Search Results (1,425)

Internet of Things deployments face escalating security threats, yet systematic methods for auditing the defensive posture of IoT device networks remain underdeveloped. Current intrusion detection evaluations focus on algorithmic accuracy while neglecting operational requirements—computational efficiency, reproducibility, and interpretable risk assessment—that security audits demand. This paper introduces a reproducible security audit framework for IoT device networks, demonstrated through systematic evaluation of four machine learning models (Random Forest, LightGBM, XGBoost, Logistic Regression) on the TON_IoT dataset containing nine attack categories targeting smart environments. Our audit methodology enforces strict feature hygiene by excluding identity-revealing attributes, benchmarks both threat detection capability and computational cost, and provides complete reproducibility artifacts including preprocessing pipelines and trained models. The framework evaluates security posture through dual lenses: binary classification (distinguishing compromised from legitimate traffic) and multiclass classification (attributing threats to specific attack types). Binary audit results show ensemble models achieve 99.8–99.9% accuracy with perfect ROC-AUC (100%) and sub-15 ms inference latency per 1000 flows, confirming reliable attack detection. Multiclass auditing reveals more nuanced findings: while overall accuracy reaches 99.0% with macro-F1 near 97%, rare attack types expose critical blind spots—man-in-the-middle threats achieve only 78% F1 despite representing serious security risks. LightGBM provides optimal audit performance, balancing 99.93% detection accuracy with 2.76 MB deployment footprint. We translate audit findings into actionable security recommendations (network segmentation, rate-limiting, TLS metadata collection) and compare against twenty published studies, demonstrating that our framework achieves competitive detection rates while uniquely delivering the transparency, efficiency metrics, and reproducibility required for credible security assessment of production IoT networks. Full article

In this paper, we propose the Adaptive Volcano Support Vector Machine (AVSVM)—a novel classification model inspired by the dynamic behavior of volcanic eruptions—for the purpose of enhancing malware detection. Unlike conventional SVMs that rely on static decision boundaries, AVSVM introduces biologically inspired mechanisms such as pressure estimation, eruption-triggered kernel perturbation, lava flow-based margin refinement, and an exponential cooling schedule. These components work synergistically to enable real-time adjustment of the decision surface, allowing the classifier to escape local optima, mitigate class overlap, and stabilize under high-dimensional, noisy, and imbalanced data conditions commonly found in malware detection tasks. Extensive experiments were conducted on the UNSW-NB15 and KDD Cup 1999 datasets, comparing AVSVM to baseline classifiers including traditional SVM, PSO-SVM, and CNN under identical computational settings. On the UNSW-NB15 dataset, AVSVM achieved an accuracy of 96.7%, recall of 95.4%, precision of 96.1%, F₁-score of 95.75%, and a false positive rate of only 3.1%, outperforming all benchmarks. Similar improvements were observed on the KDD dataset. In addition, AVSVM demonstrated smooth convergence behavior and statistically significant gains (p < 0.05) across all pairwise comparisons. These results validate the effectiveness of incorporating biologically motivated adaptivity into classical margin-based classifiers and position AVSVM as a promising tool for intelligent malware detection systems. Full article

Objectives: We propose a flow-signal-based classification of retinal artery macroaneurysms (RAMs) using Optical Coherence Tomography Angiography (OCTA) and compare the findings with fundus fluorescein angiography (FFA). Methods: A retrospective review of 49 RAM cases observed over 6 years (October 2017–March 2023) at a medical retina clinic at the University Hospital Southampton, UK. Electronic clinical records, FFA, and OCTA images (en face and B-scan) were reviewed to identify pathology and assess RAM flow profiles. Results: In total, 30 eyes from 30 patients were included. The mean age of the patients was 76 years (range 49–91), with 17 females and 13 males. All eyes underwent OCTA, enabling classification of RAMs into three flow signal types: high (9 eyes), low (10 eyes), and absent (9 eyes), while 2 eyes had haemorrhage-related artefacts. A subgroup of 13 eyes also underwent FFA, allowing direct comparison, which showed flow profiles similar to those of OCTA: high (4 eyes), low (6 eyes), and absent (2 eyes), with 1 ungradable case due to subretinal haemorrhage masking. A discrepancy in flow was observed in one case where FFA indicated flow, but OCTA did not. Despite this, FFA and OCTA generally agreed on the flow levels, with a Spearman correlation of r = 0.79 (p = 0.004). Conclusions: OCTA flow profiles were directly comparable to FFA. OCTA effectively identified different levels of blood flow signal behaviour in RAMs. The proposed flow-based RAM classification may aid in prognosis, treatment indications, follow-up, and safe repeat imaging in clinical practice without systemic risk to the patient. Full article

