Search Results (808)

Lipid-rich matrices can protect probiotics by limiting direct exposure to moisture, oxygen, and gastrointestinal stressors. However, triglycerides are chemically dynamic during storage, and differences in fatty acid structure and oxidation susceptibility may determine whether they exert protective or detrimental effects. In this study, tricaprylin, triolein, and trilinolein were evaluated as post-drying lipid storage matrices for freeze-dried Lactiplantibacillus plantarum 124 during storage at 25 °C, and 37 °C was used to accelerate lipid oxidation. Survival during storage, fatty acid chain retention, oxidation-derived products, stress tolerance, qualitative confocal imaging of membrane-integrity patterns, and transcriptomic responses were analyzed. After 30 d at 25 °C, log reductions were 0.30, 0.24, and 0.24 log10 units in the tricaprylin, triolein, and trilinolein groups, respectively, compared with 0.60 log10 units in the oil-free control, defining a common protective window. Tricaprylin showed the highest fatty acid chain retention, whereas trilinolein exhibited the greatest loss and strongest oxidative change. GC–MS detected no representative oxidation-derived products in tricaprylin, while unsaturated triglycerides generated dimethyl azelate and methyl 9-oxononanoate. Cells recovered from the protective window showed improved acid and bile tolerance, with survival increasing from 30.48% to 53.43–55.65% and from 14.85% to 22.49–27.79%, respectively. Confocal imaging further showed better-preserved membrane-integrity patterns in the triglyceride-treated groups after acid exposure. Transcriptomics indicated enhanced transport, redox, DNA-related, and cell-envelope responses, accompanied by reduced translation and growth-associated metabolism. Within the experimental conditions tested, these results suggest that triglyceride-associated attenuation of viability loss varies among triglyceride systems and across storage stages, while excessive oxidation of unsaturated triglycerides may weaken probiotic stability. Full article

Short-term financial risk in intelligent trading systems is reflected not only in prices, trading volumes, and textual sentiment but also in infrastructure operating states, including server workload, device power consumption, network latency, and packet loss rate. We propose HSF-LLMNet, a hardware sensor and large language model semantic fusion network for jointly modeling external information shocks and infrastructure responses. A large language model extracts event category, sentiment polarity, risk intensity, and semantic uncertainty from financial texts. Reliability-aware temporal modeling handles sensor missingness, drift, and abnormal noise, while asynchronous soft alignment, bidirectional cross-attention, and reliability-aware gated fusion integrate irregular textual events with continuous hardware signals. The model jointly predicts market direction, realized volatility, and three-level risk over the subsequent 30 min. Experiments were conducted on eight Chinese A-share indices: the SSE Composite Index (000001.SH), SSE 50 Index (000016.SH), CSI 300 Index (000300.SH), STAR 50 Index (000688.SH), CSI 500 Index (000905.SH), CSI 1000 Index (000852.SH), Shenzhen Component Index (399001.SZ), and ChiNext Index (399006.SZ). The common observation period for market, textual, and hardware data extended from 1 March 2024 to 30 June 2025. After data cleaning, timestamp matching, and multimodal temporal alignment, 169,208 aligned asset–time prediction windows were retained for the 30 min forecasting task. Realized volatility was defined as the square root of the sum of squared one-minute log returns over the future 30 min interval. The three-level risk label was constructed from future realized volatility, absolute 30 min return, and liquidity stress, with all thresholds estimated exclusively from the training portion of each fold. A sample was labeled high risk when at least two of the three indicators exceeded their 85th-percentile thresholds or when any indicator exceeded its 95th-percentile threshold. It was labeled medium risk when, after excluding high-risk samples, at least two indicators exceeded their 60th-percentile thresholds or any indicator exceeded its 85th-percentile threshold; all remaining samples were labeled low risk. Results showed that HSF-LLMNet achieved an accuracy of 78.62%, a precision of 78.14%, a recall of 77.83%, an F1-score of 77.98%, an area under the receiver operating characteristic curve of 84.91%, and a Matthews correlation coefficient of 57.36% for directional prediction. For realized-volatility regression, the MAE, RMSE, MAPE, and $R^2$ were 0.0089, 0.0135, 9.21%, and 0.812, respectively. For high-risk-event warning, the mean effective warning time, defined as the interval between the first valid alarm and the corresponding event, was 15.37 min; the false-alarm rate and missed-alarm rate were 6.82% and 8.14%, respectively. Ablation experiments showed performance reductions after removing semantic encoding, sensor-reliability estimation, asynchronous alignment, bidirectional cross-attention, gated fusion, or multi-task learning. These results indicate that textual events and infrastructure operating states provide complementary information for quantitative risk analytics and fintech applications. Full article

