Search Results (73)

Mental stress is a psychological or emotional strain that typically occurs because of threatening, challenging, and overwhelming conditions and affects human behavior. Various factors, such as professional, environmental, and personal pressures, often trigger it. In recent years, various deep learning (DL)-based schemes using electroencephalograms (EEGs) have been proposed. However, the effectiveness of DL-based schemes is challenging because of the intricate DL structure, class imbalance problems, poor feature representation, low-frequency resolution problems, and complexity of multi-channel signal processing. This paper presents a novel hybrid DL framework, BDDNet, which combines a deep convolutional neural network (DCNN), bidirectional long short-term memory (BiLSTM), and deep belief network (DBN). BDDNet provides superior spectral–temporal feature depiction and better long-term dependency on the local and global features of EEGs. BDDNet accepts multiple EEG features (MEFs) that provide the spectral and time-domain features of EEGs. A novel improved crow search algorithm (ICSA) was presented for channel selection to minimize the computational complexity of multichannel stress detection. Further, the novel employee optimization algorithm (EOA) is utilized for the hyper-parameter optimization of hybrid BDDNet to enhance the training performance. The outcomes of the novel BDDNet were assessed using a public DEAP dataset. The BDDNet-ICSA offers improved recall of 97.6%, precision of 97.6%, F1-score of 97.6%, selectivity of 96.9%, negative predictive value NPV of 96.9%, and accuracy of 97.3% to traditional techniques. Full article

This paper proposes an Adam-optimized Deep Belief Networks (Adam-DBNs) denoising method for throat-attached piezoelectric signals. The method aims to process mechanical vibration signals captured through polyvinylidene fluoride (PVDF) sensors attached to the throat region, which are typically contaminated by environmental noise and physiological noise. First, the short-time Fourier transform (STFT) is utilized to convert the original signals into the time–frequency domain. Subsequently, the masked time–frequency representation is reconstructed into the time domain through a diagonal average-based inverse STFT. To address complex nonlinear noise structures, a Deep Belief Network is further adopted to extract features and reconstruct clean signals, where the Adam optimization algorithm ensures the efficient convergence and stability of the training process. Compared with traditional Convolutional Neural Networks (CNNs), Adam-DBNs significantly improve waveform similarity by 6.77% and reduce the local noise energy residue by 0.099696. These results demonstrate that the Adam-DBNs method exhibits substantial advantages in signal reconstruction fidelity and residual noise suppression, providing an efficient and robust solution for throat-attached piezoelectric sensor signal enhancement tasks. Full article

Ransomware, a significant cybersecurity threat, encrypts files and causes substantial damage, making early detection crucial yet challenging. This paper introduces a novel multi-phase framework for early ransomware detection, designed to enhance accuracy and minimize false positives. The framework addresses the limitations of existing methods by integrating operational data with situational and threat intelligence, enabling it to dynamically adapt to the evolving ransomware landscape. Key innovations include (1) data augmentation using a Bi-Gradual Minimax Generative Adversarial Network (BGM-GAN) to generate synthetic ransomware attack patterns, addressing data insufficiency; (2) Incremental Mutual Information Selection (IMIS) for dynamically selecting relevant features, adapting to evolving ransomware behaviors and reducing computational overhead; and (3) a Deep Belief Network (DBN) detection architecture, trained on the augmented data and optimized with Uncertainty-Aware Dynamic Early Stopping (UA-DES) to prevent overfitting. The model demonstrates a 4% improvement in detection accuracy (from 90% to 94%) through synthetic data generation and reduces false positives from 15.4% to 14%. The IMIS technique further increases accuracy to 96% while reducing false positives. The UA-DES optimization boosts accuracy to 98.6% and lowers false positives to 10%. Overall, this framework effectively addresses the challenges posed by evolving ransomware, significantly enhancing detection accuracy and reliability. Full article

