Search Results (6,137)

Reliable randomness assessment is essential for evaluating random number generators used in cryptographic systems. This study presents the Bidirectional Convolutional Multi-head Attention Network (BCM-Net), which combines convolutional layers, a bidirectional gated recurrent unit, and multi-head attention for empirical discrimination between candidate and reference sequences. The evaluation uses Random.org reference data, linear congruential generators (LCGs) with moduli from $2^{26}$ to $2^{34}$, and an amplified spontaneous emission-based quantum random number generator under three post-processing settings. BCM-Net flags $X_{LCG-30}$ and $X_{LCG-32}$, although they pass the reported NIST SP 800-22 tests. In the baseline comparison on $X_{LCG-32}$, its absolute difference between mean output scores is 52.38 percentage points (pp), compared with 32.88 pp for LSTM, 9.66 pp for CNN, and 0.02 pp for FNN. For the QRNG data, the 11-LSB output is flagged, while the 8-LSB and Toeplitz outputs are not. Under the generator configurations, finite observation lengths, preprocessing, and evaluation protocol examined here, these findings support empirical sequence discrimination as a complementary screening method. They do not establish detection performance for untested fractions of a generator period or certify randomness or cryptographic security. Full article

Post-traumatic stress disorder (PTSD) is associated with immune and autonomic disturbances, but molecular programs linking PTSD-related blood transcriptional signals with autonomic–cardiac biology remain incompletely characterized. The public Gene Expression Omnibus (GEO) cohorts analyzed here did not directly measure palpitations or autonomic dysfunction. Peripheral-blood transcriptomes from GSE81761 and GSE63878 were integrated with a prespecified pathway-derived autonomic–cardiac gene set constructed from Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Genes overlapping the curated pathway set were assessed using functional enrichment and protein–protein interaction (PPI) analyses. Multi-algorithm comparison and a separate random-forest/SHapley Additive exPlanations (SHAP) analysis were performed for internal model assessment and biological prioritization. GSE64813 and GSE97356 were analyzed using targeted single-gene analyses and cohort-specific multivariable logistic models. Model performance was examined using repeated nested cross-validation, a fully nested sensitivity analysis, and locked-transfer testing to GSE64813 and GSE97356. Interleukin 15 (IL15) and protein phosphatase 2 scaffold subunit Aalpha (PPP2R1A) were evaluated using external-cohort analyses, immune-cell deconvolution, and descriptive postmortem hippocampal single-nucleus data. Among 1679 nominally significant PTSD-associated candidate genes, 92 overlapped the curated pathway set and were enriched for cytokine signaling, chemotaxis, apoptosis, calcium transport, and PP2A-related functions. PPI analysis yielded 37 recurrent candidate hub genes. The original model comparison ranked support-vector machine (SVM) the highest across the merged and source-cohort summaries [mean area under the receiver-operating-characteristic curve (AUC) 0.853], but these estimates represent internal discovery-stage performance. IL15 was recurrently prioritized by network-based analyses, whereas PPP2R1A was a component of an enriched phosphatase-related module. IL15 and PPP2R1A showed modest single-gene effects in GSE64813 (AUC 0.566 and 0.612) and GSE97356 (AUC 0.558 and 0.593). Cohort-specific refitted models showed apparent AUC values of 0.834 and 0.749, respectively. Repeated nested cross-validation conditional on the preselected 29-gene feature set identified L2-regularized logistic regression as the best-performing algorithm (mean AUC = 0.690). Importantly, a more stringent fully nested sensitivity analysis, in which differential-expression screening and pathway intersection were repeated within each outer training fold, yielded a mean repeated outer-cross-validated AUC of 0.593 [standard deviation (SD) = 0.033; range, 0.541–0.625], indicating limited predictive performance. Locked-transfer AUCs were 0.664 in GSE64813 and 0.534 in GSE97356. Single-nucleus summaries suggested donor- and nucleus-type-dependent expression patterns for IL15 and PPP2R1A. These findings identify IL15-related cytokine signaling and PPP2R1A-related phosphatase regulation as candidate molecular programs linking PTSD-associated transcriptional variation with pathway-derived autonomic–cardiac biology. The results should therefore be interpreted as hypothesis-generating rather than direct molecular evidence for unmeasured palpitation symptoms. Prospective validation in independent, clinically well-characterized cohorts with standardized autonomic and cardiac phenotyping is warranted. Full article