Background: Streptococcus suis serotype 2 is a major zoonotic pathogen responsible for severe systemic infections in pigs and humans, including septicemia, meningitis, and high mortality outcomes. Cefquinome, a fourth-generation β-lactam antibiotic widely used in veterinary medicine, is commonly applied for the treatment of S. suis infections. However, optimized dosing strategies remain insufficiently defined, particularly under conditions of varying bacterial burden, inoculum size, and reduced susceptibility or resistance phenotypes. These factors may significantly alter pharmacodynamic responses and compromise the predictive value of conventional MIC-based approaches. Objectives: This study aimed to characterize the pharmacokinetics (PK) and pharmacodynamics (PD) of cefquinome against S. suis serotype 2 using an integrated ex vivo serum time-kill experiments and semi-mechanistic PK/PD modeling. A secondary objective was to evaluate optimized dosing regimens across different inoculum levels and susceptibility phenotypes, including a cefquinome-resistant mutant. Methods: Cefquinome pharmacokinetics following intramuscular administration at 2 and 4 mg/kg in piglets were described using a two-compartment model. Dose proportionality, exposure linearity, and clearance parameters were assessed. Ex vivo serum time-kill experiments were conducted using a parental strain and a cefquinome-resistant mutant (M1) under normal-inoculum (NI), high-inoculum (HI), and mutant/resistant (MS) conditions. A semi-mechanistic PK/PD model incorporating logistic bacterial growth, sigmoidal Emax killing, nutrient limitation, and a time-delay function was developed to describe dynamic bacterial responses. Model parameters (k₀, k_max, EC₅₀) were estimated using nonlinear least-squares regression (Scientist v2.0), and simulations were performed by integrating time-varying PK input functions. Results: Cefquinome demonstrated linear pharmacokinetics with dose-proportional increases in Cmax and AUC between 2 and 4 mg/kg, with comparable clearance across doses. Ex vivo studies revealed time-dependent antibacterial activity with a pronounced inoculum effect. Higher bacterial burdens significantly reduced bactericidal efficiency and promoted regrowth during declining drug exposure. No tested concentrations achieved ≥3-log₁₀ killing in HI or MS conditions, whereas the NI group achieved a maximal reduction of 3.5-log₁₀ CFU/mL. MIC values in serum and medium were consistent (0.03, 0.06, and 0.24 µg/mL for NI, HI, and MS, respectively), indicating minimal protein binding influence. The semi-mechanistic model accurately described observed bacterial dynamics (R² > 0.99; MSC > 1.5), capturing delayed drug effects, inoculum-dependent growth suppression, and regrowth phenomena. Growth rates were reduced under serum conditions, reflecting nutrient limitation. Importantly, inoculum size exerted a stronger impact on pharmacodynamic outcomes than resistance phenotype, as reflected by reductions in kmax and increases in EC₅₀ under HI conditions. Although %T>MIC exceeded conventional β-lactam targets (>40%) in most regimens, MIC-based indices poorly correlated with observed dynamic killing responses. Conclusions: Cefquinome exhibited time-dependent antibacterial activity against S. suis serotype 2, strongly modulated by inoculum size and reduced susceptibility. The developed semi-mechanistic PK/PD model provided robust prediction of bacterial time-kill behavior and outperformed MIC-based metrics in guiding dose optimization. Simulation results support 2 mg/kg every 24 h for normal infections and 2 mg/kg every 12 h for high-inoculum or less susceptible infections, emphasizing the value of model-informed dosing strategies for optimizing β-lactam therapy. Full article

