Search Results (39)

Earthquakes remain among the most destructive natural hazards, necessitating reliable monitoring and early warning systems for effective risk mitigation. Recent advances in machine learning (ML) and deep learning (DL) have significantly improved seismic signal analysis, enabling more accurate event detection, phase picking, classification, and magnitude estimation. This study presents a systematic review of ML- and DL-based approaches for earthquake monitoring, with particular emphasis on Distributed Acoustic Sensing (DAS) as an emerging technology for high-resolution, real-time seismic observation. Following the PRISMA 2020 guidelines, a systematic literature search was conducted across Scopus, Web of Science, IEEE Xplore, and Google Scholar, yielding 252,223 initial records. After applying the predefined publication period, removing duplicate records, conducting relevance screening, and performing eligibility assessment, 138 peer-reviewed studies published between 2021 and 2025 were retained for detailed analysis and synthesis. The review reveals a significant transition from conventional signal-processing techniques to advanced artificial intelligence-based approaches, including Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, Bidirectional Long Short-Term Memory (BiLSTM) networks, Transformer-based architectures, hybrid models, and Bayesian learning methods for uncertainty quantification. The findings further demonstrate that Distributed Acoustic Sensing (DAS) has emerged as a transformative sensing technology because of its dense spatial coverage, high spatial resolution, and continuous monitoring capability. However, several challenges remain, including the lack of standardized datasets, limited model generalization across diverse geological settings, insufficient model interpretability, high computational complexity, and the limited integration of uncertainty-aware approaches for real-time seismic monitoring. This review identifies these critical research gaps and highlights promising future research directions, including multimodal data fusion, interpretable artificial intelligence, physics-informed learning, self-supervised learning, and robust uncertainty quantification for next-generation intelligent seismic monitoring systems. Unlike previous review studies that primarily focus on individual machine learning techniques or conventional seismic monitoring, this review provides a comprehensive and systematic synthesis of recent advances in machine learning, deep learning, and Distributed Acoustic Sensing (DAS), identifies current research gaps, and offers practical recommendations to guide future research on intelligent earthquake monitoring systems. Full article

The development of drug delivery systems (DDSs) is moving from empirical, trial-and-error formulation toward data-driven, artificial intelligence (AI)-guided, and automated workflows. Microfluidics provides precise control of microscale fluids, enabling high-throughput production of relatively homogeneous drug carriers such as liposomes, lipid nanoparticles (LNPs), and polymeric micelles. However, the high-dimensional parameter space of microfluidic reactors often exceeds the capacity of manual optimization. This review examines the integration of AI, particularly machine learning (ML), deep learning (DL), and Bayesian optimization, into microfluidic platforms for accelerating DDS design, optimizing critical quality attributes (CQAs), and supporting self-driving laboratory workflows. We discuss the technical foundations of AI-enabled microfluidics, applications in nanocarrier synthesis and phenotypic screening, and the evolving regulatory landscape for AI-assisted pharmaceutical development. Full article

This work investigates the deployment of Deep Learning (DL) models for network intrusion detection on resource-constrained IoT devices, using the public CICIoT2023 dataset. In particular, we consider the following DL models: Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), Temporal Convolutional Network (TCN), Multi-Layer Perceptron (MLP). Bayesian optimization is employed to fine-tune the models’ hyperparameters and ensure reliable performance evaluation across both binary (2-class) and multi-class (8-class, 34-class) intrusion detection. Then, the computational complexity of each DL model is analyzed—in terms of the number of Multiply–ACCumulate operations (MACCs), RAM usage, and inference time—through the STMicroelectronics Cube.AI Analyzer tool, with models being deployed on an STM32H7S78-DK board. To assess the practical deployability of the considered DL models, a trade-off score (balancing classification accuracy and computational efficiency) is introduced: according to this score, our experimental results indicate that MLP and TCN outperform the other models. Furthermore, Post-Training Quantization (PTQ) to 8-bit integer precision is applied, allowing the model size to be reduced by more than $90\%$ with negligible performance degradation. This demonstrates the effectiveness of quantization in optimizing DL models for real-world deployment on resource-constrained IoT devices. Full article