Dural arteriovenous fistulas (dAVFs) represent a unique subset of intracranial vascular malformations characterized by pathologic shunting between dural arteries and venous sinuses or cortical veins. Although once considered rare and uniformly high-risk, modern imaging and therapeutic innovations have revealed a spectrum of biological behavior ranging from benign to aggressive. The past decade has witnessed a paradigm shift from purely anatomic classification toward individualized, hemodynamic-based decision-making that incorporates endovascular, microsurgical, and radiosurgical techniques. This Perspective reviews the evolving management of dAVFs, emphasizing early recognition of cortical venous drainage, endovascular innovation, venous sinus reconstruction, and the emerging role of artificial intelligence and personalized medicine in risk stratification. Accordingly, we seek to delineate how a precision approach based on angioarchitecture, patterns of venous flow, and clinical phenotype has transformed the treatment of dAVFs from a purely reactive to a potentially curative discipline. Full article

More than half of the world’s highly regulated rivers are currently experiencing an unsustainable balance between ecological protection and economic development. The value realization of river eco-products is considered a key pathway to addressing this challenge; however, its effectiveness remains to be empirically verified. Therefore, the objective of this study is to develop an integrated framework for evaluating the sustainability of river ecological protection and economic development through eco-product value realization. The framework integrates the classification of river eco-products, the estimation of their potential and realized values, and the analysis of value realization pathways. Taking the Baoji section of the Weihe River (BSWHR) as a case study, the framework is applied with hydrological, hydraulic, and socio-economic datasets to empirically evaluate the coordination between ecological protection and economic development. The main results showed that: (1) River eco-products are divided into three types: public, operational, and physical operational eco-products; (2) The potential ecological value of all river eco-products in the BSWHR is estimated at 549 million CNY; (3) The realized value of all river eco-products is 288.75 million CNY under current realization paths, corresponding to a sustainability index of 0.63, indicating that the BSWHR is less sustainable and represents an asset liability river; and (4) Enhancing the protection level of river ecological flow (e-flow) and establishing a multi-stakeholder compensation mechanism can improve the sustainability of ecological protection and economic development in highly regulated rivers. The proposed framework provides a practical basis for assessing river sustainability and guiding the effective allocation of ecological protection funds. Full article

In the field of remote sensing, accurate semantic segmentation is crucial for applications such as environmental monitoring and urban planning. Effective fusion of multi-modal data is a key factor in improving land cover classification accuracy. To address the limitations of existing methods, such as inadequate feature fusion, noise interference, and insufficient modeling of long-range dependencies, this paper proposes ASGT-Net, an enhanced multi-modal fusion network. The network adopts an encoder-decoder architecture, with the encoder featuring a symmetric dual-branch structure based on a ResNet50 backbone and a hierarchical feature extraction framework. At each layer, Adaptive Weighted Fusion (AWF) modules are introduced to dynamically adjust the feature contributions from different modalities. Additionally, this paper innovatively introduces an alternating mechanism of Learnable Sparse Attention (LSA) and Adaptive Gating Fusion (AGF): LSA selectively activates salient features to capture critical spatial contextual information, while AGF adaptively gates multi-modal data flows to suppress common conflicting noise. These mechanisms work synergistically to significantly enhance feature integration, improve multi-scale representation, and reduce computational redundancy. Experiments on the ISPRS benchmark datasets (Vaihingen and Potsdam) demonstrate that ASGT-Net outperforms current mainstream multi-modal fusion techniques in both accuracy and efficiency. Full article