Conventional logging and elemental logging-based lithology identification for volcanic buried-hill reservoirs is hampered by overlapping logging responses, low vertical resolution of elemental measurements, and inter-sample class imbalance, which lead to unsatisfactory classification accuracy. Existing cross-plot empirical methods cannot reliably distinguish lithologies with similar geophysical signatures (e.g., volcanic breccia versus andesite), especially for complex Mesozoic volcanic successions in offshore Bohai Bay Basin. To fill this technical gap, this work proposes an improved random-forest workflow integrating multi-source log-data fusion, depth-aligned linear interpolation, mutual-information feature screening, and SMOTE oversampling. The core methodological innovations lie in: (1) depth matching between conventional continuous logs and sparsely sampled elemental logging by linear interpolation to construct complete multi-feature datasets; (2) eliminating redundant input variables via mutual-information-based feature selection; and (3) mitigating lithology-sample imbalance with SMOTE synthetic-sample generation prior to random-forest training, rather than directly applying off-the-shelf random-forest classifiers. Six dominant lithologies are recognized within the Mesozoic buried-hill of Block KL16-1: basalt, andesite, rhyolite, volcanic breccia, tuff, and tuffaceous conglomerate. The proposed improved random-forest model yields an overall lithology-identification accuracy of 87% and average recall of 85.9%, substantially outperforming traditional cross-plot approaches. Confusion-matrix error analysis demonstrates that the workflow greatly reduces misclassification between easily confused lithological pairs (volcanic breccia andesite, tuff andesite). This study not only delivers a practical tool for fine reservoir evaluation and reservoir-facies prediction in Bohai Mesozoic buried-hill plays but also provides a reproducible reference for machine-learning-driven lithology interpretation in analogous offshore volcanic-reservoir settings. Full article

Privacy is a major concern in the Internet Healthcare of Things (IoHT), where threat actors may intrude systems to access personally identifiable data. Federated Learning (FL) is a well suited Machine Learning (ML) approach to preserve confidentiality, availability, and integrity in such settings during data analysis. In this work, we introduce a Network of Quantum ML (N-QML) approach for IoHT intrusion detection, integrating quantum PCA (QPCA) with Quantum FL (QPCA+QFL), tested on a subset of the data set WUSTL-EHMS-2020 using classical and quantum computing experiments across three seeds, compared against conventional ML (CML) on three feature dimensions. Among CML techniques, the integration of PCA and ANN (PCA+ANN) attained the best mean accuracy, 76.6%, using two features. Among QML techniques, QPCA+QNN achieved the best centralized accuracy, 74.4%, when two features are used, while PCA+SVM outperformed QPCA+QSVM using ten features (70.1% versus 66.9%). As a result of the study, during federation of the quantum model, accuracy was realized to be reduced steadily from 62.9% to 54.7% as dimensionality changed and highest variance attainment occurred at the smallest dimension; this is a pattern that was not previously documented under matched, multi-seed validation. We attribute this to FedAvg interacting with the loss landscape of quantum-derived features on small client partitions, contributing this finding and the framework as groundwork for quantum-aware federated aggregation. Full article

Integrated Energy Systems (IESs) are characterized by strong nonstationarity, seasonal heterogeneity, and stochastic fluctuations in electric, cooling, and heating loads, making accurate and reliable probabilistic forecasting challenging. To address this issue, this study proposes an Improved Osprey Optimization Algorithm-based TCN-BiLSTM-Quantile Regression (IOOA-TCN-BiLSTM-QR) model for short-term probabilistic forecasting of multi-energy loads. The model integrates Logistic chaotic mapping and Lévy flight to enhance the optimization capability of the Osprey Optimization Algorithm, while TCN and BiLSTM are employed to capture multiscale local features and long-term temporal dependencies, respectively. Quantile regression is further introduced to construct 90% prediction intervals and quantify forecasting uncertainty. Using year-round hourly operational data from a university IES, the proposed model achieves average PICP and PINAW values of 90.50% and 0.1513, respectively, indicating a favorable balance between interval coverage and width. For point forecasting, the model achieves test-set R² values of 0.9385, 0.9781, and 0.9731 for electric, cooling, and heating loads, respectively, with corresponding MAPE values of 2.61%, 2.67%, and 2.36%. Comparative and ablation experiments demonstrate the effectiveness of the integrated optimization and temporal feature extraction framework, while cross-dataset validation further confirms its applicability to different multi-energy load conditions. Overall, the proposed model provides accurate point forecasts and 90% prediction intervals with average coverage close to the nominal level, offering quantitative information for reserve capacity allocation, operational scheduling, and risk-aware decision-making in IESs. Full article

