Search Results (62)

This work investigates a data-driven approach for detecting structural damage in the wing of a Cessna 172 aircraft using reduced-order finite element (FE) models. This study focuses on the ability of machine learning methods to generalize across different structural conditions, aiming to support reliable Structural Health Monitoring (SHM) in aeronautical applications. The wing was first modeled in detail using the FiniteElement Method, followed by the development of a simplified FE model to reduce computational cost while maintaining accuracy. The similarity between the two models was evaluated through modal analysis and the Modal Assurance Criterion (MAC). Dynamic excitation representing turbulence effects was applied to simulate healthy and damaged conditions, producing acceleration data used to train one-dimensional and two-dimensional neural network classifiers. The 1D models processed raw vibration signals, while the 2D models used image representations of the same data. Both architectures were tested against results from the detailed FE model to assess their generalization capability. The 2D networks achieved higher classification accuracy, demonstrating improved robustness in identifying both minor and severe damage. The findings highlight the potential of combining reduced FE models with data-driven methods for efficient and accurate aircraft wing damage detection. Full article

Slicing functionality makes it possible for an operator to share a 5G physical infrastructure between several virtual networks operated by different institutions. The deployed slices can support a wide range of applications with conflicting QoS targets. The coexistence of these slices on top of a common infrastructure is challenging and remains an open issue. Identifying traffic associated with a given type of slice is required to operate and control network resources in an efficient and secure way. This work proposes new algorithms operating at the physical and MAC layers. The solutions designed identify traffic generated by URLLC and eMBB slices by defining a new LightGBM framework. The algorithms can operate at the base station level in an O-RAN-type architecture. They provide a valuable input to radio resource management and traffic steering procedures. Full article

In the context of tightening sustainability regulations and rising demands for transparent and responsible capital allocation, understanding how digital financial innovations influence market efficiency has become increasingly important. This study examines the impact of Financial Technology (FinTech) solutions and crowdfunding platforms on sustainable market efficiency, volatility dynamics, and risk structures in the United Kingdom. Using weekly data for the Financial Times Stock Exchange 100 (FTSE 100) index from January 2010 to June 2025, the analysis applies the Lo–MacKinlay variance ratio test to assess compliance with the Random Walk Hypothesis as a proxy for informational efficiency. Firm-level proxies for FinTech and crowdfunding activity are constructed using the Nomenclature of Economic Activities (NACE) and Standard Industrial Classification (SIC) systems. The empirical results indicate substantial deviations from random-walk behavior in crowdfunding-related market segments, where persistent positive autocorrelation and elevated volatility reflect liquidity constraints and informational frictions. By contrast, FinTech-dominated segments display milder inefficiencies and faster information absorption, pointing to more stable price-adjustment mechanisms. After controlling for structural distortions through heteroskedasticity-consistent corrections and volatility adjustments, variance ratios converge toward unity, suggesting a restoration of informational efficiency. The results provide relevant insights for investors, regulators, and policymakers seeking to align financial innovation with the objectives of sustainable financial systems. Full article

Background/Objectives: To analyze the parent-perceived quality of life (QoL) in children with cerebral palsy (CP) and to study the relationship between sociodemographic and clinical factors and this perception, under the perspective of the International Classification of Functioning, Disability and Health (ICF). Methods: A cross-sectional study was conducted with 95 participants (ages 5–19 years) with CP. Participants’ parents were asked about sociodemographic and clinical characteristics and compiled Cerebral Palsy Quality of Life (CP-QoL) and Pediatric Disability Inventory-Computer Adaptive Test (PEDI-CAT). Participants were assessed and classified into the following functional domains: gross motor function (GMFM-88, GMFCS), manual ability (MACS), eating and drinking abilities (EDACS), and communication function (CFCS). Correlations between CP-QoL domains and variables were investigated using Spearman’s correlation coefficient and multivariate predictive models were used to investigate the variables predicting CP-QoL scores for each domain. Results: A total of 95 children with a mean age of 12.4 ± 3.5 years (range 5–19 years) were included. Participants demonstrated moderate-high GMFM-88 level (228.8 ± 44.7) and high functional performance across PEDI-CAT domains: Activity (57.2 ± 5.1), Mobility (63.1 ± 5.6), and Social/Cognitive (70.2 ± 4.3). Parent-perceived QoL was significantly higher when children did not require AFOs, botulinum toxin, or recent hospitalizations, and lower among children who attended physical therapy >2 h/week. Moderate correlations were consistently found between the ‘Feelings about Functioning’ domain and functional variables, being positive for GMFM-88 and all PEDI-CAT domains, and negative for GMFCS, MACS, EDACS and CFCS. That domain of CP-QoL was best explained by the regression model (R² = 0.619, p < 0.001), with the combination of three variables: GMFM-88, PEDI-CAT Activity and PEDI-CAT Social/Cognitive. Among them, PEDI-CAT Activity was the strongest predictor (β = 0.1436). Conclusions: In children with CP, to enhance family well-being, interventions should prioritize social participation and carefully balance the intensity and frequency of therapy against family burden and daily life demands, as QoL is primarily driven by manual ability and functional performance. Full article