Accurate crack width measurement is essential for the condition assessment of reinforced concrete (RC) bridges, yet most unmanned aerial vehicle (UAV) inspections remain limited to pixel-level observations that cannot be converted into reliable metric units without an external scale reference. This paper presents a lightweight, drone-agnostic UAV-mounted metrology system that enables standards-aligned metric crack width measurement directly from inspection imagery. The payload integrates a focusable diffractive optical element (DOE) red laser that projects a cross pattern of known angular geometry, three TF-Luna time-of-flight (ToF) distance sensors, and an ESP-WROOM-32 microcontroller that provides dual-rate sampling, Bluetooth Low Energy (BLE) streaming, and on-board logging. A two-stage calibration links the synchronized distance measurements to the physical length of the projected cross, yielding an image-specific pixel-to-millimeter scale that is applied to pixel-level crack widths obtained from a vision-based segmentation pipeline. The system is field-deployed on the Puente Huamani Bridge in Pisco, Peru, where measurements of 39 cracks classified under AASHTO MBEI condition states are compared against independent manual measurements by six inspectors. The proposed system reduces measurement variability across all condition states (CS), lowering the average coefficient of variation from 0.36 to 0.10 for fine CS1 cracks, from 0.27 to 0.11 for CS2, and from 0.22 to 0.07 for CS3. Cross-platform adaptability is demonstrated through an additional deployment on a DJI Matrice 350 RTK at the Low Level Bridge in Edmonton, Canada. The results indicate that the system provides a practical, low-cost, and scalable solution for repeatable, standards-aligned UAV-based bridge crack assessment. Full article

Inland container depot (ICD) truck schedules must absorb new orders, service delays, appointment changes, congestion, and port cut-offs without destabilizing an already executed plan. This study asks whether event-triggered local repair can be separated into an explicit business-rule audit and a learned ranking of feasible task–vehicle actions. The proposed decision-support framework connects a static baseline, candidate task chains, six modeled hard-feasibility predicates, a Transformer encoder trained with proximal policy optimization (Transformer-PPO), and discrete-event execution logs. A five-seed, 120-episode confirmation gave Transformer-PPO a held-out online completion proxy (α_online) of 0.3226 and reward of 110.58, compared with 0.2581 and 61.87 for the matched multilayer perceptron (MLP); deterministic rules and search remained competitive. An independent audit of 4,968,000 action cells across 552 decision states found no disagreement with an independently coded oracle for the implemented hard predicates, while a reward-weight screen exposed the expected efficiency-stability trade-off. Together with a rolling-horizon comparator and a three-scale by three-disturbance stress test, the evidence supports an auditable system-integration contribution, not a new generic reinforcement learning (RL) algorithm or universal performance superiority. Claims are limited to synthetic simulation-based decision support. Full article

Background: Systemic inflammation significantly impacts graft survival and clinical outcomes following living donor liver transplantation (LDLT). The systemic immune-inflammation index (SII), which integrates peripheral neutrophil, platelet, and lymphocyte counts, has shown prognostic value in various clinical settings, but its role in LDLT has not been thoroughly investigated. Methods: This retrospective cohort study included 378 consecutive adult patients with end-stage liver disease (ESLD) who underwent primary LDLT between March 2016 and February 2025. The SII was calculated as (neutrophil × platelet)/lymphocyte. Spearman correlation analysis was used to assess the relationships between the preoperative SII, neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) and clinical outcomes, including the preoperative Model for End-Stage Liver Disease (MELD) score, duration of mechanical ventilation, and lengths of stay in the intensive care unit (ICU) and hospital. Receiver operating characteristic (ROC) curve analysis determined the optimal cut-off values for predicting 30-day mortality, and the areas under the curve (AUROCs) were compared using the DeLong test. Multivariable logistic regression was used to identify independent predictors of 30-day mortality. Results: Among 378 analyzable patients, the 30-day mortality rate was 7.9% (30/378). Both the SII and the NLR were significantly higher in non-survivors than in survivors (SII: median 368.8 vs. 169.1, p < 0.001; NLR: 6.0 vs. 2.4, p < 0.001), whereas the PLR did not differ significantly. On ROC analysis for 30-day mortality, the NLR and the SII showed comparable discrimination (NLR AUROC = 0.729, 95% CI: 0.63–0.82; SII AUROC = 0.705, 95% CI: 0.60–0.80; DeLong p = 0.43), both exceeding the PLR (AUROC = 0.541). The optimal SII cut-off was 275 × 10⁹ cells/L (sensitivity 70.0%; specificity 69.5%). Patients with an SII ≥ 275 × 10⁹ cells/L had significantly lower 30-day survival than those below the cut-off (83.5% vs. 96.4%; log-rank p < 0.001). On multivariable logistic regression adjusting for age and MELD score as continuous variables, an SII ≥ 275 × 10⁹ cells/L remained an independent predictor of 30-day mortality (aOR = 3.78; 95% CI: 1.60–8.93; p = 0.002); the MELD score was also independently predictive (aOR = 1.04 per point; 95% CI: 1.01–1.08; p = 0.018). The association persisted when the SII was modelled continuously (aOR = 1.60 per unit log SII; 95% CI: 1.08–2.38; p = 0.020). At the 275 cut-off, sensitivity for 30-day death was 70.0% and the positive predictive value 16.5%. Conclusions: The preoperative SII and NLR are simple, inexpensive, CBC-derived inflammatory indices significantly associated with 30-day mortality after LDLT for ESLD. An elevated preoperative SII independently predicts early post-transplant mortality and may aid perioperative risk stratification, although it does not outperform the NLR. Full article