Truck platooning refers to a group of heavy-duty vehicles travelling in close succession through cooperative driving technologies and inter-vehicle communication. This transport solution is increasingly investigated as a promising strategy to enhance the efficiency and sustainability of road freight transport. The expected benefits include fuel and operational cost savings, reduced emissions, improved traffic flow and congestion mitigation, as well as enhanced safety for both platoon drivers and surrounding traffic. This paper presents a literature review of truck platooning, with a specific focus on the expected benefits and on how they are addressed across two fundamental perspectives: planning and control. Planning encompasses issues related to platoon formation, maintenance and reconfiguration during transport operations, whereas control focuses on the methods and schemes used to coordinate vehicle behaviour within and between platoons. The reviewed contributions are further analysed according to the methodology adopted, the level of vehicle automation, and the specific control approaches implemented. The resulting classification provides an integrated view of how different research streams contribute to economic, environmental, safety and social benefits. Finally, the current gaps and promising research directions are outlined to support future developments in large-scale platooning deployment. Full article

The growing demand for precision diagnostics in cystic fibrosis and other genetic disorders, such as cancers, is driving the need for sequencing platforms that combine analytical robustness, scalability, and cost-efficiency. In this study, we performed a direct comparison between two leading Next-Generation Sequencing (NGS) platforms, MiSeq (Illumina, CA, USA) and DNBSEQ-G99RS (MGI Tech Co., Shenzhen, China), using a CE-IVD-certified CFTR panel (Devyser AB), selected for its complexity and variant spectrum, including SNVs, CNVs, and intronic polymorphisms. A total of 47 genomic DNA samples from routine clinical activity were analyzed on both platforms. Illumina sequencing covered all CFTR variants using standard workflows, while MGI data were generated from residual diagnostic DNA, with informed consent. Sequencing data were processed using Amplicon Suite v3.7.0 for variant calling, annotation, and ACMG classification. Quality control metrics and platform-specific parameters were also evaluated. Both platforms demonstrated complete concordance in variant detection, including SNVs, CNVs, and complex alleles (e.g., Poly-T/TG). Illumina exhibited slightly superior basecalling quality and allelic frequency uniformity, while MGI achieved higher sequencing depth (mean ~2793×) and demultiplexing efficiency. No false positives, false negatives, or discordant HGVS annotations were observed. The use of full-gene CFTR sequencing enabled granular and technically rigorous cross-platform validation. These findings confirm the analytical equivalence of Illumina and MGI for diagnostic genotyping. Moreover, MGI’s greater data output and flow cell capacity may offer tangible advantages in high-throughput settings, including somatic applications such as liquid biopsy and molecular oncology workflows. Full article

The exponential growth of the Internet of Things (IoT) has made it increasingly vulnerable to cyberattacks, where malicious manipulation of network and sensor data can lead to incorrect data classification. IoT data are inherently heterogeneous, comprising sensor readings, network flow records, and device metadata that differ significantly in scale and structure. This diversity motivates transforming tabular IoT data into image-based representations to facilitate the recognition of intrusion patterns and the analysis of spatial correlations. Many deep learning models offer robust detection performance, including CNNs, LSTMs, CNN–LSTM hybrids, and Transformer-based networks, but many of these architectures are computationally intensive and require significant training resources. To address this challenge, this study introduces an evolutionary-driven framework that mathematically formalizes the transformation of tabular IoT data into image-encoded matrices and optimizes feature selection through metaheuristic algorithms. Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Variable Neighborhood Search (VNS) are employed to identify optimal feature subsets for Random Forest (RF) and Extreme Gradient Boosting (XGBoost) classifiers. The approach enhances discrimination by optimizing multi-objective criteria, including accuracy and sparsity, while maintaining low computational complexity suitable for edge deployment. Experimental results on benchmark IoT intrusion datasets demonstrate that VNS-XGBoost configurations performed better on the IDS2017 and IDS2018 benchmarks, achieving accuracies up to 0.99997 and a significant reduction in Type II errors (212 and 6 in tabular form, reduced to 4 and 1 using image-encoded representations). These results confirm that integrating evolutionary optimization with image-based feature modeling enables accurate, efficient, and robust intrusion detection across large-scale IoT systems. Full article

