Search Results (7,496)

Natural disasters demand swift and accurate impact assessment, yet traditional field-based methods remain prohibitively slow. While semi-automatic techniques leveraging remote sensing and drone imagery have accelerated evaluations, existing datasets predominantly emphasize Western infrastructure, offering limited representation of African contexts. The EDDA dataset (a Mozambique post-disaster building damage dataset developed under the Efficient Humanitarian Aid Through Intelligent Image Analysis project), addresses this critical gap by capturing rural and urban damage patterns in Mozambique following Cyclone Idai. Despite encouraging early results, significant challenges persist due to task complexity, severe class imbalance, and substantial architectural diversity across regions. Building upon EDDA, this study introduces a two-stage building damage assessment pipeline that decouples localization from classification. We employ lightweight You Only Look Once (YOLO)-based detectors—RTMDet, YOLOv7, and YOLOv8—for building localization, followed by dedicated damage severity classification using state-of-the-art architectures including Compact Convolutional Transformers, EfficientNet, and ResNet. This approach tests whether separating feature extraction tasks—assigning detectors solely to localization and specialized classifiers to damage assessment—yields superior performance compared to multi-class detection models that jointly learn both objectives. Comprehensive evaluation across 640+ model combinations demonstrates that our two-stage pipeline achieves competitive performance (mAP 0.478) with enhanced modularity compared to multi-class detection baselines (mAP 0.455), offering improved robustness across diverse building types and imbalanced damage classes. Full article

Fundamental frequency estimation, also known as F0 estimation, is a crucial task in speech processing and analysis, with significant applications in areas such as speech recognition, speaker identification, and emotion detection. Traditional algorithms, while effective, often encounter challenges in real-time environments due to computational limitations. Recent advances in deep learning, especially in the use of recurrent neural networks (RNNs), have opened new opportunities for enhancing F0 estimation accuracy and efficiency. This paper introduces a novel RNN-based F0 estimation method with an attention mechanism and evaluates its performance against selected state-of-the-art F0 estimation approaches, including standard baseline methods, as well as neural-network-based regression and classification models. By integrating attention mechanisms, the model eliminates the necessity for post-processing steps and enables a more efficient seq2scal estimation process. While the self-attention mechanism used in Transformers captures all pairwise temporal dependencies at a quadratic computational cost, the proposed method’s implementation of the attention mechanism enables it to selectively focus on the most relevant acoustic cues for F0 prediction, enhancing robustness without increasing the model’s complexity. Experimental results using the LibriSpeech and Common Voice datasets demonstrate superior computational efficiency of the proposed method compared to current state-of-the-art RNN-based seq2seq models, while maintaining comparable estimation accuracy. Furthermore, the proposed “RNN-based F0 estimation method with an attention mechanism” achieves the lowest computational complexity among all compared models, while maintaining high accuracy, making it suitable for low-latency, resource-limited deployments and competitive even with standard baseline methods, such as pYIN or CREPE. Finally, the performance of the developed RNN-based F0 estimation method with attention mechanism in terms of RMSE and FLOPs demonstrates the potential of attention mechanisms and sequence modelling in achieving high accuracy alongside lightweight F0 estimation suitable for modern speech processing applications, which aligns with the growing trend towards deploying intelligent systems on resource-constrained devices. Full article

The Learning with Errors (LWE) problem, particularly its efficient ternary variant where secrets and errors are small, is a fundamental building block for numerous post-quantum cryptographic schemes. Combinatorial attacks provide a potent approach to cryptanalyzing ternary LWE. While classical attacks have achieved complexities close to their asymptotic ${\mathcal{S}}^{0.25}$ bound for a search space of size $\mathcal{S}$, their quantum counterparts have faced a significant gap: the attack by van Hoof et al. (vHKM) only reached a concrete complexity of ${\mathcal{S}}^{0.251}$, far from its asymptotic promise of ${\mathcal{S}}^{0.193}$. This work introduces an efficient quantum combinatorial attack that substantially narrows this gap. We present a quantum walk adaptation of the locality-sensitive hashing algorithm by Kirshanova and May, which fundamentally removes the need for guessing error coordinates—the primary source of inefficiency in the vHKM approach. This crucial improvement allows our attack to achieve a concrete complexity of approximately ${\mathcal{S}}^{0.225}$, markedly improving over prior quantum combinatorial methods. For concrete parameters of major schemes including NTRU, BLISS, and GLP, our method demonstrates substantial runtime improvements over the vHKM attack, achieving speedup factors ranging from $2^{16}$ to $2^{60}$ across different parameter sets and establishing the new state-of-the-art for quantum combinatorial attacks. As a second contribution, we address the challenge of polynomial quantum memory constraints. We develop a hybrid approach combining the Kirshanova–May framework with a quantum claw-finding technique, requiring only $\mathcal{O}\left(n\right)$ qubits while utilizing exponential classical memory. This work provides the first comprehensive concrete security analysis of real-world LWE-based schemes under such practical quantum resource constraints, offering crucial insights for post-quantum security assessments. Our results reveal a nuanced landscape where our combinatorial attacks are superior for small-weight parameters, while lattice-based attacks maintain an advantage for others. Full article