Efficient forward and inverse design of laminated non-pneumatic tires requires repeated evaluation of their load-bearing and tire–ground contact responses. The high-fidelity laminated beam–grounding analysis (LB-GA) formulation captures the coupled band and spoke mechanics but requires iterative solution of 18 differential equations with unknown regional boundaries. This study derives explicit reduced-order relations for vertical stiffness and average contact pressure by combining laminated curved-beam mechanics with double-sided compression-ring theory. Joint fitting to 500 high-fidelity LB-GA solutions yields the empirical screening criterion $R_{\mathrm{opt}}=N^{0.68}{n}^{0.83}>40$. In an additional set of 5000 independently generated cases spanning the investigated design and material domain, 97.68% of the cases satisfying this criterion have a maximum response error no greater than 10%. The explicit relations are subsequently used as a domain-masked mechanics constraint in a multi-fidelity physics-informed neural network trained with independent high-fidelity labels. With $5\times {10}^5$ labels, the model gives stiffness and pressure NRMSEs of 4.47% and 4.68%, whereas a data-driven model using $5\times {10}^6$ labels gives 5.00% and 8.37%. Model-preparation time decreases from 220.1 to 32.5 h. Multiobjective inverse design, a newly manufactured-tire experiment, and six reconstructed finite-element designs produce validation errors below 10%. The framework therefore enables accurate tire design with substantially fewer high-fidelity numerical labels. Full article

Background/Objectives: Accelerating population ageing has made community-based frailty prevention among older adults an important public health priority. Guided by the Reach, Effectiveness, Adoption, Implementation, and Maintenance Qualitative Evaluation for Systematic Translation (RE-AIM QuEST) framework, this mixed-methods study assessed the implementation outcomes and contextual factors of the multi-component Exercise, Fruit intake and Vitamin Supplementation (EFVF) program and identified factors relevant to its sustainability and potential adaptation to other settings. Methods: This implementation evaluation was embedded within a cluster-randomized controlled trial (ClinicalTrials.gov Identifier: NCT06225271) conducted across 14 community clusters in Wuzhong District, Suzhou, China, from March 2024 to April 2025. Quantitative data from participation records, implementation supervision records, and post-intervention follow-up questionnaires were combined with qualitative data from semi-structured interviews with 17 stakeholders, including 3 CDC administrators and 14 frontline primary care providers. Qualitative data were analyzed using a hybrid framework-guided inductive–deductive approach. Quantitative and qualitative findings were analyzed separately and integrated within the RE-AIM QuEST dimensions at the interpretation stage. Results: Of the 1380 eligible older adults in the intervention clusters, 475 (34.42%) initiated the EFVF program. Compared with 2248 eligible non-participants, participants showed similar sex distributions (p = 0.880) but differed significantly in age (p < 0.001), educational level (p = 0.002), and marital status (p < 0.001). Material incentives, health-seeking needs, community trust, and accessible venues facilitated participation, whereas agricultural work, family responsibilities, limited health literacy, and geographic barriers constrained Reach. Frontline providers perceived improvements in health awareness, dietary practices, exercise behaviors, and social interaction, while also reporting variation in participant responsiveness and limited maintenance of some behaviors after structured support ended. All seven intervention community health centers adopted the program, and 34 of 75 eligible frontline providers participated in delivery (45.33%). All 168 directly observed site-sessions met the prespecified core fidelity criteria, and 393 participants (82.74%) completed both intervention months. At follow-up, provider willingness for continued delivery varied, with 21.43% expressing unconditional willingness to continue the program. Conclusions: The EFVF program was implemented with high fidelity under trial-supported community primary care conditions, while frontline workload, behavioral maintenance, and organizational support remained important implementation considerations. These findings support feasibility under the study conditions but do not establish routine-care feasibility or scalability to other settings. Future adaptation should consider local population characteristics, staffing and workflow capacity, flexible delivery arrangements, and sustained organizational support. Multi-site pragmatic evaluation, including implementation cost assessment and longer-term maintenance, is needed before broader implementation can be established. Full article