The coexistence of class imbalance and overlap poses a major challenge in classification and significantly limits model accuracy. Data-level methods alleviate class imbalance by generating samples, but without ensuring their rationality, which may introduce noise. Algorithm-level methods are designed based on the model training process, avoiding noise introduction. However, existing methods often fail to consider the potential multiclass scenarios within overlap regions or design targeted solutions for different overlap patterns. This paper proposes an ensemble imbalanced classification framework via dual-perspective overlapping analysis with multi-resolution metrics. The dataset is divided into multiple resolutions for independent analysis, capturing distributional information from local to global levels. For each independent resolution, overlap is analyzed from the perspectives of “feature overlap” and “instance overlap” to derive more refined overlap scores. Flow model mapping and importance weighting are, respectively, applied to refine overlapping samples according to the two criteria. During testing, classifiers are adaptively selected based on the overlap degree of test samples under different criteria, and predictions across resolutions are integrated for the final decision. Experiments on 39 datasets demonstrate that the proposed method outperforms typical imbalanced classification methods in F-measure and G-mean, with particularly notable gains on 15 severely overlapping datasets. Full article

Background: Early and accurate cancer detection remains a critical challenge in global healthcare. Deep learning has shown strong diagnostic potential, yet widespread adoption is limited by dependence on high-performance hardware, centralized servers, and data-privacy risks. Methods: This study introduces a browser-based multi-cancer classification framework that performs real-time, client-side inference using TensorFlow.js—eliminating the need for external servers or specialized GPUs. The proposed model fine-tunes the Xception architecture, leveraging depthwise separable convolutions for efficient feature extraction, on a large multi-cancer dataset of over 130,000 histopathological and cytological images spanning 26 cancer types. It was benchmarked against VGG16, ResNet50, EfficientNet-B0, and Vision Transformer. Results: The model achieved a Top-1 accuracy of 99.85% and Top-5 accuracy of 100%, surpassing all comparators while maintaining lightweight computational requirements. Grad-CAM visualizations confirmed that predictions were guided by histopathologically relevant regions, reinforcing interpretability and clinical trust. Conclusions: This work represents the first fully browser-deployable, privacy-preserving deep learning framework for multi-cancer diagnosis, demonstrating that high-accuracy AI can be achieved without infrastructure overhead. It establishes a practical pathway for equitable, cost-effective global deployment of medical AI tools. Full article

Frequent extreme climate events have intensified landslide hazards in mountainous regions, necessitating efficient identification and classification to understand movement mechanisms and mitigate risks. This study develops a novel, non-contact InSAR framework that seamlessly integrates three key steps—Identification, Inversion, and Classification—to address this challenge. By applying this framework to ascending and descending Sentinel-1 data in the complex terrain of the Jishi Mountain region, we first introduce geometric distortion masking and a C-Index deformation consistency check, which enables the reliable identification of 530 active landslides, with 154 detected in both orbits. Second, we employ a local parallel flow model to invert the landslide movement geometry without relying on DEM-derived prior assumptions, successfully retrieving the two-dimensional (sliding and normal direction) deformation fields for all 154 consistent landslides. Finally, by synthesizing these 2D deformation patterns with geomorphological features, we achieve a systematic classification of movement types, categorizing them into retrogressive translational (31), progressive translational (66), rotational (19), composite (24), and earthflows (14). This integrated methodology provides a validated, transferable solution for deciphering landslide mechanisms and assessing risks in remote, complex mountainous areas. Full article

Background: The development of anti-drug antibodies (ADAs) and resulting immune complexes are key mechanisms behind the secondary loss of response to adalimumab in Crohn’s disease (CD). Despite their clinical importance, routine immunogenicity assays are limited, underscoring the need for alternative predictive approaches. Objective: This study aimed to develop interpretable artificial neural network (ANN) models to predict immune complex formation and estimate serum adalimumab levels using routinely available clinical and laboratory data from CD patients. Methods: A prospective analysis was performed on 58 CD patients on maintenance adalimumab. Immune complexes and serum adalimumab were measured via ELISA and lateral flow assays. ANN and ensemble regression models were trained on demographic, clinical, and inflammatory data, with performance evaluated by five-fold cross-validation. Interpretability was enhanced using Garson’s algorithm and permutation importance. Results: The ANN-based classification model accurately predicted ADA immune complex formation, achieving an accuracy of 77.47% and an area under the curve (AUC) of 82.63%. The main predictive variables included extraintestinal manifestations, perianal disease, disease behavior, and age at diagnosis. For estimating serum adalimumab levels measured by ELISA, the model performed modestly (accuracy 59.89%, AUC 79.72%), incorporating factors such as Montreal classification, perianal disease, C-reactive protein, immunosuppressant use, and disease duration. Conclusions: Interpretable ANN models robustly predict anti-adalimumab immune complexes and, to a lesser extent, serum adalimumab, using clinically available data, including perianal disease. This proof-of-concept study is limited by the relatively small, single-center dataset (n = 58), which may affect model generalizability and increase the risk of overfitting. External validation in larger and multicenter cohorts is required before clinical implementation. Full article