Older adults make up about 12% of the public sector, primary care, and hospital use and represent a large proportion of the users of healthcare services. Older people are also more vulnerable to serious injury from unexpected falls due to tripping, slipping, or illness. This underscores the immediate necessity of stable and cost-effective e-health technologies in maintaining independent living. Artificial intelligence (AI) and machine learning (ML) offer promising solutions for early fall prediction and continuous health monitoring. This paper introduces a novel cooperative AI model that forecasts the risk of future falls in the elderly based on behavioral and health abnormalities. Two AI models’ predictions are combined to produce accurate predictions: The ${AI}_1$ model is based on vital signs using Fuzzy Logic, and the ${AI}_2$ model is based on Activities of Daily Living (ADLs) using a Deep Belief Network (DBN). A meta-model then combines the outputs to generate a total fall risk prediction. The results show 85.71% sensitivity, 100% specificity, and 90.00% prediction accuracy when compared to the Morse Falls Scale (MFS). This emphasizes how deep learning-based cooperative systems can improve well-being for older adults living alone, facilitate more precise fall risk assessment, and improve preventive care. Full article

The big data collected from structural health monitoring systems (SHMs), combined with the rapid advances in machine learning (ML), have enabled data-driven methods in practical SHM applications. These methods typically use ML algorithms to identify patterns within features extracted from data representing structural conditions, thereby inferring damage from changes in these patterns. However, data-driven models often struggle to generalize effectively to unseen datasets. This study addresses this challenge through three key contributions: dataset augmentation, an efficient feature representation, and a probabilistic modeling approach. First, a data augmentation method leveraging the symmetric properties of bridge structures is introduced to enhance dataset diversity. Second, a novel damage indicator named Fre-GraRMSC1 is proposed, capable of distinguishing both damage locations and severity. Finally, a probabilistic generative model based on a deep belief network (DBN) is developed to predict damage locations and degrees. The proposed methods are validated using vibration data from a numerical three-span continuous bridge subjected to random vehicle excitations. Results demonstrate high accuracy in damage identification and improved generalization performance. Full article

The growing digitalization of Industrial Control Systems (ICSs) presents both significant benefits and security challenges, especially for small and medium-sized factories with limited resources. Effective anomaly detection is essential to safeguard these facilities and prevent costly disruptions. Although current research has advanced anomaly detection, it is still challenging for algorithms to be capable of effectively balancing the interplay between training speed, computational cost, and accuracy while simultaneously exhibiting robust stability and adaptability. This gap often leaves small and medium-sized factories without efficient solutions. To address these issues, this work introduces a deep belief network-based boosting adversarial autoencoder termed DBN-BAAE, a novel lightweight anomaly detection mechanism based on boosting adversarial learning. The proposed lightweight mechanism saves computational overhead, enhances autoencoder training stability with an improved deep belief network (DBN) for pre-training, boosts encoder expression through ensemble learning, achieves high detection accuracy via an adversarial decoder, and employs a dynamic threshold to enhance adaptability and reduce the need for retraining. Experiments reveal that the mechanism not only achieves an F1 score of 0.82, surpassing the best baseline by 1%, but also accelerates training speed by 2.2 times, demonstrating its effectiveness and efficiency in ICS environments, particularly for small and medium-sized factories. Full article

Sea surface temperature (SST) is crucial for weather forecasting, climate modeling, and environmental monitoring. This study proposes a novel prediction model that achieves a 60-day forecast with a root mean square error (RMSE) consistently below 0.9 °C. The model combines the nonlinear feature extraction of a deep belief network (DBN) with the high-precision regression of support vector regression (SVR), enhanced by spatiotemporal secondary calibration (SSC) to better capture SST variation patterns. Using satellite-derived remote sensing data, the DBN-SVR model outperforms baseline methods in both the Indian Ocean and North Pacific regions, demonstrating strong applicability across diverse marine environments. This work advances long-term SST prediction capabilities, providing a reliable foundation for extended-range marine forecasts. Full article

Early detection of diabetic retinopathy is critical for preserving vision in diabetic patients. The classification of lesions in Retinal fundus images, particularly macular edema, is an essential diagnostic tool, yet it presents a significant learning curve for both novice and experienced ophthalmologists. To address this challenge, a novel Convolutional Deep Belief Network (CDBN) is proposed to classify image patches into three distinct categories: two types of macular edema—microhemorrhages and hard exudates—and a healthy category. The method leverages high-level feature extraction to mitigate issues arising from the high similarity of low-level features in noisy images. Additionally, a Real-Coded Genetic Algorithm optimizes the parameters of Gabor filters and the network, ensuring optimal feature extraction and classification performance. Experimental results demonstrate that the proposed CDBN outperforms comparative models, achieving an F1 score of 0.9258. These results indicate that the architecture effectively overcomes the challenges of lesion classification in retinal images, offering a robust tool for clinical application and paving the way for advanced clinical decision support systems in diabetic retinopathy management. Full article