This study explores the impact of integrating formal and non-formal learning in secondary school education through a mixed-methods experimental design. A total of 120 students (aged 14–16) from two secondary schools were randomly assigned to an experimental group (n = 60) and a control group (n = 60). The experimental group participated in a twelve-week interdisciplinary programme that combined traditional curricular subjects with non-formal educational practices such as sports, theatre, art, and community engagement, supported by digital learning platforms. Quantitative data were collected through validated instruments, while qualitative data were gathered through observations, focus groups, and semi-structured interviews with students, teachers, and parents. Statistical analyses (t-tests and ANOVA) revealed significant improvements in intrinsic motivation, psychological well-being, and sense of belonging among students in the experimental group compared to the control group. Thematic analysis of qualitative data confirmed these findings, highlighting increased collaboration, engagement, and inclusion. The results indicate that integrating formal and non-formal education fosters holistic learning, strengthens community ties, and promotes emotional and cognitive development. These findings provide empirical support for policies and pedagogical practices aimed at developing flexible, participatory, and sustainable educational models. Full article

Whole slide imaging has transformed the field of pathology by enabling high-resolution digitization of histopathological slides. However, the large image size and variability in morphology, tissue processing, and imaging can pose challenges for robust computational analysis. When working with specific tasks in digital pathology, conventional feature extractors pretrained on general images may not provide features as relevant as those trained on histopathological images. To address this, foundation models pretrained on histopathological images have been developed. Yet, their large size and computational demands might limit widespread adoptions to specific tasks. To facilitate the low-cost adoption of these models, we utilized low-rank adaptation for finetuning the model and developed evolving prototype-based multiple instance learning (EP-MIL). Our method’s capabilities were demonstrated by applying it to the classification of two histological subtypes of lung cancer. The results show that our approach achieves competitive performance when benchmarked against a state-of-the-art technique (CLAM), while offering improvements in efficiency. Specifically, our proposed method requires 8.3 times less training runtime compared with CLAM, uses less than 200.0 MB of memory during training, and enables 73.8 times faster inference runtime. These efficiency gains, combined with competitive performance, suggest that utilizing evolving prototypes with LoRA-tuned foundation models offers a more efficient and practical approach for broader use of foundation models in resource-constrained clinical settings. Full article

Background/Objectives: Having access to antiretroviral therapy (ART) has altered the health status of people living with HIV (PLHIV) to that of having a chronic condition, with a greater life expectancy. The development of the Veterans Aging Cohort Study (VACS) Index has allowed for the prediction of 5-year mortality in PLHIV, using both HIV-related and non-HIV-related markers. The modified Charlson Index describes the comorbidity burden and is indicated to predict 10-year mortality. This study validates the Veterans Aging Cohort Study (VACS) Index 1.0 and the modified Charlson Index in a contemporary European cohort, with the aim of better predicting mortality. Methods: An observational, multicenter study was conducted using data from the eVIHa cohort in the Balearic Islands (Spain) from 2000 to 2023. The VACS Index 1.0 and the modified Charlson Index were calculated. Model discrimination was assessed using Harrell’s C-statistic, and observed mortality was estimated using Kaplan–Meier analysis. Results: Of 6913 eligible PLHIV, 4480 (64.8%) had sufficient data for VACS Index calculation and were included in the primary analysis. The excluded group (N = 2433) had significantly higher mortality (27.7% vs. 9.4%) and a greater proportion of people who inject drugs. In the analyzed cohort, the VACS Index 1.0 showed good discrimination for 5-year all-cause mortality (C-statistic: 0.759), outperforming the modified Charlson Index (C-statistic: 0.729). Discrimination was the highest for deaths from liver disease (C: 0.875) and non-HIV-related infections (C: 0.853). Conclusions: In our analyzed cohort, the VACS Index 1.0 accurately predicted 5-year mortality. However, its performance in populations with higher rates of people who inject drugs and irregular follow-up is unknown and likely to be lower. Clinicians should be aware of these limitations when applying the index in practice. Full article