The Internet of Military Things (IoMT) relies on Long-Range Wide Area Networks (LoRaWAN) for low-power, long-range communication in critical applications like border security and soldier health monitoring. However, conventional priority-based flow control mechanisms, which rely on static classification thresholds, lack the adaptability to handle the nuanced, continuous nature of physiological data and dynamic network states. To overcome this rigidity, this paper introduces a novel, domain-adaptive Fuzzy Logic Flow Control (FFC) protocol specifically tailored for LoRaWAN-based IoMT. While employing established Mamdani inference, the FFC system innovatively fuses multi-parameter physiological data (body temperature, blood pressure, oxygen saturation, and heart rate) into a continuous Health Score, which is then mapped via a context-optimised sigmoid function to dynamic transmission intervals. This represents a novel application-layer semantic integration with LoRaWAN’s constrained MAC and PHY layers, enabling cross-layer flow optimisation without protocol modification. Simulation results confirm that FFC significantly enhances reliability and energy efficiency while reducing latency relative to traditional static priority architectures. Seamlessly integrated into the NS-3 LoRaWAN simulation framework, the FFC protocol demonstrates superior performance in IoMT communications. Simulation results confirm that FFC significantly enhances reliability and energy efficiency while reducing latency compared with traditional static priority-based architectures. It achieves this by prioritising high-priority health telemetry, proactively mitigating network congestion, and optimising energy utilisation, thereby offering a robust solution for emergent, health-critical scenarios in resource-constrained environments. Full article

In automated warehousing scenarios, mobile UHF RFID robots typically operate along preset fixed paths to collect basic information from goods tags. They lack the ability to perceive shelf layouts and goods distribution, leading to problems such as missing reads and low inventory efficiency. To address this issue, this paper proposes a cross-layer sensing method for mobile UHF RFID robot reading states based on multiple linear regression-orthogonal least squares (MLR-OLS) and random forest. For shelf state sensing, a position sensing model is constructed based on the physical layer, and MLR-OLS is used to estimate shelf positions and interaction time. For good state sensing, combining physical layer and MAC layer features, a K-means-based tag density classification method and a missing tag count estimation algorithm based on frame states and random forest are proposed to realize the estimation of goods distribution and the number of missing goods. On this basis, according to the read state sensing results, this paper further proposes an adaptive reading strategy for RFID robots to perform targeted reading on missing goods. Experimental results show that when the robot is moving at medium and low speeds, the proposed method can achieve centimeter-level shelf positioning accuracy and exhibit high reliability in goods distribution sensing and missing goods count estimation, and the adaptive reading strategy can significantly improve the goods read rate. This paper realizes cross-layer sensing and read optimization of the RFID robot system, providing a theoretical basis and technical route for the application of mobile UHF RFID robot systems. Full article