While low-impact development and retention strategies are widely studied in urban and agricultural contexts, a distinct knowledge gap remains regarding the quantitative evaluation of dispersed, natural microretention structures in small, ungauged, mountainous forested catchments under complex topographic conditions. To address this limitation, this study provides a novel quantitative assessment of how natural bioretention interventions—specifically arcuate deadwood log barriers, cascading reservoir systems, and strategic afforestation—influence runoff reduction and substrate infiltration dynamics. Focusing on the 4.57 km² basin of the Stankowice Stream in southwestern Poland, the research integrates field geodetic and hydrological measurements with Iszkowski’s empirical flow formulas and high-resolution digital elevation modeling (SCALGO platform). Delineation of 10 key subcatchments revealed that surface runoff potential is heavily concentrated within specific flow pathways rather than determined solely by subbasin area. In unit No. 9, deploying an arcuate arrangement of 19 deadwood logs achieved an 11% reduction in surface runoff (retaining 8662.50 m³), whereas coupling these log structures with a downstream cascading two-dam system significantly enhanced retention performance by establishing 79,065.68 m³ of depression storage and driving 264,066.16 m³ of subsurface infiltration. Furthermore, multi-scenario land use modeling demonstrated that transforming land cover to forest reduced surface runoff by over 70% in topographically steep subcatchments (e.g., unit No. 7). These findings demonstrate that effective flood mitigation in headwater catchments requires a systemic, targeted hybrid strategy combining decentralized bioretention with localized storage nodes, offering a transferable framework for sustainable regional water governance. Full article

Outdoor Internet of Things (IoT) sensor nodes require battery systems for which their behaviour and implementation limits are characterized under low-temperature and variableload conditions. This study documents a LiFePO₄ battery-pack prototype integrating a BQ29729DSET protection IC and a configured MAX17055ETB+T fuel gauge and descriptively compares its discharge runtime with that of a reference pack with a similar nominal capacity, comprising three parallel Samsung ICR18650-26H cells. Tests were conducted at −30 °C, 8 °C, and 25 °C under nominal load settings of 50, 100, and 200 mA. At 8 °C and 25 °C, the two configurations showed similar runtimes and nominal-current-based calculated capacities. At −30 °C, the LiFePO₄ assembly ran for 89.1 versus 66.0 h at 50 mA and 44.7 versus 37.2 h at 100 mA. An analysis based on the typical MCP1700 dropout characteristic bounds the portions of these LiFePO₄ runtimes recorded below the assumed regulation threshold at approximately 1.1 h and 0.9 h, respectively; even subtracting those complete intervals leaves positive differences of 33.3% and 17.7% relative to the reference runtimes. Complete current logs were unavailable; therefore, capacity and energy are reported only as nominal-current estimates. In the ICR18650-26H reference pack at −30 °C, the calculated capacity increased anomalously from 3302 to 4087 mAh as the nominal setting increased from 50 to 200 mA. The ICR cutoff remained above the estimated regulator-dropout thresholds, so dropout does not explain the anomaly; temperature, conditioning, and run-order effects cannot be excluded. Protection trip points and fuel-gauge accuracy were not experimentally verified. Our contribution is therefore reproducible design documentation combined with preliminary low-temperature runtime evidence rather than validation of a fully monitored and protected battery pack. Full article

