Search Results (1,698)

Background/Objectives: Tuberculosis remains a significant global health problem, particularly in resource-limited environments. Its mortality and spread can be considerably decreased by early and precise detection via chest X-ray imaging. This study introduces a novel approach based on hybrid deep learning for Tuberculosis detection from chest X-ray images. Methods: The introduced approach combines GhostNet, a lightweight convolutional neural network tuned for computational efficiency, and MobileViT, a transformer-based model that can capture both local spatial patterns and global contextual dependencies. Through such integration, the model attains a balanced trade-off between classification accuracy and computational efficiency. The architecture employs feature fusion, where spatial features from GhostNet and contextual representations from MobileViT are globally pooled and concatenated, which allows the model to learn discriminative and robust feature representations. Results: The suggested model was assessed on two publicly available chest X-ray datasets and contrasted against several cutting-edge convolutional neural network architectures. Findings showed that the introduced hybrid model surpasses individual baselines, attaining 99.52% accuracy on dataset 1 and 99.17% on dataset 2, while keeping low computational cost (7.73M parameters, 282.11M Floating Point Operations). Conclusions: These outcomes verify the efficacy of feature-level fusion between a convolutional neural network and transformer branches, allowing robust tuberculosis detection with low inference overhead. The model is ideal for clinical deployment and resource-constrained contexts due to its high accuracy and lightweight design. Full article

Within the framework of the ongoing development of application of Machine Learning models in Medicine and Physical Therapy, the development of accurate prognosis algorithms for postoperative patients during the rehabilitation phase remains an area requiring further refinement. This paper examines hybrid Deep Learning models that integrate Convolution Neural Networks, Long Short-Term Memory and Gated Recurrent Unit networks, as well as genetic algorithm optimization for feature selection for predicting the time needed for a patient to rehabilitate. Patient data included features like age, passive range of available movements (preoperative and postoperative) and total rehabilitation time. Genetic Algorithm optimization for feature selection indicated that 4 out of the 16 available features are adequate for predicting rehabilitation time. Hybrid Deep Learning models achieved a Root Mean Squared Error (RMSE) of 12 days (less than 0.4 months) in rehabilitation time prediction, demonstrating good performance on a relatively small dataset of 120 patients. Full article

The reliability of electric vehicle (EV) batteries requires detecting surface defects but also precisely locating them on the physical module for automated inspection, repair, and process optimisation. Conventional 2D computer vision methods, though accurate in image-space, do not provide traceable, real-world defect coordinates on complex or curved battery surfaces, limiting utility for digital twins, root cause analysis, and automated quality control. This work proposes a hybrid inspection pipeline that produces millimetre-level three-dimensional (3D) defect maps for EV battery modules. The approach integrates (i) calibrated dual-view multi-view geometry to project defect points onto the CAD geometry and triangulate them where dual-view coverage is available, (ii) single-image neural 3D shape inference calibrated to the module geometry to complement regions with limited multi-view coverage, and (iii) generative, physically informed augmentation of rare or complex defect types. Defects are first detected in 2D images using a convolutional neural network (CNN), then projected onto a dense 3D CAD model of each module, complemented by a single-image depth prediction in regions with limited dual-view coverage, yielding true as-built localisation on the battery’s surface. GenAI methods are employed to expand the dataset with synthetic defect variations. Synthetic, physically informed defect examples are incorporated during training to mitigate the scarcity of rare defect types. Evaluation on a pilot industrial dataset, with a physically measured reference subset, demonstrates that the hybrid 3D approach achieves millimetre-scale localisation accuracy and outperforms a per-view CNN baseline in both segmentation and 3D continuity. Full article