Current functional safety mechanisms mainly control the access points and perimeters of manufacturing cells without guaranteeing the integrity of their internal components or the absence of unauthorized humans or objects. In this work, we present a novel deep learning (DL)-based safety system that enhances the safety circuit designed according to functional safety principles, detecting, with great reliability, the presence of persons within the cell and, with high precision, anomalous elements of any kind. Our approach follows a two-stage DL methodology that combines contrastive learning with Bayesian clustering. First, a supervised contrastive scheme learns the characteristics of safe scenarios and distinguishes them from unsafe ones caused by workers remaining inside the cell. Next, a Bayesian mixture models the latent space of safe scenarios, quantifying deviations and enabling the detection of previously unseen anomalous objects without any specific fine-tuning. To further improve robustness, we introduce an ensemble-based hybrid latent-space methodology that maximizes performance regardless of the underlying encoders’ characteristics. The experiments are conducted on a real dataset captured in a belt-picking cell in production. The proposed system achieves 100% accuracy in distinguishing safe scenarios from those with the presence of workers, even in partially occluded cases, and an average area-under-the-curve of 0.9984 across seven types of anomalous objects commonly found in manufacturing environments. Finally, for interpretability analysis, we design a patch-based feature-ablation framework that demonstrates the model’s reliability under uncertainty and the absence of learning biases. The proposed technique enables the deployment of an innovative high-performance safety system that, to our knowledge, does not exist in the industry. Full article

Accurate stock price forecasting is crucial for supporting informed investment decisions, effective risk management, and the identification of profitable market opportunities. Financial time series present considerable challenges for prediction due to their complex, nonlinear dynamics and sensitivity to a wide range of economic factors. Although various statistical methods have been developed to model the multidimensional relationships inherent in such datasets, advancements in big data technologies have greatly facilitated the recording, analysis, and interpretation of large-scale financial data, thereby accelerating the adoption of deep learning (DL) algorithms in this domain. In the present study, RNN-, LSTM-, and GRU-based models were developed to forecast the closing prices of two airline stocks, with hyperparameter optimization conducted via the Bayesian optimization algorithm. The dataset consisted of daily closing prices of THYAO and PGSUS stocks obtained from Yahoo Finance. Comparative analysis demonstrated that the GRU model yielded the highest accuracy for THYAO stock price prediction, achieving a MAPE of 3.05% and an RMSE of 3.195, whereas for PGSUS, the model achieved a MAPE of 3.97% and an RMSE of 3.232. Beyond its empirical contribution, this study also emphasizes the conceptual relevance of symmetry in financial forecasting. The proposed deep learning framework captures the balanced relationships and nonlinear interactions inherent in stock market behavior, reflecting both symmetry and asymmetry in market responses to economic factors. Full article

The prediction of Remaining Useful Life (RUL) constitutes a vital aspect of Prognostics and Health Management (PHM), providing capabilities for the assessment of mechanical component health status and prediction of failure instances. Recent studies on feature extraction, time-series modeling, and multi-task learning have shown remarkable advancements. However, most deep learning (DL) techniques predominantly focus on unimodal data or static feature extraction techniques, resulting in a lack of RUL prediction methods that can effectively capture the individual differences among heterogeneous sensors and failure modes under complex operational conditions. To overcome these limitations, an adaptive RUL prediction framework named ADAPT-RULNet is proposed for mechanical components, integrating the feature extraction capabilities of attention-enhanced deep learning (DL) and the decision-making abilities of deep reinforcement learning (DRL) to achieve end-to-end optimization from raw data to accurate RUL prediction. Initially, Functional Alignment Resampling (FAR) is employed to generate high-quality functional signals; then, attention-enhanced Dynamic Time Warping (DTW) is leveraged to obtain individual degradation stages. Subsequently, an attention-enhanced of hybrid multi-scale RUL prediction network is constructed to extract both local and global features from multi-format data. Furthermore, the network achieves optimal feature representation by adaptively fusing multi-source features through Bayesian methods. Finally, we innovatively introduce a Deep Deterministic Policy Gradient (DDPG) strategy from DRL to adaptively optimize key parameters in the construction of individual degradation stages and achieve a global balance between model complexity and prediction accuracy. The proposed model was evaluated on aircraft engines and railway freight car wheels. The results indicate that it achieves a lower average Root Mean Square Error (RMSE) and higher accuracy in comparison with current approaches. Moreover, the method shows strong potential for improving prediction accuracy and robustness in varied industrial applications. Full article