Background/Objectives: Total en bloc spondylectomy (TES) remains one of the most invasive and selectively used procedures in spine oncology. Its role has become more selective in the modern era of stereotactic body radiotherapy, separation surgery, targeted systemic therapy, immunotherapy, and multidisciplinary cancer care. This study evaluated long-term survival, imaging-defined systemic disease burden, operative morbidity, patient-reported outcomes, and frailty-related variables after TES in a rare single-center cohort, with the goal of identifying selection-relevant survival and treatment-burden signals rather than developing a validated decision algorithm. Methods: We performed a retrospective single-center cohort study of consecutive adults who underwent TES for spinal tumors between 2011 and 2022. Of the 36 screened patients, 30 had sufficient clinical and survival data for analysis; patients without reliable survival or last-contact data were not included. Contrast-enhanced CT and PET-CT were reviewed for extraspinal metastases, lymphadenopathy, pleural effusion, and soft-tissue extension. Survival was analyzed using Kaplan–Meier methods, log-rank testing, and exploratory univariate Cox regression. Patient-reported outcomes included the Oswestry Disability Index (ODI) and SF-36 when available; frailty was summarized with the modified frailty index-5 (mFI-5) when component data were present. Results: The cohort included 13 men and 17 women with a mean age of 54.8 ± 15.2 years. At final follow-up, 18 patients had died, and 12 were alive. Five-year overall survival was approximately 76% in the full cohort. Extraspinal metastases were present in 72.2% of deceased patients compared with 8.3% of survivors and showed the clearest exploratory association with increased mortality (HR 3.46, 95% CI 1.23–9.78; p = 0.019). Metastatic disease demonstrated inferior survival compared with primary bone or soft-tissue tumors. Perioperative blood loss and transfusion burden were substantial but were not associated with survival in univariate analysis. ODI and SF-36 data were available only in small subsets and were therefore interpreted as descriptive signals of treatment burden. Conclusions: TES remains relevant in modern spine oncology, but only as an increasingly selective intervention. In this rare cohort, systemic disease burden, particularly extraspinal metastases, was the clearest selection-relevant survival signal, while blood loss, transfusion requirements, complications, and limited patient-reported outcomes illustrated substantial treatment burden. These findings do not establish a validated selection algorithm but support a contemporary decision threshold that integrates tumor biology, systemic disease status, anticipated margins, physiologic reserve, operative morbidity, and patient goals. Full article

Background: Patients with end-stage renal disease (ESRD) undergoing radical nephroureterectomy (RNU) for upper tract urothelial carcinoma (UTUC) represent a uniquely high-risk population. While prior studies have demonstrated worse postoperative outcomes among dialysis-dependent patients, no study has systematically compared oncological outcomes between peritoneal dialysis (PD) and hemodialysis (HD) modalities following RNU. This multicenter study evaluates whether dialysis modality (peritoneal dialysis versus hemodialysis) acts as an independent prognostic factor following radical nephroureterectomy. Methods: Using the Taiwan UTUC Collaboration Group Registry—a multicenter, nationwide database comprising 21 tertiary and regional medical centers —we identified 350 ESRD patients who underwent RNU for UTUC between September 1988 and December 2023. Patients were categorized by dialysis modality at the time of surgery: 310 on HD and 40 on PD. Propensity score overlap weighting was applied to account for baseline differences. Multivariate Cox proportional hazards and Fine-Gray competing risk regression models were utilized to evaluate overall survival (OS), cancer-specific survival (CSS), progression-free survival (PFS), and non-UTUC mortality. Results: After overlap weighting, PD was independently associated with significantly worse OS (HR = 2.34, 95% CI 1.29–4.25, p = 0.005) and PFS (HR = 2.06, 95% CI 1.16–3.66, p = 0.013) compared to HD and exhibited a trend toward worse CSS in univariate analysis (log-rank p = 0.094). Survival curve divergence between PD and HD was most pronounced from 12 to 24 months post-surgery onward. Notably, this survival disadvantage persisted despite PD patients being significantly younger (mean age 58.8 vs. 65.1 years) and receiving adjuvant chemotherapy more frequently (20.0% vs. 6.5%). PD was also independently associated with higher non-UTUC mortality on multivariate competing risk analysis (sHR = 2.17, 95% CI 1.22–3.88, p = 0.009). Conclusions: In this first multicenter systematic comparison of PD versus HD patients undergoing RNU for UTUC, PD modality was independently associated with worse OS and PFS and exhibited a trend toward worse CSS in univariate analysis, compared to HD. This survival disadvantage persists despite favorable baseline characteristics and higher rates of adjuvant chemotherapy. We hypothesize these outcomes may be driven by a dual vulnerability: impaired systemic tumor control and elevated non-cancer mortality following surgical disruption of the peritoneal environment. To mitigate these risks, prioritizing minimally invasive or retroperitoneal surgical approaches to preserve peritoneal integrity, combined with modality-specific multidisciplinary surveillance, is recommended. Full article