Because Internet of Things (IoT) networks are widely deployed, they have become attractive targets for botnet and distributed denial of service (DDoS) attacks, which require effective intrusion detection. Convolutional neural networks (CNNs) can achieve strong detection performance, but their many hyperparameters are usually tuned manually, which is costly and time-consuming. This paper proposes a new hybrid metaheuristic optimizer, FW-CNN, that combines Grey Wolf Optimization and Red Fox Optimization to automatically tune the key hyperparameters of a one-dimensional CNN for IoT intrusion detection. The Red Fox component enhances exploration and helps the search escape local optima, while the Grey Wolf component strengthens exploitation and guides convergence toward high-quality solutions. The proposed model is evaluated using the N-BaIoT dataset and compared with a feedforward neural network as well as a metaheuristic-optimized model based on the Adaptive Particle Swarm Optimization–Whale Optimization Algorithm-CNN. It achieves a final accuracy of 95.56%, improving on the feedforward network by 12.56 percentage points and outperforming the Adaptive Particle Swarm Optimization–Whale Optimization Algorithm-based CNN model by 1.02 percentage points. It also yields higher average precision, Kappa coefficient, and Jaccard similarity, and significantly reduces Hamming loss. These results indicate that the proposed hybrid optimizer is stable and effective for multi-class IoT intrusion detection in real environments. Full article

Tube hydroforming is a versatile forming process widely used in lightweight structural applications, where accurate characterization of the hoop mechanical behavior is crucial for reliable design and simulation. The ring hoop tensile test (RHTT) provides valuable experimental data for evaluating the elastoplastic response of anisotropic tubes in the hoop direction, but frictional effects often distort the measured force–displacement response. This study proposes a deep learning-based inverse identification framework to accurately recover the true hoop stress–strain behavior from RHTT data. Convolutional and recurrent neural network architectures, including CNN, long short term memory (LSTM), gated recurrent unit (GRU), bidirectional GRU (BiGRU), bidirectional LSTM (BiLSTM) and ConvLSTM, were trained using numerically generated datasets from finite element simulations. Data augmentation and hyperparameter tuning were applied to generalization. The hybrid ConvLSTM model achieved superior performance, with a minimum mean absolute error (MAE) of 0.08 and a coefficient of determination (R²) value of approximately 0.97, providing a close match to the Hill48 yield criterion. The proposed approach demonstrates the potential of deep neural networks as an efficient and accurate alternative to traditional inverse methods for characterizing anisotropic tubular materials. Full article

Upstream processes strongly influence downstream analysis in sequential data-processing workflows, particularly in machine learning, where data quality directly affects model performance. Conventional statistical imputations often fail to capture nonlinear dependencies, while deep learning approaches typically lack uncertainty quantification. We introduce a hybrid imputation model that integrates a deep learning autoencoder with Convolutional Neural Network (CNN) layers and a Transformer-based contextual modeling architecture to address systematic variation across heterogeneous data sources. Performing multiple imputations in the autoencoder–transformer latent space and averaging representations provides implicit batch correction that suppresses context-specific remains without explicit batch identifiers. We performed experiments on datasets in which 10% of missing data was artificially introduced by completely random missing data (MCAR) and non-random missing data (MNAR) mechanisms. They demonstrated practical performance, jointly ranking first among the imputation methods evaluated. This imputation technique reduced the root mean square error (RMSE) by 50% compared to denoising autoencoders (DAE) and by 46% compared to iterative imputation (MICE). Performance was comparable for adversarial models (GAIN) and attention-based models (MIDA), and both provided interpretable uncertainty estimates (CV = 0.08–0.15). Validation on datasets from multiple sources confirmed the robustness of the technique: notably, on a forensic dataset from multiple laboratories, our imputation technique achieved a practical improvement over GAIN (0.146 vs. 0.189 RMSE), highlighting its effectiveness in mitigating batch effects. Full article

The Internet of Things (IoT) has transformed industries, healthcare, and smart environments, but introduces severe security threats due to resource constraints, weak protocols, and heterogeneous infrastructures. Traditional Intrusion Detection Systems (IDS) fail to address critical challenges including scalability across billions of devices, interoperability among diverse protocols, real-time responsiveness under strict latency, data privacy in distributed edge networks, and high false positives in imbalanced traffic. This study provides a systematic comparative evaluation of three representative AI models, CNN-BiLSTM, Random Forest, and XGBoost for IoT intrusion detection on the NSL-KDD and UNSW-NB15 datasets. The analysis quantifies the achievable detection performance and inference latency of each approach, revealing a clear accuracy–latency trade-off that can guide practical model selection: CNN-BiLSTM offers the highest detection capability (F1 up to 0.986) at the cost of higher computational overhead, whereas XGBoost and Random Forest deliver competitive accuracy with significantly lower inference latency (sub-millisecond on conventional hardware). These empirical insights support informed deployment decisions in heterogeneous IoT environments where accuracy-critical gateways and latency-critical sensors coexist. Full article