Six small interference RNAs (siRNAs) have been approved as therapeutics since 2018 making them promising nanosystems due to selective gene knockdown activity. siRNA design is complex due to various factors, where the chemical modifications are crucial to improve its half-life and stability. Machine learning (ML) enabled more efficient analysis of siRNA data, moreover predicting efficacy and off-target effects. This work proposes a novel pipeline for predicting gene knockdown activity of chemically modified siRNAs across the whole range of activities leveraging both descriptors of siRNA chemical composition-aware property matrices and large language model (LLM) embeddings for target gene encoding. Several general-purpose and domain-specific fine-tuned LLMs were benchmarked on the target task, where the Mistral 7B general-purpose model slightly outperformed even the models pre-trained on genomic data. Proposed two-stage probability-enhanced model successfully mitigates data imbalance towards moderate-to-high active constructs and achieves state-of-the-art (SOTA) quality with R² = 0.84 and a RMSE = 12.27% on unseen data, where the probabilistic outputs of classifiers trained with F-scores up to 0.92 were used for regression model supervision. Moreover, leave-one-gene-out (LOGO) experiments show that the model is able to extrapolate on unseen genes, which further shows representativeness of siRNA features and gene embeddings. By filling the gap in the field of advanced chemical composition-aware siRNA design, our model aims to improve the efficacy of developed siRNA-based therapies. Full article

Rock glaciers have been recognized as key indicators of geomorphic and climatic processes in high mountain environments. In this study, Sentinel-2 MSI imagery and topographic data were integrated to construct enhanced feature sets for rock glacier identification. Three state-of-the-art deep learning models (U-Net, DeepLabV3+, and HRnet) were employed to perform semantic segmentation for extracting rock glacier boundaries in the Hunza River Basin, located in the eastern Karakoram Mountains. The combination of spectral and terrain features significantly improved the differentiation of rock glaciers from surrounding landforms, establishing a robust basis for model training. A series of comparative experiments were conducted to evaluate the performance of each model. The HRnet model achieved the highest overall accuracy, exhibiting superior capabilities in high-resolution feature representations and generalization. Using the HRnet framework, a total of 597 rock glaciers were identified, covering an area of 183.59 km². Spatial analysis revealed that these rock glaciers are concentrated between elevations of 4000 m and 6000 m, with maximum density near 5000 m, and a predominant south and southwest orientation. These spatial patterns reflect the combined influences of topography, thermal conditions, and snow accumulation on the formation and preservation of rock glaciers. The results confirm the effectiveness of deep learning-based semantic segmentation for large-scale rock glacier mapping. The proposed framework establishes a technical foundation for automated monitoring of alpine landforms and supports future assessments of rock glacier dynamics under climate variability. Full article

The presence of subtle mucosal abnormalities makes small bowel Crohn’s disease (SBCD) and other gastrointestinal lesions difficult to detect, as these features are often very subtle and can closely resemble other disorders. Although the Kvasir and Esophageal Endoscopy datasets offer high-quality visual representations of various parts of the GI tract, their manual interpretation and analysis by clinicians remain labor-intensive, time-consuming, and prone to subjective variability. To address this, we propose a generalizable ensemble deep learning framework for gastrointestinal lesion detection, capable of identifying pathological patterns such as ulcers, polyps, and esophagitis that visually resemble SBCD-associated abnormalities. Further, the classical convolutional neural network (CNN) extracts shallow high-dimensional features; due to this, it may miss the edges and complex patterns of the gastrointestinal lesions. To mitigate these limitations, this study introduces a deep learning ensemble framework that combines the strengths of EfficientNetB5, MobileNetV2, and multi-head self-attention (MHSA). EfficientNetB5 extracts detailed hierarchical features that help distinguish fine-grained mucosal structures, while MobileNetV2 enhances spatial representation with low computational overhead. The MHSA module further improves the model’s global correlation of the spatial features. We evaluated the model on two publicly available DBE datasets and compared the results with four state-of-the-art methods. Our model achieved classification accuracies of 99.25% and 98.86% on the Kvasir and Kaither datasets. Full article