Monitoring Peripheral Oxygen Saturation (SpO₂) is an important vital sign both in Intensive Care Units (ICUs), during surgery and convalescence, and as part of remote medical consultations after of the COVID-19 pandemic. This has made the development of new SpO₂-measurement tools an area of active research and opportunity. In this paper, we present a new Deep Learning (DL) combined strategy to estimate SpO₂ without contact, using pre-magnified facial videos to reveal subtle color changes related to blood flow and with no calibration per subject required. We applied the Eulerian Video Magnification technique using the Hermite Transform (EVM-HT) as a feature detector to feed a Three-Dimensional Convolutional Neural Network (3D-CNN). Additionally, parameters and hyperparameter Bayesian optimization and an ensemble technique over the dataset magnified were applied. We tested the method on 18 healthy subjects, where facial videos of the subjects, including the automatic detection of the reference from a contact pulse oximeter device, were acquired. As performance metrics for the SpO₂-estimation proposal, we calculated the Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and other parameters from the Bland–Altman (BA) analysis with respect to the reference. Therefore, a significant improvement was observed by adding the ensemble technique with respect to the only optimization, obtaining 14.32% in RMSE (reduction from 0.6204 to 0.5315) and 13.23% in MAE (reduction from 0.4323 to 0.3751). On the other hand, regarding Bland–Altman analysis, the upper and lower limits of agreement for the Mean of Differences (MOD) between the estimation and the ground truth were 1.04 and −1.05, with an MOD (bias) of −0.00175; therefore, MOD $\pm 1.96\sigma$ = −0.00175 ± 1.04. Thus, by leveraging Bayesian optimization for hyperparameter tuning and integrating a Bagging Ensemble, we achieved a significant reduction in the training error (bias), achieving a better generalization over the test set, and reducing the variance in comparison with the baseline model for SpO₂ estimation. Full article

Lung-cancer incidence is projected to rise by 50% by 2035, underscoring the need for accurate yet accessible risk-stratification tools. We trained a Bayesian neural network on 300 annotated chest-CT scans from the public LIDC–IDRI cohort, integrating clinical metadata. Hamiltonian Monte-Carlo sampling (10 000 posterior draws) captured parameter uncertainty; performance was assessed with stratified five-fold cross-validation and on three independent multi-centre cohorts. On the locked internal test set, the model achieved 99.0% accuracy, AUC = 0.990 and macro-F1 = 0.987. External validation across 824 scans yielded a mean AUC of 0.933 and an expected calibration error $<0.034$, while eliminating false positives for benign nodules and providing voxel-level uncertainty maps. Uncertainty-aware Bayesian deep learning delivers state-of-the-art, well-calibrated lung-cancer risk predictions from a single CT scan, supporting personalised screening intervals and safe deployment in clinical workflows. Full article

Background: Septic shock is a heterogeneous syndrome with diverse clinical presentations and pathophysiology, yet current management guidelines largely treat it as a homogenous entity. Early risk stratification relies on lactate and different predictive scoring systems, which may not capture the underlying heterogeneity in host responses. Aim: To identify discrete subphenotypes of septic shock using admission demographics and laboratory parameters, and to evaluate their relationship with in-hospital outcomes. Methods: We conducted a retrospective multicenter cohort study of 10,462 adult patients with ICD-10-defined septic shock admitted to intensive care units between 2014 and 2015. We used Two-Step Cluster Analysis using log-likelihood distance and the Bayesian Information Criterion to identify two distinct phenotypes. We compared clusters on baseline characteristics, in-hospital outcomes including mortality, days on mechanical ventilation, vasopressor use, acute kidney injury (AKI), AKI requiring renal replacement therapy (RRT), and ICU and hospital lengths of stay. Results: We identified two clusters (Cluster 1, n = 5355 and Cluster 2, n = 5107) in our study. Cluster 1 showed greater biochemical severity at presentation, including higher median lactate (2.40 vs. 2.20 mmol L⁻¹; p < 0.001), serum creatinine (1.39 vs. 1.20 mg dL⁻¹; p < 0.001), blood urea nitrogen (28 vs. 25 mg dL⁻¹; p < 0.001), and neutrophil-to-lymphocyte ratio (11.12 vs. 10.38; p < 0.001), and a higher mean SOFA score (7.05 ± 3.85 vs. 6.76 ± 3.87; p < 0.001). Despite this, Cluster 1 required mechanical ventilation more frequently (46.1% vs. 42.2%; p < 0.001) and had a higher incidence of AKI (58.1% vs. 55.6%; p = 0.009), including more stage 3 AKI (17.2% vs. 15.2%; p < 0.001) and dialysis (6.6% vs. 5.2%; p = 0.005), yet experienced similar in-hospital mortality (15.4% vs. 15.8%; p = 0.615) and comparable ICU (2.18 vs. 2.26 days; p = 0.254) and hospital lengths of stay (6.63 vs. 6.80 days; p = 0.251). Conclusions: Two septic shock phenotypes were identified, one with marked early organ dysfunction (Cluster 1) and another with milder initial derangements (Cluster 2), yet both showed convergent short-term mortality and lengths of stay despite divergent support needs. These results challenge reliance on single-parameter severity markers and underscore the need for phenotype-guided risk stratification and personalized management strategies in septic shock. Full article