The central vein sign (CVS) is a supportive imaging biomarker for multiple sclerosis (MS), but its manual assessment on susceptibility-weighted imaging with phase enhancement (SWIp) is time-consuming and reader-dependent. Existing automated methods rely on multi-contrast volumetric data and prerequisite lesion masks. We study a deliberately different formulation: a structured end-to-end pipeline operating on a single axial SWIp sequence with sparse slice-wise annotations and no prerequisite volumetric mask, combining anatomical tiling, tiled lesion detection, cross-slice reconstruction of lesion identity, and two complementary branches that independently assess the same lesion crops—direct CVS classification (Pathway A) and vein-presence gating, vein segmentation, and geometric centrality (Pathway B). A cohort of 64 clinical studies annotated by four readers was merged into a consolidated reference. The main contribution is an explicit decomposition of where such a pipeline fails. On an eight-study evaluation, the pipeline recovered 78 of 111 reference lesions (70.3%); lesion-level F1 was 0.667 for Pathway A and 0.706 for Pathway B, rising to 0.785 and 0.818 when restricted to detected lesions, and a paired test found no significant difference between the pathways ($p=1.00$). Detection is therefore the binding constraint, and a threshold sweep shows the associated overcounting of CVS-positive lesions is not separable from it by detector confidence alone. All operating thresholds were selected on these same eight studies, which were acquired on a single scanner, so the figures are single-centre development-set estimates rather than measures of clinical performance and are likely optimistic. Larger-cohort, acquisition-diverse independent evaluation is required before clinical use. Full article

Tree-induced single-phase high-impedance faults (THIFs) threaten the reliability of distribution networks because their fault currents typically fall below the operating thresholds of conventional overcurrent relays. Accurately identifying tree species based on fault records is crucial for implementing differentiated vegetation management; however, the limited size of sample sets obtained from field experiments leads to severe overfitting in deep learning models. This paper proposes a framework that integrates physics-guided and data-driven approaches. First, a branch utilizing manually designed physical features extracts time-domain and frequency-domain metrics, while a MiniRocket branch captures multi-resolution temporal patterns. A bootstrap-based stability selection procedure is employed to retain discriminative features, and a rigorous “leave-one-file-out” cross-validation (LOFO-CV) scheme is used to train a strongly regularized Ridge classifier. Additionally, ablation studies are conducted to compare the feature information content at sampling rates of 100 kHz and 10 kHz under the experimental conditions. Finally, a carbonization degradation index (CDI) threshold is proposed. Experiments involving 37 fault records across five tree species demonstrate that the Hybrid + Ridge method achieves a file-level F1 score of 97.3%, achieving a file-level F1 of 97.3%, higher than standalone handcrafted features and MiniRocket. Full article