Accurate sky–obstacle segmentation in hemispherical fisheye imagery is essential for solar irradiance forecasting, photovoltaic system design, and environmental monitoring. However, existing methods often rely on expensive all-sky imagers and region-specific training data, produce coarse sky–obstacle boundaries, and ignore the optical properties of fisheye lenses. We propose a low-cost segmentation framework designed for fisheye imagery that combines synthetic data generation, lens-aware augmentation, and a hybrid deep-learning pipeline. Synthetic fisheye training images are created from publicly available street-view panoramas to cover diverse environments without dedicated hardware, and lens-aware augmentations model fisheye projection and photometric effects to improve robustness across devices. On this dataset, we train a convolutional neural network (CNN) and refine its output with gradient-boosted decision trees (GBDT) to sharpen sky–obstacle boundaries. The method is evaluated on real fisheye images captured with smartphones and low-cost clip-on lenses across multiple sites, achieving an Intersection over Union (IoU) of 96.63% and an F1 score of 98.29%, along with high boundary accuracy. An additional evaluation on an external panoramic baseline dataset confirms strong cross-dataset generalization. Together, these results show that the proposed framework enables accurate, low-cost, and widely deployable hemispherical sky segmentation for practical solar and environmental imaging applications. Full article

The proliferation of Internet of Things (IoT) devices has amplified botnet risks, while traditional network-based intrusion detection systems (IDSs) struggle under encrypted and/or sparse traffic. Power consumption offers an effective side channel for device-level detection. Yet, prior studies typically focus on a single model family (often a convolutional neural network (CNN)) and rarely assess generalization across devices or compare broader model classes. In this paper, we conduct unified benchmarking and comparison of classical (SVM and RF), deep (CNN, LSTM, and 1D Transformer), and hybrid (CNN + LSTM, CNN + Transformer, and CNN + RF) models on the CHASE’19 dataset and a newly curated three-class botnet dataset, using consistent preprocessing and evaluation across single- and cross-device settings, reporting both accuracy and efficiency (latency and throughput). Experimental results demonstrate that Random Forest achieves the highest single-device accuracy (99.43% on the Voice Assistant with Seed 42), while CNN + Transformer shows a strong accuracy–efficiency trade-off in cross-device scenarios (94.02% accuracy on the combined dataset at ∼60,000 samples/s when using the best-performing Seed 42). These results offer practical guidance for selecting models under accuracy, latency, and throughput constraints and establish a reproducible baseline for power-side-channel IDSs. Full article

The exponential growth of candidate data from large-scale radio pulsar surveys has created a pressing need for efficient and accurate classification methods. This paper presents a novel hybrid pulsar candidate recognition algorithm that integrates diagnostic plot images and structured numerical features using a multi-scale DenseNet framework. The proposed model combines convolutional neural networks (CNNs) for extracting spatial patterns from pulsar diagnostic plots and feedforward neural networks (FNNs) for processing scalar features such as SNR, DM, and pulse width. By fusing these multimodal representations, the model achieves superior classification performance, particularly in class-imbalanced settings standard to pulsar survey data. Evaluated on a synthesized dataset constructed from FAST and HTRU survey characteristics, the model demonstrates robust performance, achieving an F1-score of 0.904 and AUC-ROC of 0.978. Extensive ablation and cross-validation analyses confirm the contribution of each data modality and the model’s generalizability. Furthermore, the system maintains low inference latency (4.2 ms per candidate) and a compact architecture (~2.3 million parameters), indicating potential for real-time deployment once validated on real observational datasets. The proposed approach offers a scalable and interpretable multimodal framework for automated pulsar classification and provides a foundation for future validation and potential integration into observatories such as FAST and the Square Kilometre Array (SKA). Full article