Accurate indoor floor classification is essential for wireless positioning systems. However, the performance of conventional received signal strength indictor (RSSI)-based fingerprinting approaches is often limited by signal fluctuations and insufficient feature representation. To address these challenges, this paper introduces a heterogeneous graph neural network (GNN) framework that models WiFi signals using two types of nodes: reference points and Media Access Control (MAC) address. The edges between reference points and MAC addresses are weighted by normalized RSSI values, allowing the model to capture signal strength interactions and perform relation-aware message passing. Through this graph-based representation, the model can learn spatial and signal dependencies more effectively than conventional vector-based approaches. The proposed model was extensively evaluated under both benchmark and practical settings. On small-scale datasets, it achieved performance comparable to that of a conventional convolutional neural network trained on large-scale datasets, confirming its effectiveness with limited samples. In addition, the proposed model consistently outperformed other models under noisy conditions, achieving 93.88% accuracy on the widely used UJIIndoorLoc dataset and 97.3% accuracy in real-time experiments conducted at a test site. These values are significantly higher than those achieved using conventional machine learning (ML) baselines, highlighting the ability of the proposed model to handle real-world signal variations. These findings highlight that the heterogeneous GNN effectively captures spatial and signal-level dependencies, offering a robust and scalable solution for accurate indoor floor classification. Overall, this work presents a promising pathway for improving the performance and reliability of future wireless positioning systems. Full article

Wideband spectrum sensing is a crucial technology for the efficient utilization of limited frequency resources in cognitive radio. While deep learning models have yielded promising results, they typically rely on either time-domain (I/Q) or frequency-domain (FFT) data alone, which can limit their performance. This study proposes IPFSCNN (IQ-Parallel FFT-Serial CNN), a novel asymmetric hybrid architecture that synergistically fuses both data representations. The key idea of its design is an asymmetric architecture that employs two specialized streams: a parallelized branch to efficiently capture temporal features from I/Q data, and a deep serial branch to extract spectral patterns from FFT data. These complementary features are fused to perform a multi-label classification task. Experiments on an LTE-M dataset demonstrate that the proposed IPFSCNN achieves a higher detection performance than state-of-the-art models, including DeepSense and ParallelCNN, particularly in low signal-to-noise ratio conditions. Furthermore, IPFSCNN achieves this superior accuracy while maintaining high computational efficiency, requiring 15% fewer parameters and only one-third of the multiply-accumulate (MAC) operations compared to the DeepSense model. Crucially, a comprehensive ablation study validates this asymmetric design, proving that the proposed ‘IQ-Parallel FFT-Serial’ combination is demonstrably superior to other hybrid configurations. Full article

Human Activity Recognition (HAR) is widely used for healthcare, but few works focus on Manual Material Handling (MMH) activities, despite their diffusion and impact on the workers’ health. We propose four Deep Learning algorithms for HAR in MMH: Bidirectional Long Short-Term Memory (BiLSTM), Sparse Denoising Autoencoder (Sp-DAE), Recurrent Sp-DAE, and Recurrent Convolutional Neural Network (RCNN). We explored different hyperparameter combinations to maximize the classification performance (F1-score,) using wearable sensors’ data gathered from 14 subjects. We investigated the best three-parameter combinations for each network using the full dataset to select the two best-performing networks, which were then compared using 14 datasets with increasing subject numerosity, 70–30% split, and Leave-One-Subject-Out (LOSO) validation, to evaluate whether they may perform better with a larger dataset. The benchmarking network DeepConvLSTM was tested on the full dataset. BiLSTM performs best in classification and complexity (95.7% 70–30% split; 90.3% LOSO). RCNN performed similarly (95.9%; 89.2%) with a positive trend with subject numerosity. DeepConvLSTM achieves similar classification performance (95.2%; 90.3%) but requires $\times 57.1$ and $\times 31.3$ more Multiply and ACcumulate (MAC) and $\times 100.8$ and $\times 28.3$ more Multiplication and Addition (MA) operations, which measure the complexity of the network’s inference process, than BiLSTM and RCNN, respectively. The BILSTM and RCNN perform close to DeepConvLSTM while being computationally lighter, fostering their use in embedded systems. Such lighter algorithms can be readily used in the automatic ergonomic and biomechanical risk assessment systems, enabling personalization of risk assessment and easing the adoption of safety measures in industrial practices involving MMH. Full article