The rapid growth of user-generated content in the digital space has increased the necessity of properly and interpretively analyzing sentiment and emotion systems. This research paper presents a new hybrid model, HAM (Hybrid Attention-based Model), a Transformer-based contextual embedding model combined with deep sequential modeling and multi-layer explainability. The suggested framework integrates the BERT/RoBERTa encoders, Bidirectional LSTM, and Graph Attention that can be used to embrace semantic and aspect-level sentiment correlation. Additionally, an enhanced Explainability Module, including Attention Heatmaps, Aspect-Level Interpretations, and SHAP/Integrated Gradients analysis, contributes to the increased model transparency and interpretive reliability. Four benchmark datasets, namely GoEmotions-1, GoEmotions-2, GoEmotions-3, and Amazon Cell Phones and Accessories Reviews, were experimented on in order to have a strong cross-domain assessment. The 28 emotion words of GoEmotions were merged into five sentiment-oriented classes to harmonize the dissimilarity in the emotional granularities to fit the schema of the Amazon dataset. The proposed HAM model had a highest accuracy of 96.4% and F1-score of 94.9%, which was significantly higher than the state-of-the-art baselines like BERT (89.8%), RoBERTa (91.7%), and RoBERTa+BiLSTM (92.5%). These findings support the idea that HAM is a better solution to finer-grained emotional details and is still interpretable as a vital move towards creating open, exposible, and domain-tailored sentiment intelligence systems. Future endeavors will aim at expanding this architecture to multimodal fusion, cross-lingual adaptability, and federated learning systems to increase the scalability, generalization, and ethical application of AI. Full article

Repackaged Android applications pose a significant threat to mobile ecosystems, acting as common vectors for malware distribution and intellectual property infringement. Addressing the challenges of existing repackaging detection methods—such as scalability, reliance on app pairs, and high computational costs—this paper presents a novel hybrid approach that combines supervised learning and symptom discovery. We develop a lightweight feature extraction and analysis framework that leverages only 20 discriminative features, including inter-component communication (ICC) patterns, sensitive API usage, permission profiles, and a structural anomaly metric derived from string offset order. Our experiments, conducted on 8441 Android applications sourced from the RePack dataset, demonstrate the effectiveness of our approach, achieving a maximum F1 score of 85.9% and recall of 98.8% using Support Vector Machines—outperforming prior state-of-the-art models that utilized over 500 features. We also evaluate the standalone predictive power of AndroidSOO’s string offset order feature and highlight its value as a low-cost repackaging indicator. This work offers an accurate, efficient, and scalable alternative for automated detection of repackaged mobile applications in large-scale Android marketplaces. Full article

Automated pollen recognition is a foundational tool for diverse scientific domains, including paleoclimatology, biodiversity monitoring, and agricultural science. However, conventional methods create a critical data bottleneck, limiting the temporal and spatial resolution of ecological analysis. Existing deep learning models often fail to achieve the requisite localization accuracy for microscopic pollen grains, which are characterized by their minute size, indistinct edges, and complex backgrounds. To overcome this, we introduce HieraEdgeNet, a novel object detection framework. The core principle of our architecture is to explicitly extract and hierarchically fuse multi-scale edge information with deep semantic features. This synergistic approach, combined with a computationally efficient large-kernel operator for fine-grained feature refinement, significantly enhances the model’s ability to perceive and precisely delineate object boundaries. On a large-scale dataset comprising 44,471 annotated microscopic images containing 342,706 pollen grains from 120 classes, HieraEdgeNet achieves a mean Average Precision of 0.9501 (mAP@0.5) and 0.8444 (mAP@0.5:0.95), substantially outperforming state-of-the-art models such as YOLOv12n and the Transformer-based RT-DETR family in terms of the accuracy–efficiency trade-off. This work provides a powerful computational tool for generating the high-throughput, high-fidelity data essential for modern ecological research, including tracking phenological shifts, assessing plant biodiversity, and reconstructing paleoenvironments. At the same time, we acknowledge that the current two-dimensional design cannot directly exploit volumetric Z-stack microscopy and that strong domain shifts between training data and real-world deployments may still degrade performance, which we identify as key directions for future work. By also enabling applications in precision agriculture, HieraEdgeNet contributes broadly to advancing ecosystem monitoring and sustainable food security. Full article

