetB7 (73.03%). For the 4-class, EfficientNetB7 (88.08%) led with MobileNetV3Large in the last (87.08%), but MobileNetV3Large led the 3-way with 97.88% with EfficientNetB7 again in the last (96.55%). These results indicate that MobileNetV3 can effectively distinguish and diagnose thoracic diseases from chest X-rays, even with interclass similarity and supports the use of computer-aided detection systems in thoracic disease classification