Background/Objectives: Feeding and swallowing difficulties are common in young children with cerebral palsy (CP), yet no validated tool has been available in Korea for those under 3 years. The Mini-Eating and Drinking Ability Classification System (Mini-EDACS) was designed for children aged 18–36 months. This study aimed to translate the Mini-EDACS into Korean and evaluate its reliability and validity. Methods: Translation followed international guidelines, including forward–backward translation and Delphi consensus with experts in pediatric dysphagia. Forty-eight children with CP (mean age 27.1 ± 5.0 months) were assessed. Caregivers and speech–language pathologists (SLPs) independently rated Mini-EDACS and assistance levels. Inter-rater reliability was examined using Cohen’s κ. Construct validity was tested by Spearman’s correlations with the Gross Motor Function Classification System (GMFCS), Mini-MACS, the Communication Function Classification System (CFCS), the Visual Function Classification System (VFCS), and the Functional Oral Intake Scale for Children (FOIS-C). Results: Agreement between caregivers and SLPs was excellent (κ = 0.90; weighted κ = 0.98). Assistance-level ratings also showed almost perfect concordance (κ = 0.97). Mini-EDACS correlated strongly with FOIS-C (ρ = −0.86, p < 0.001) and with assistance levels (ρ = 0.81, p < 0.001). Moderate-to-strong positive correlations were observed with GMFCS (ρ = 0.56), Mini-MACS (ρ = 0.64), CFCS (ρ = 0.61), and VFCS (ρ = 0.61), supporting construct validity. Conclusions: The Korean Mini-EDACS is a reliable and valid tool for classifying eating and drinking abilities in children with CP under 3 years. It enables standardized communication between caregivers and clinicians, complements existing functional classification systems, and may facilitate earlier identification and intervention for feeding difficulties. Full article

Accurate skin lesion classification is crucial for the early detection of malignant lesions, including melanoma, as well as improved patient outcomes. While convolutional neural networks (CNNs) excel at capturing local morphological features, they struggle with global context modeling essential for comprehensive lesion assessment. Vision transformers address this limitation but suffer from quadratic computational complexity O(n²), hindering deployment in resource-constrained clinical environments. We propose DermaMamba, a novel dual-branch fusion architecture that integrates CNN-based local feature extraction with Vision Mamba (VMamba) for efficient global context modeling with linear complexity O(n). Our approach introduces a state space fusion mechanism with adaptive weighting that dynamically balances local and global features based on lesion characteristics. We incorporate medical domain knowledge through multi-directional scanning strategies and ABCDE (Asymmetry, Border irregularity, Color variation, Diameter, Evolution) rule feature integration. Extensive experiments on the ISIC dataset show that DermaMamba achieves 92.1% accuracy, 91.7% precision, 91.3% recall, and 91.5% mac-F1 score, which outperforms the best baseline by 2.0% accuracy with 2.3× inference speedup and 40% memory reduction. The improvements are statistically significant based on a significance test (p < 0.001, Cohen’s d > 0.8), with greater than 79% confidence also preserved on challenging boundary cases. These results establish DermaMamba as an effective solution bridging diagnostic accuracy and computational efficiency for clinical deployment. Full article

Ocular disease (OD) represents a complex medical condition affecting humans. OD diagnosis is a challenging process in the current medical system, and blindness may occur if the disease is not detected at its initial phase. Recent studies showed significant outcomes in the identification of OD using deep learning (DL) models. Thus, this work aims to develop a multi-classification DL-based model for the classification of seven ODs, including normal (NOR), age-related macular degeneration (AMD), diabetic retinopathy (DR), glaucoma (GLU), maculopathy (MAC), non-proliferative diabetic retinopathy (NPDR), and proliferative diabetic retinopathy (PDR), using color fundus images (CFIs). This work proposes a custom model named the ocular disease detection model (ODDM) based on a CNN. The proposed ODDM is trained and tested on a publicly available ocular disease dataset (ODD). Additionally, the SMOTE Tomek (SM-TOM) approach is also used to handle the imbalanced distribution of the OD images in the ODD. The performance of the ODDM is compared with seven baseline models, including DenseNet-201 (R₁), EfficientNet-B0 (R₂), Inception-V3 (R₃), MobileNet (R₄), Vgg-16 (R₅), Vgg-19 (R₆), and ResNet-50 (R₇). The proposed ODDM obtained a 98.94% AUC, along with 97.19% accuracy, a recall of 88.74%, a precision of 95.23%, and an F1-score of 88.31% in classifying the seven different types of OD. Furthermore, ANOVA and Tukey HSD (Honestly Significant Difference) post hoc tests are also applied to represent the statistical significance of the proposed ODDM. Thus, this study concludes that the results of the proposed ODDM are superior to those of baseline models and state-of-the-art models. Full article