In the context of global warming, renewable energy sources, particularly wind and solar power, have garnered increasing attention in recent decades. Accurate forecasting of the energy output in microgrids (MGs) is essential for optimizing energy management, reducing maintenance costs, and prolonging the lifespan of energy storage systems. This study proposes an innovative approach to solar irradiance forecasting based on the theory of belief functions, introducing a novel and flexible evidential method for short-to-medium-term predictions. The proposed machine learning model is designed to effectively handle missing data and make optimal use of available information. By integrating multiple predictive models, each focusing on different meteorological factors, the approach enhances forecasting accuracy. The Yager combination method and pignistic transformation are utilized to aggregate the individual models. Applied to a publicly available dataset, the method achieved promising results, with an average root mean square error (RMS) of 27.83 W/m² calculated from eight distinct forecast days. This performance surpasses the best reported results of 30.21 W/m² from recent comparable studies for one-day-ahead solar irradiance forecasting. Comparisons with deep learning-based methods, such as long short-term memory (LSTM) networks and recurrent neural networks (RNNs), demonstrate that the proposed approach is competitive with state-of-the-art techniques, delivering reliable predictions with significantly less training data. The full potential and limitations of the proposed approach are also discussed. Full article

Railway traction motor bearings (RTMB) are critical components in high-speed trains (HST) that are particularly susceptible to failure due to the high stress and rotational frequency they experience. To address the challenge of high false-positive rates in existing monitoring systems, this paper introduces a novel sensorless monitoring scheme that leverages stator current to detect fault-related characteristics, eliminating the need for additional sensors. This approach employs a hybrid signal preprocessing algorithm that integrates adaptive notch filtering (ANF) with envelope spectrum analysis (ESA) to effectively sparse the stator current and extract relevant fault features. A deep belief network (DBN) is utilized for the classification of the health status of the RTMB. To validate the scheme’s feasibility and effectiveness, we conducted experiments on a 1:1 scale high-speed railway traction motor, demonstrating that mechanical defects in RTMB can be reliably indicated by changes in stator current. Based on the analysis of experimental results, it was concluded that the fault detection accuracy of RTMB based on stator current is at least 17.3% higher than that of the fault diagnosis methods based on vibration in diagnosing whether the system has a fault. Among them, the method proposed in this paper is the best in diagnosing the presence and type of faults, with an accuracy that is at least 8.9% higher than other methods. This study not only presents a new method for RTMB monitoring but also contributes to the field by offering a more accurate and efficient alternative to current practices. Full article

The precise detection of liquid flow and viscosity is a crucial challenge in industrial processes and environmental monitoring due to the variety of liquid samples and the complex reflective properties of energetic liquids. Traditional methods often struggle to maintain accuracy under such conditions. This study addresses the complexity arising from sample diversity and the reflective properties of energetic liquids by introducing a novel model based on computer vision and deep learning. We propose the DBN-AGS-FLSS, an integrated deep learning model for high-precision, real-time liquid surface pointer detection. The model combines Deep Belief Networks (DBN), Feedback Least-Squares SVM classifiers (FLSS), and Adaptive Genetic Selectors (AGS). Enhanced by bilateral filtering and adaptive contrast enhancement algorithms, the model significantly improves image clarity and detection accuracy. The use of a feedback mechanism for reverse judgment dynamically optimizes model parameters, enhancing system accuracy and robustness. The model achieved an accuracy, precision, F1 score, and recall of 99.37%, 99.36%, 99.16%, and 99.36%, respectively, with an inference speed of only 1.5 ms/frame. Experimental results demonstrate the model’s superior performance across various complex detection scenarios, validating its practicality and reliability. This study opens new avenues for industrial applications, especially in real-time monitoring and automated systems, and provides valuable reference for future advancements in computer vision-based detection technologies. Full article

Universal circuit breakers are crucial devices in power systems, and the accuracy of their fault diagnosis is vital. However, existing diagnostic models suffer from low feature extraction rates and low diagnostic accuracy. In this paper, we propose a novel approach for fault diagnosis of universal circuit breakers based on analyzing vibration signals generated during the closing operation. Firstly, the vibration signal was decomposed into multiple modal components using Variable Mode Decomposition (VMD), and the modal components were subjected to time and frequency domain feature extraction. Then, the extracted features were fused and normalized to construct a training dataset for the proposed model. We propose a Deep Belief Network (DBN) diagnostic model based on the Whale Optimization Algorithm (WOA), where the WOA is employed to optimize the hyperparameters of the DBN. Experimental results demonstrate that the proposed VMD and WOA-DBN model achieved an average accuracy of 96.63%. This method enhanced the accuracy of feature extraction from vibration signals and outperformed traditional diagnostic models when using a single vibration signal for fault diagnosis of universal circuit breakers. It provides a novel solution for early fault diagnosis of universal circuit breakers. Full article