In 5G wireless communication, network slicing is considered one of the key network elements, which aims to provide services with high availability, low latency, maximizing data throughput, and ultra-reliability and save network resources. Due to the exponential expansion of cellular networking in the number of users along with the new applications, delivering the desired Quality of Service (QoS) requires an accurate and fast network slicing mechanism. In this paper, hybrid deep learning (DL) approaches are investigated using convolutional neural networks (CNNs), Long Short-Term Memory (LSTM), recurrent neural networks (RNNs), and Gated Recurrent Units (GRUs) to provide an accurate network slicing model. The proposed hybrid approaches are CNN-LSTM, CNN-RNN, and CNN-GRU, where a CNN is initially used for effective feature extraction and then LSTM, an RNN, and GRUs are utilized to achieve an accurate network slice classification. To optimize the model performance in terms of accuracy and model complexity, the hyperparameters of each algorithm are selected using the Bayesian optimization algorithm. The obtained results illustrate that the optimized hybrid CNN-GRU algorithm provides the best performance in terms of slicing accuracy (99.31%) and low model complexity. Full article

Estimating the downlink (DL) channel covariance matrix (CCM) is crucial for beamforming and capacity optimization in massive MIMO frequency division duplexing (FDD) systems, yet it poses significant challenges due to the lack of direct channel reciprocity. To address this issue, a turbo channel covariance conversion (Turbo-CCC) algorithm is proposed to enhance estimation accuracy and robustness by utilizing the angular power spectrum (APS) reciprocity. Specifically, based on the electromagnetic wave propagation characteristics, we model the APS as multikernel functions. On this basis, we then develop the Turbo-CCC algorithm by integrating the orthogonal approximate message passing (OAMP) algorithm and the multikernel adaptive filtering (MKAF) algorithm based on a Bayesian framework. The OAMP module estimates the APS from the uplink (UL) CCM regardless of its structural characteristics, whereas the MKAF module refines the APS estimation by leveraging its structural characteristics. These two modules operate iteratively, progressively improving the accuracy of the DL CCM estimation. Simulation results demonstrate that the proposed algorithm noticeably enhances the estimation performance and exhibits strong adaptability to diverse APS distributions and propagation environments, offering a novel approach for the DL CCM estimation in massive MIMO FDD systems. Full article

This study explores the application of time series, machine learning (ML), and deep learning (DL) models to predict the prices and performance of covered call ETFs. Utilizing historical data from major covered call ETFs like QYLD, XYLD, JEPI, JEPQ, and RYLD, the research assesses the predictive accuracy and reliability of different forecasting approaches. It compares traditional time series methods, including ARIMA and Heterogeneous Autoregressive Model (HAR), with advanced ML techniques such as Random Forests (RF) and Support Vector Regression (SVR), as well as DL models like Recurrent Neural Networks (RNN) and Convolutional Neural Networks (CNN). Model performance is evaluated using metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC). Results indicate that the DL models are effective at identifying the nonlinear patterns and temporal dependencies in the price movements of covered call ETFs, outperforming both traditional time series and ML techniques. These findings enhance the existing financial forecasting literature and offer valuable insights for investors and portfolio managers aiming to improve their strategies using covered call ETFs. Full article