Deep learning methods based on convolutional neural networks (CNNs) and Mamba have advanced medical image segmentation, yet two challenges remain: (1) trade-off in feature extraction, where CNNs capture local details but miss global context, and Mamba captures global dependencies but overlooks fine structures, and (2) limited feature aggregation, as existing methods insufficiently integrate inter-layer common information and delta details, hindering robustness to subtle structures. To address these issues, we propose a hybrid cascade and dual-path adaptive aggregation network (HCDAA-Net). For feature extraction, we design a hybrid cascade structure (HCS) that alternately applies ResNet and Mamba modules, achieving a spatial balance between local detail preservation and global semantic modeling. We further employ a general channel-crossing attention mechanism to enhance feature expression, complementing this spatial modeling and accelerating convergence. For feature aggregation, we first propose correlation-aware aggregation (CAA) to model correlations among features of the same lesions or anatomical structures. Second, we develop a dual-path adaptive feature aggregation (DAFA) module: the common path captures stable cross-layer semantics and suppresses redundancy, while the delta path emphasizes subtle differences to strengthen the model’s sensitivity to fine details. Finally, we introduce a residual-gated visual state space module (RG-VSS), which dynamically modulates information flow via a convolution-enhanced residual gating mechanism to refine fused representations. Experiments on diverse datasets demonstrate that our HCDAA-Net outperforms some state-of-the-art (SOTA) approaches. Full article

Milk adulteration through centrifugation, which artificially reduces the somatic cell count (SCC), represents a significant challenge to food authenticity and public health. This fraudulent practice alters the native molecular architecture of milk, masking inflammatory conditions such as subclinical mastitis and distorting product quality. Conventional analytical and microscopic techniques remain insufficiently sensitive to detect the subtle physicochemical changes associated with centrifugation, highlighting the need for molecular-level, data-driven diagnostics. The dataset included 128 paired raw milk samples and approximately 25,000 bright-field micrographs acquired across multiple microscopes, of which 95% were confirmed to be of high quality. In this study, advanced machine learning (ML) and deep learning (DL) approaches were applied to identify centrifugation-induced alterations in raw milk microstructure. Bright-field micrographs (pixel size 0.27 µm) of paired unprocessed and centrifuged samples were obtained under standardized optical conditions and analyzed using convolutional neural networks (ResNet-18/50, Inception-v3, Xception, NasNet-Mobile) and hybrid attention architectures (MaxViT, CoAtNet). Model performance was evaluated using the harmonic average of recalls across five micrographs per sample (HAR₅). Human microscopy experts (n = 4) achieved only 18% classification accuracy—below the random baseline (25%)—confirming that centrifugation-induced modifications are not visually discernible. In contrast, DL architectures reached up to 97% accuracy (HAR₅, Xception), successfully identifying subtle molecular cues. Class activation and sensitivity analyses indicated that models focused not on milk fat globule (MFG) boundaries but on high-frequency nanoscale variations related to the reorganization of casein micelles and solid non-fat fractions. The findings strongly suggest that centrifugation adulteration constitutes a molecular reorganization event rather than a morphological alteration. The integration of optical microscopy with AI-driven molecular analytics establishes deep learning as a precise and objective tool for detecting fraudulent milk processing and improving food integrity diagnostics. Full article