Protected wetlands in the Mediterranean are vital biodiversity hotspots that face growing pressure from climate change and human activity. Traditional bird monitoring relies on professional field surveys, which, although highly standardized, are resource-constrained and limited in temporal frequency. Here, we present ValWet-Birds, a multi-source spatiotemporal dataset and fusion framework that harmonises and integrates professional counts, eBird citizen science registries, and passive acoustic monitoring (via a BirdNET classifier deployed on a Raspberry Pi 5 node) for three protected wetlands in Alicante, Spain (El Hondo, Santa Pola, and La Mata–Torrevieja), from 2010 to 2025. We normalize incompatible observation protocols into a unified monthly relative observation proportion target across 12 sub-regions and 394 taxa (349 resolved to species level after a taxonomic audit). Spatiotemporal prediction models are implemented using Multi-Layer Perceptron (MLP) and Long Short-Term Memory (LSTM) neural networks. A preliminary evaluation using non-matched test sets suggested a 42.7% reduction in Test Mean Squared Error for the LSTM under multi-source augmentation. Re-evaluating every source condition on an identical held-out set of professional-census observations shows that this benefit does not hold in aggregate: both baselines outperform naive temporal reference predictors by a wide margin, but adding eBird and BirdNET data does not reduce error relative to census-only training overall. The exception is conservation-relevant taxa (a 50-species subset cross-referenced against Annex I of the EU Birds Directive), for which multi-source augmentation reduces LSTM error by 29% and MLP error by 9%, plausibly because these less-common species have sparser census history to draw on. We report this reversal explicitly as a methodological finding in its own right. Finally, we describe the design and interactive user flows of the deployed web visualization platform Avistory, which presents observation summaries and model outputs; it should not be interpreted as a validated population-monitoring or conservation-decision system. Full article

Anomaly detection becomes considerably harder when labels are scarce and the data are described by several heterogeneous views. A handful of labelled samples rarely delineates the normal region well, and the unlabelled pool is itself often contaminated by anomalies, so treating unlabelled data as normal and feeding them into a one-class boundary constraint injects incorrect supervision. We address this with SMDL-SVDD, a semi-supervised multi-view support vector data description (SVDD) framework built on dictionary representation learning. The guiding idea is to let unlabelled samples act at the level of representation rather than the SVDD boundary, so no class assumptions are imposed on them. SMDL-SVDD jointly learns a synthesis and an analysis dictionary in each view to obtain sparse representations, and constrains the SVDD hypersphere using only the small sets of labelled normal and labelled anomalous samples. A graph Laplacian regulariser over all training samples preserves the local manifold structure, while a cross-view consistency term on the unlabelled samples exploits their geometric distribution and shares information across views. The view-specific decision functions are then fused into a single anomaly score. We solve the joint problem by alternating convex search and analyse its convergence. Across 23 anomaly detection tasks derived from six public datasets, SMDL-SVDD gives the highest area under the receiver operating characteristic curve (AUC) on 21, improving the mean AUC by 8.87–10.62 percentage points over single-view one-class methods, by 5.10–7.64 points over representative semi-supervised detectors and by 4.11–5.24 points over multi-view one-class methods. Significance, noise, label-ratio, parameter, convergence, ablation and runtime studies confirm that the gains are stable; the ablation shows that manifold regularisation contributes substantially to the improvement and acts complementarily to the cross-view consistency constraint. Full article

This paper presents a preliminary smartphone-based inertial measurement unit (IMU) system for recognizing three common motion types, namely walking, running, and cycling, under different carrying modes. The proposed system is implemented on an Android smartphone and uses accelerometer and gyroscope signals to support both offline evaluation and on-device recognition. The current dataset contains six motion-carrying combinations formed by three motion types and two carrying modes, with more than 15 data segments collected for each category. Each segment lasts approximately 40–60+ s and was collected on flat ground. The raw sensor streams were processed using a sampling rate of 50 Hz, an 8 s sliding window, and a 4 s step size. To reduce data leakage, session-level splitting was adopted in the offline experiments. A K-nearest neighbor (KNN) classifier was used as the baseline recognition model, and different sensor combinations and feature settings were compared. The best offline configuration, using fused accelerometer and gyroscope features with the enhanced feature set and K = 3, achieved an accuracy of 86.57% and a macro-F1 score of 88.42%. On-device tests further showed stable recognition performance under hand-held conditions, while performance decreased under an unseen briefcase-style carrying condition, especially for walking and running. These results indicate that the proposed system is feasible as a preliminary smartphone-based motion recognition pipeline. At the current stage, the study should be interpreted as a pilot-scale feasibility evaluation under limited and partially unbalanced multi-subject data conditions, rather than a strict subject-independent generalization study. The findings nevertheless highlight the importance of carrying-mode diversity and more rigorous cross-subject evaluation in future work. Full article