This paper introduces a novel approach to handwritten digit recognition based on directional flood simulation and topological feature extraction. While traditional pixel-based methods often struggle with noise, partial occlusion, and limited data, our method leverages the structural integrity of digits by simulating water flow from image boundaries using a modified breadth-first search (BFS) algorithm. The resulting flooded regions capture stroke directionality, spatial segmentation, and closed-area characteristics, forming a compact and interpretable feature vector. Additional parameters such as inner cavities, perimeter estimation, and normalized stroke density enhance classification robustness. For efficient prediction, we employ the Annoy approximate nearest neighbors algorithm using ensemble-based tree partitioning. The proposed method achieves high accuracy on the MNIST (95.9%) and USPS (93.0%) datasets, demonstrating resilience to rotation, noise, and limited training data. This topology-driven strategy enables accurate digit classification with reduced dimensionality and improved generalization. Full article

Background: Interstitial lung diseases (ILDs) often progress quickly and are associated with a poor prognosis. New noninvasive biomarkers to assist in the classification and prognostication of ILD are needed. Exhaled nitric oxide (FeNO), Cavity nitric oxide (CaNO), and carbon monoxide (eCO) are biomarkers of airway inflammation, widely used in respiratory inflammatory diseases such as asthma and chronic obstructive pulmonary disease (COPD). However, their value in ILD remains unclear. Objective: To evaluate the potential diagnostic and prognostic value of FeNO, CaNO, and eCO in ILD, and explore their integration into clinical practice. Methods: A total of 237 patients were recruited for the study, including 14 with idiopathic pulmonary fibrosis (IPF), 46 with interstitial pneumonia with autoimmune features (IPAF), 19 with mixed connective tissue disease–associated ILD (MCTD-ILD), 65 with polymyositis/dermatomyositis-associated ILD (PM/DM-ILD), 17 with rheumatoid arthritis-associated ILD (RA-ILD), 7 with systemic lupus erythematosus-associated ILD (SLE-ILD), 19 with Sjögren’s syndrome-associated ILD (SS-ILD), and 50 with systemic sclerosis-associated ILD (SSc-ILD). Multiple-flow FeNO and eCO analyses were performed in this population. The associations of these biomarkers with pulmonary function, acute exacerbations, and radiologic fibrosis classification were evaluated. Results: Patients with IPF exhibited significantly higher levels of FeNO at 50 mL/s (FeNO50) compared to those with connective tissue disease-associated ILD (CTD-ILD) and IPAF. Both CaNO and eCO were negatively correlated with pulmonary function parameters, particularly forced vital capacity (FVC) and diffusing capacity of the lung for carbon monoxide (DLCO). Receiver operating characteristic (ROC) curve analysis indicated that CaNO is a reliable biomarker for acute exacerbation, with an area under the ROC curve (AUC) of 0.8887, and a cutoff value of 6.35. Additionally, CaNO > 6.35 was associated with a relative risk (RR) of 12.87 for acute exacerbation (AE) compared to CaNO ≤ 6.35. Moreover, both CaNO and eCO levels were significantly higher in the fibrotic ILD group compared to the non-fibrotic group, with ROC analysis indicating AUCs of 0.7173 for CaNO and 0.6875 for eCO. Conclusions: FeNO, CaNO, and eCO can provide strong support for the early diagnosis and monitoring of ILD, especially with CaNO playing a crucial role in predicting acute exacerbations. Integrating these biomarkers into clinical practice can help doctors more accurately assess the progression of ILD and develop personalized treatment plans, ultimately improving the prognosis of ILD patients. Future research is needed to validate the effectiveness of these biomarkers in clinical management, facilitating their integration as standard tools for clinical monitoring. Full article