Maternal and child health during pregnancy is an important issue in global public health, and the classification accuracy of fetal cardiotocography (CTG), as a key tool for monitoring fetal health during pregnancy, is directly related to the effectiveness of early diagnosis and intervention. Due to the serious category imbalance problem of CTG data, traditional models find it challenging to take into account a small number of categories of samples, increasing the risk of leakage and misdiagnosis. To solve this problem, this paper proposes a two-step innovation: firstly, we design a method of adaptive adjustment of misclassification loss function weights (MAAL), which dynamically identifies and increases the focus on misclassified samples based on misclassification rates. Secondly, a primary and secondary correction network model (MAC-NET) is constructed to carry out secondary correction for the misclassified samples of the primary model. Experimental results show that the method proposed in this paper achieves 99.39% accuracy on the UCI publicly available fetal health dataset, and also obtains excellent performance on other domain imbalance datasets. This demonstrates that the model is not only effective in alleviating the problem of category imbalance, but also has very high clinical utility. Full article

Accurate antimicrobial susceptibility testing (AST) of cefiderocol remains a diagnostic challenge, especially in infections caused by metallo-β-lactamase (MBL)-producing Klebsiella pneumoniae. While disk diffusion offers a cost-effective alternative to broth microdilution, it is highly sensitive to factors such as media composition and the presence of atypical colony morphology. The objective of this study was to evaluate how different agar media and interpretations of isolated colonies affect the performance and reliability of cefiderocol AST by disk diffusion. A total of 50 clinical K. pneumoniae MBL isolates were tested using disk diffusion on Columbia with blood, MacConkey, and chromogenic agars from three manufacturers. Inhibition zones were compared with MICs from broth microdilution. Statistical analyses included paired t-tests and Spearman correlation to assess media effects and zone morphology impact. Variability in inhibition zone diameters was observed between media, notably with chromogenic agar. The most consistent results were obtained using Graso Biotech and Thermo Fisher Columbia with blood agar. Isolated colonies were observed in over half the samples and, depending on how they were interpreted, led to major changes in classification accuracy. Up to 64% of results fell into the EUCAST area of technical uncertainty (ATU), and categorical agreement varied across media and interpretive criteria. Disk diffusion for cefiderocol may be used in resource-limited settings but only if rigorously standardized using validated media, consistent zone reading, and ATU-aware interpretive strategies. In borderline cases or when morphological anomalies are present, broth microdilution should be considered the sole reliable method. Clinical microbiologists are advised to exercise caution with ambiguous results and seek expert or confirmatory testing when needed. Full article

Edge-computing applications demand ultra-low-power architectures for both feature extraction and classification tasks. In this manuscript, a Keyword Spotting (KWS) system tailored for energy-constrained portable environments is proposed. A 16-channel analog filter bank is employed for audio feature extraction, followed by a digital Gated Recurrent Unit (GRU) classifier. The filter bank is behaviorally modeled, making use of second-order band-pass transfer functions, simulating the analog front-end (AFE) processing. To enable efficient deployment, the GRU classifier is trained using a Learned Step Size (LSQ) and Look-Up Table (LUT)-aware quantization method. The resulting quantized model, with 4-bit weights and 8-bit activation functions (W4A8), achieves 91.35% accuracy across 12 classes, including 10 keywords from the Google Speech Command Dataset v2 (GSCDv2), with less than 1% degradation compared to its full-precision counterpart. The model is estimated to require only 34.8 kB of memory and 62,400 multiply–accumulate (MAC) operations per inference in real-time settings. Furthermore, the robustness of the AFE against noise and analog impairments is evaluated by injecting Gaussian noise and perturbing the filter parameters (center frequency and quality factor) in the test data, respectively. The obtained results confirm a strong classification performance even under degraded circuit-level conditions, supporting the suitability of the proposed system for ultra-low-power, noise-resilient edge applications. Full article