Unobserved fruit crop illnesses are a major threat to agricultural productivity worldwide and frequently cause farmers to suffer large financial losses. Manual field inspection-based disease detection techniques are time-consuming, unreliable, and unsuitable for extensive monitoring. Deep learning approaches, in particular convolutional neural networks, have shown promise for automated plant disease identification, although they still face significant obstacles. These include poor generalization across complicated visual backdrops, limited resilience to different illness sizes, and high processing needs that make deployment on resource-constrained edge devices difficult. We suggest a Hybrid Multi-Scale Neural Network (HMCT-AF with GSAF) architecture for precise and effective fruit crop disease identification in order to overcome these drawbacks. In order to extract long-range dependencies, HMCT-AF with GSAF combines a Vision Transformer-based structural branch with multi-scale convolutional branches to capture both high-level contextual patterns and fine-grained local information. These disparate features are adaptively combined using a novel HMCT-AF with a GSAF module, which enhances model interpretability and classification performance. We conduct evaluations on both PlantVillage (controlled environment) and CLD (real-world in-field conditions), observing consistent performance gains that indicate strong resilience to natural lighting variations and background complexity. With an accuracy of up to 93.79%, HMCT-AF with GSAF outperforms vanilla Transformer models, EfficientNet, and traditional CNNs. These findings demonstrate how well the model captures scale-variant disease symptoms and how it may be used in real-time agricultural applications using hardware that is compatible with the edge. According to our research, HMCT-AF with GSAF presents a viable basis for intelligent, scalable plant disease monitoring systems in contemporary precision farming. Full article

Fractional-order models (FOMs) have been recognized as superior tools for capturing the complex electrochemical dynamics of lithium-ion batteries, outperforming integer-order models in accuracy, robustness, and adaptability. Parameter identification (PI) is essential for FOMs, as its accuracy directly affects the model’s ability to predict battery behavior and estimate critical states such as state of charge (SOC) and state of health (SOH). In this study, a hybrid deep learning approach has been introduced for FOM PI, representing the first application of deep learning in this domain. A simulation platform was developed to generate datasets using Sobol and Monte Carlo sampling methods. Five deep learning models were constructed: long short-term memory (LSTM), gated recurrent unit (GRU), one-dimensional convolutional neural network (1DCNN), and hybrid models combining 1DCNN with LSTM and GRU. Hyperparameters were optimized using Optuna, and enhancements such as Huber loss for robustness to outliers, stochastic weight averaging, and exponential moving average for training stability were incorporated. The primary contribution lies in the hybrid architectures, which integrate convolutional feature extraction with recurrent temporal modeling, outperforming standalone models. On a test set of 1000 samples, the improved 1DCNN + GRU model achieved an overall root mean square error (RMSE) of 0.2223 and a mean absolute percentage error (MAPE) of 0.27%, representing nearly a 50% reduction in RMSE compared to its baseline. This performance surpasses that of the improved LSTM model, which yielded a MAPE of 9.50%, as evidenced by tighter scatter plot alignments along the diagonal and reduced error dispersion in box plots. Terminal voltage prediction was validated with an average RMSE of 0.002059 and mean absolute error (MAE) of 0.001387, demonstrating high-fidelity dynamic reconstruction. By advancing data-driven PI, this framework is well-positioned to enable real-time integration into battery management systems. Full article

Accurate short- and medium-term forecasting of photovoltaic (PV) power generation is vital for grid stability and renewable energy integration. This study presents a comparative scenario-based approach using Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and Gated Recurrent Unit (GRU) models trained with one year of real-time meteorological and production data from a 250 kWp grid-connected PV system located at Dicle University in Diyarbakır, Southeastern Anatolia, Turkey. The dataset includes hourly measurements of solar irradiance (average annual GHI 5.4 kWh/m²/day), ambient temperature, humidity, and wind speed, with missing data below 2% after preprocessing. Six forecasting scenarios were designed for different horizons (6 h to 1 month). Results indicate that the LSTM model achieved the best performance in short-term scenarios, reaching R² values above 0.90 and lower MAE and RMSE compared to CNN and GRU. The GRU model showed similar accuracy with faster training time, while CNN produced higher errors due to the dominant temporal nature of PV output. These results align with recent studies that emphasize selecting suitable deep learning architectures for time-series energy forecasting. This work highlights the benefit of integrating real local meteorological data with deep learning models in a scenario-based design and provides practical insights for regional grid operators and energy planners to reduce production uncertainty. Future studies can improve forecast reliability by testing hybrid models and implementing real-time adaptive training strategies to better handle extreme weather fluctuations. Full article