As technology advances, developers continually create innovative solutions to enhance smartphone security. However, the rapid spread of Android malware poses significant threats to devices and sensitive data. The Android Operating System (OS)’s open-source nature and Software Development Kit (SDK) availability mainly contribute to this alarming growth. Conventional malware detection methods, such as signature-based, static, and dynamic analysis, face challenges in detecting obfuscated techniques, including encryption, packing, and compression, in malware. Although developers have created several visualization techniques for malware detection using deep learning (DL), they often fail to accurately identify the critical malicious features of malware. This research introduces MalVis, a unified visualization framework that integrates entropy and N-gram analysis to emphasize meaningful structural and anomalous operational patterns within the malware bytecode. By addressing significant limitations of existing visualization methods, such as insufficient feature representation, limited interpretability, small dataset sizes, and restricted data access, MalVis delivers enhanced detection capabilities, particularly for obfuscated and previously unseen (zero-day) malware. The framework leverages the MalVis dataset introduced in this work, a publicly available large-scale dataset comprising more than 1.3 million visual representations in nine malware classes and one benign class. A comprehensive comparative evaluation was performed against existing state-of-the-art visualization techniques using leading convolutional neural network (CNN) architectures, MobileNet-V2, DenseNet201, ResNet50, VGG16, and Inception-V3. To further boost classification performance and mitigate overfitting, the outputs of these models were combined using eight distinct ensemble strategies. To address the issue of imbalanced class distribution in the multiclass dataset, we employed an undersampling technique to ensure balanced learning across all types of malware. MalVis achieved superior results, with 95% accuracy, 90% F1-score, 92% precision, 89% recall, 87% Matthews Correlation Coefficient (MCC), and 98% Receiver Operating Characteristic Area Under Curve (ROC-AUC). These findings highlight the effectiveness of MalVis in providing interpretable and accurate representation features for malware detection and classification, making it valuable for research and real-world security applications. Full article

This study presents a comparative evaluation of state-of-the-art Machine Learning (ML) and Explainable Artificial Intelligence (XAI) methods, specifically SHAP and LIME, for predicting electromagnetic field (EMF) strength in urban environments. Using more than 19,000 in situ EMF measurements across Catalonia, Spain, combined with high-resolution geospatial features such as building height, built-up volume, and population density, six ML algorithms were trained and assessed over 50 randomized train–test splits. The k-Nearest Neighbors (kNN) model achieved the highest predictive accuracy (RMSE = 0.623), followed by XGBoost (RMSE = 0.711) and LightGBM (RMSE = 0.717). Explainability analysis showed that SHAP consistently identified built-up volume, building height, degree of urbanization, and population density as the dominant global predictors of EMF strength, whereas LIME revealed that degree of urbanization, population density, and building height were the most influential at the local (micro-scale) level. The results demonstrate that integrating interpretable ML frameworks with enriched geospatial datasets improves both predictive performance and transparency in EMF exposure modeling, supporting data-driven urban planning and public health assessment. Full article

Recently, the convergence of advanced sensor technologies and innovations in artificial intelligence and robotics has highlighted facial emotion recognition (FER) as an essential component of human–computer interaction (HCI). Traditional FER studies based on handcrafted features and shallow machine learning have shown a limited performance, while convolutional neural networks (CNNs) have improved nonlinear emotion pattern analysis but have been constrained by local feature extraction. Vision transformers (ViTs) have addressed this by leveraging global correlations, yet both CNN- and ViT-based single networks often suffer from overfitting, single-network dependency, and information loss in ensemble operations. To overcome these limitations, we propose ArecaNet, an assembled residual enhanced cross-attention network that integrates multiple feature streams without information loss. The framework comprises (i) channel and spatial feature extraction via SCSESResNet, (ii) landmark feature extraction from specialized sub-networks, (iii) iterative fusion through residual enhanced cross-attention, (iv) final emotion classification from the fused representation. Our research introduces a novel approach by integrating pre-trained sub-networks specialized in facial recognition with an attention mechanism and our uniquely designed main network, which is optimized for size reduction and efficient feature extraction. The extracted features are fused through an iterative residual enhanced cross-attention mechanism, which minimizes information loss and preserves complementary representations across networks. This strategy overcomes the limitations of conventional ensemble methods, enabling seamless feature integration and robust recognition. The experimental results show that the proposed ArecaNet achieved accuracies of 97.0% and 97.8% using the public databases, FER-2013 and RAF-DB, which were 4.5% better than the existing state-of-the-art method, PAtt-Lite, for FER-2013 and 2.75% for RAF-DB, and achieved a new state-of-the-art accuracy for each database. Full article