Wire Arc Additive Manufacturing (WAAM) based on Cold Metal Transfer (CMT) welding produces near-net-shape metallic components, but post-process machining remains essential for dimensional accuracy and surface quality. This study investigates the dry turning machinability of EN AW-5083 aluminium alloy parts produced by CMT-based WAAM, focusing on tool-holder temperature (T_th) as a relative thermal indicator and surface roughness. T_th was monitored with a K-type thermocouple embedded in the tool holder and an Arduino-based data acquisition unit, while surface roughness (R_a) was measured with a portable contact profilometer after each pass. Three machining parameters—cutting speed, feed rate, and depth of cut—were investigated using a Taguchi L18 orthogonal array, with significance assessed through S/N ratio analysis and ANOVA. Grey Relational Analysis (GRA) was applied for multi-response optimisation, benchmarked against published machinability data for wrought and WAAM-fabricated aluminium alloys. Surface roughness was governed predominantly by feed rate, while T_th was most strongly influenced by depth of cut. The combined Taguchi–GRA optimisation identified a parameter set minimising both responses simultaneously. Benchmarking against three literature sources showed that the surface roughness achieved for the investigated CMT-WAAM material falls within the range reported for wrought EN AW-5083 under comparable machining conditions. These findings provide practical guidance for post-process machining of WAAM-produced aluminium components in hybrid manufacturing chains. Full article

Accurate service life (SL) data for building components are a critical yet poorly constrained input for life cycle assessment (LCA). Reference service life (RSL) values published in national frameworks are typically derived from technical standards rather than empirical observation. This paper compares two independent, methodologically aligned empirical datasets for windows, heating systems, and roofs for Belgium and The Netherlands. SL distributions are estimated using Weibull survival analysis, selected with AICc model comparison, and differences between survival functions are assessed using a Bonferroni-corrected Log-Rank test. Estimates are benchmarked against Belgian and Dutch RSLs. Two sets of insights emerge. First, on SL estimates: robust cross-national divergence is found only for flat roofs, is sensitivity-dependent for windows and pitched roofs and could not be assessed for heating systems due to incompatible definitions across surveys. Second, on survey methodology: both datasets query a single renovation cycle, biasing SL estimates towards longer values, particularly for shorter-lived components, underscoring that definitional alignment and multi-cycle survey design are essential for valid cross-national comparison. Together with the empirical SL estimates, these methodological lessons, of broader relevance beyond the present datasets, provide practical input for LCA and renovation planning in Belgium and The Netherlands, and guide future survey-based SL research. Full article

Multi-label text feature selection aims to identify compact and discriminative feature subsets for documents associated with multiple correlated labels. In federated environments, this task is further complicated by high-dimensional search spaces, non-independent and non-identically distributed data, complex label dependencies, and the requirement that raw feature and label data remain local. To address these challenges, this paper proposes the Federated Anchor-Guided Frequency-Domain Optimization Algorithm (Fed-AGFDO), a federated multi-label text feature selection framework. At each client, the Anchor-Guided Frequency-Domain Optimization (AGFDO) module transforms the encoded search population into the frequency domain and combines random spectral exploration, elite-guided mixing, diversity-aware anchor screening, and multi-anchor direction aggregation to balance exploration and exploitation. Candidate feature-weight vectors are evaluated using a manifold-regularized fitness function that jointly considers sample structure, label-consensus regularization, sparsity regularization, and a soft global feature-weight consistency penalty. The server aggregates only local feature-weight vectors through sample-size-weighted aggregation followed by an exponential moving average update to smooth inter-round changes in the global feature weights while keeping raw data local. Experiments on eight multi-label text datasets demonstrate that Fed-AGFDO achieves strong classification and label-ranking performance with compact feature subsets. Under the representative subset settings, Fed-AGFDO improves Average Precision by up to $5.83\%$ over Fuzzy Federated Multi-Label Feature Selection (Fuzzy FMFS) and reduces Ranking Loss by up to $22.65\%$ relative to Federated Multi-Label Feature Selection (FMLFS). Parameter sensitivity and ablation analyses further demonstrate robustness across the examined parameter ranges and the complementary contributions of anchor guidance, diversity screening, and multi-anchor aggregation. These results indicate that Fed-AGFDO provides an effective and data-locality-aware solution for high-dimensional federated multi-label text feature selection. Full article