To successfully meet the various requirements of modern storage systems, NAND flash memory should be highly optimized by precisely tuning a huge number of internal operating parameters. Although 3D NAND flash memory succeeds in increasing the capacity of storage systems, its complex architecture and unique error behavior make such optimization a more difficult and time-consuming process during NAND manufacturing. In this paper, we introduce HAIPO, a novel methodology for post-fabrication optimization of NAND flash memory, which is an essential step in the manufacturing process of modern 3D NAND flash memory to simultaneously meet various requirements on reliability, performance, yield, etc. HAIPO is based on simple machine-learning approaches that consist of (i) a lightweight deep-learning (DL) model to generate initial device parameters and (ii) an evolutionary algorithm (EA) to explore device parameters automatically. To more effectively explore device parameters, we introduce three key guidelines for each generation in the EA: (1) domain-specific rules, (2) recent optimization results, and (3) online Bayesian simulation, respectively, to enable quick optimization for a huge number of device parameters within the limited product turnaround time (TAT). In addition, we integrate two optimization modules with HAIPO to improve optimization efficiency even in environments with severe process variation. We demonstrate the feasibility and effectiveness of HAIPO using real 320 3D TLC/QLC NAND flash chips, showing significant performance and reliability improvements by up to 8.8% and 12% on average, respectively, within a quite limited optimization TAT. Full article

Accurately classifying road signs is crucial for autonomous driving due to the high stakes involved in ensuring safety and compliance. As Convolutional Neural Networks (CNNs) have largely replaced traditional Machine Learning models in this domain, the demand for substantial training data has increased. This study aims to compare the performance of classical Machine Learning (ML) models and Deep Learning (DL) models under varying amounts of training data, particularly focusing on altered signs to mimic real-world conditions. We evaluated three classical models: Support Vector Machine (SVM), Random Forest, and Linear Discriminant Analysis (LDA), and one Deep Learning model: Convolutional Neural Network (CNN). Using the German Traffic Sign Recognition Benchmark (GTSRB) dataset, which includes approximately 40,000 German traffic signs, we introduced digital alterations to simulate conditions such as environmental wear or vandalism. Additionally, the Histogram of Oriented Gradients (HOG) descriptor was used to assist classical models. Bayesian optimization and k-fold cross-validation were employed for model fine-tuning and performance assessment. Our findings reveal a threshold in training data beyond which accuracy plateaus. Classical models showed a linear performance decrease under increasing alteration, while CNNs, despite being more robust to alterations, did not significantly outperform classical models in overall accuracy. Ultimately, classical Machine Learning models demonstrated performance comparable to CNNs under certain conditions, suggesting that effective road sign classification can be achieved with less computationally intensive approaches. Full article

Melanoma is a highly prevalent and lethal form of skin cancer, which has a significant impact globally. The chances of recovery for melanoma patients substantially improve with early detection. Currently, deep learning (DL) methods are gaining popularity in assisting with the identification of diseases using medical imaging. The paper introduces a computational model for classifying melanoma skin cancer images using convolutional neural networks (CNNs) and vision transformers (ViT) with the HAM10000 dataset. Both approaches utilize mask-guided techniques, employing a specialized U2-Net segmentation module to generate masks. The CNN-based approach utilizes ResNet50, VGG16, and Xception with transfer learning. The training process is enhanced using a Bayesian hyperparameter tuner. Moreover, this study applies gradient-weighted class activation mapping (Grad-CAM) and Grad-CAM++ to generate heatmaps to explain the classification models. These visual heatmaps elucidate the contribution of each input region to the classification outcome. The CNN-based model approach achieved the highest accuracy at 98.37% in the Xception model with a sensitivity and specificity of 95.92% and 99.01%, respectively. The ViT-based model approach achieved high values for accuracy, sensitivity, and specificity, such as 92.79%, 91.09%, and 93.54%, respectively. Furthermore, the performance of the model was assessed through intersection over union (IOU) and other qualitative evaluations. Finally, we developed the proposed model as a web application that can be used as a support tool for medical practitioners in real-time. The system usability study score of 86.87% is reported, which shows the usefulness of the proposed solution. Full article