Thermoacoustic oscillation is indeed a phenomenon characterized by the symmetric coupling of thermal and acoustic waves. This paper introduces a novel approach for monitoring and predicting thermoacoustic combustion instability using a combination of recurrence quantification analysis (RQA) and an optimized deep belief network (DBN). Six samples of combustion state data were collected using two distinct types of burners to facilitate the training and validation of GA-DBN. The proposed methodology leverages RQA to extract intricate patterns and dynamic features from time series data representing combustion behavior. By quantifying the recurrence plot of specific patterns, the analysis provides valuable insights into the underlying thermoacoustic dynamics. Among three different feature extraction methods, RQA stands out remarkably in performance. These RQA-derived features serve as input to a carefully tuned DBN, which is trained to learn the complex relationships within the combustion process. The classification accuracy of deep belief network optimized by genetic algorithm (GA-DBN) reached an impressive 99.8%. Subsequent multiple comparisons were conducted between GA-DBN, DBN, and support vector machine (SVM), revealing that GA-DBN consistently demonstrated satisfactory classification results. This method holds significant importance in monitoring intricate combustion states. Full article

Cloud computing (CC) is an internet-enabled environment that provides computing services such as networking, databases, and servers to clients and organizations in a cost-effective manner. Despite the benefits rendered by CC, its security remains a prominent concern to overcome. An intrusion detection system (IDS) is generally used to detect both normal and anomalous behavior in networks. The design of IDS using a machine learning (ML) technique comprises a series of methods that can learn patterns from data and forecast the outcomes consequently. In this background, the current study designs a novel multi-objective seagull optimization algorithm with a deep learning-enabled vulnerability detection (MOSOA-DLVD) technique to secure the cloud platform. The MOSOA-DLVD technique uses the feature selection (FS) method and hyperparameter tuning strategy to identify the presence of vulnerabilities or attacks in the cloud infrastructure. Primarily, the FS method is implemented using the MOSOA technique. Furthermore, the MOSOA-DLVD technique uses a deep belief network (DBN) method for intrusion detection and its classification. In order to improve the detection outcomes of the DBN algorithm, the sooty tern optimization algorithm (STOA) is applied for the hyperparameter tuning process. The performance of the proposed MOSOA-DLVD system was validated with extensive simulations upon a benchmark IDS dataset. The improved intrusion detection results of the MOSOA-DLVD approach with a maximum accuracy of 99.34% establish the proficiency of the model compared with recent methods. Full article

Breast cancer (BC) has affected many women around the world. To accomplish the classification and detection of BC, several computer-aided diagnosis (CAD) systems have been introduced for the analysis of mammogram images. This is because analysis by the human radiologist is a complex and time-consuming task. Although CAD systems are used to primarily analyze the disease and offer the best therapy, it is still essential to enhance present CAD systems by integrating novel approaches and technologies in order to provide explicit performances. Presently, deep learning (DL) systems are outperforming promising outcomes in the early detection of BC by creating CAD systems executing convolutional neural networks (CNNs). This article presents an Intelligent Breast Mass Classification Approach using the Archimedes Optimization Algorithm with Deep Learning (BMCA-AOADL) technique on Digital Mammograms. The major aim of the BMCA-AOADL technique is to exploit the DL model with a bio-inspired algorithm for breast mass classification. In the BMCA-AOADL approach, median filtering (MF)-based noise removal and U-Net segmentation take place as a pre-processing step. For feature extraction, the BMCA-AOADL technique utilizes the SqueezeNet model with AOA as a hyperparameter tuning approach. To detect and classify the breast mass, the BMCA-AOADL technique applies a deep belief network (DBN) approach. The simulation value of the BMCA-AOADL system has been studied on the MIAS dataset from the Kaggle repository. The experimental values showcase the significant outcomes of the BMCA-AOADL technique compared to other DL algorithms with a maximum accuracy of 96.48%. Full article