Globally, respiratory diseases such as asthma, chronic obstructive pulmonary disease (COPD) and pneumonia affect populations significantly, requiring early and accurate diagnosis for effective clinical management. Manual auscultation and expert interpretation are the common shortcomings in conventional diagnostic approaches, as they lead to time-consuming and inconsistent analysis. To address these limitations, an artificial intelligence-driven framework for respiratory disease classification using multi-spectral feature extraction and deep learning architectures is proposed to classify four different respiratory conditions: Asthma, COPD, Pneumonia and Healthy. The dataset is collected from Kaggle’s respiratory sound database and the COUGHVID V3 database, which together contain 322 Asthma signals, 746 COPD signals, 323 Pneumonia signals and 174 Healthy signals. Subsequently, the features are extracted using four different feature extraction techniques—Constant Q Transform (CQT), a Gammatone spectrogram, Mel-Frequency Cepstral Coefficients (MFCC) and Perceptual Linear Prediction (PLP)—and these extracted spectral representations are provided as inputs to various deep learning models such as a Deep Convolutional Neural Network (Deep CNN), a Temporal Attention Network (TAN) and an Autoencoder for automated feature learning and disease classification. The proposed framework is evaluated using several performance metrics, and the experimental results clearly indicate that the performance of the proposed classification framework strongly depends on the selection of spectral feature extraction techniques and deep learning models. Among all the evaluated combinations, it is evident that the Autoencoder model integrated with CQT features exhibited the best classification performance, with an accuracy of 98.72%, precision of 98.74%, recall of 98.72%, Matthews correlation coefficient (MCC) of 98.11%, Cohen’s kappa value of 98.10% and the least log loss of 0.025. The proposed artificial intelligence (AI)-enabled respiratory disease classification framework has demonstrated the ability to produce a reliable computer-aided diagnostic system which is suitable for smart healthcare applications and automated pulmonary disease screening. Full article

The characterization of geothermal reservoirs and their genetic mechanisms is critical for understanding geothermal system evolution and evaluating geothermal resource potential. The eastern Gushi Sag of the Weihe Basin hosts three Neogene sandstone geothermal reservoirs, including the Gaoling Group, Lantian–Bahe Formation, and Zhangjiapo Formation; however, their reservoir characteristics and genetic mechanisms remain poorly constrained. This study integrates geological structures, geothermal well logging, core petrophysical properties, and hydrochemical data to characterize reservoir conditions and establish a genetic model. The results show that the Neogene reservoirs are mainly composed of feldspathic sandstone, with the Lantian–Bahe Formation identified as the primary geothermal reservoir due to its moderate porosity, low permeability, large sandstone thickness, and favorable continuity. The geothermal field exhibits an average geothermal gradient of 3.35 °C/100 m with a south-to-north decreasing trend. Hydrochemical evidence suggests that geothermal fluids originate mainly from meteoric water recharged from the northern Qinling Orogenic Belt and paleo-sedimentary water, with deep faults and pore networks controlling fluid migration and accumulation. The Quaternary strata and Zhangjiapo Formation provide effective sealing conditions. This study reveals the coupled controls of thermal conditions, reservoir architecture, fluid circulation, and preservation on sandstone geothermal systems, providing insights into geothermal resource assessment, exploration strategy optimization, and the formation mechanisms of similar sedimentary basin geothermal systems. Full article

Cloud-integrated sensor and Internet of Things (IoT) systems enable scalable data storage, processing, and intelligent monitoring, but their distributed nature exposes network-flow and host-level data to unauthorized access, tampering, and cyberattacks. This study proposes a Weighted Symmetric Hashed Blockchain (WSHB) framework that integrates mutual-information-based feature weighting, deep-learning-based attack classification, AES-256-GCM authenticated encryption, cryptographic hashing, and permissioned-ledger logging. The framework was evaluated independently using HIKARI-2021 for network-based intrusion detection and ADFA-LD for host-based intrusion detection. A transparent comparative evaluation was conducted against LSTM and CNN–RNN baselines using identical data splits, preprocessing settings, input representations, hyperparameter-search budget, and repeated initialization seeds. The proposed WSHB-DNN classifier achieved macro-F1 scores of $0.6837$ on HIKARI-2021 and $0.7914$ on ADFA-LD, showing the strongest overall classification performance among the evaluated models. The cryptographic and permissioned-ledger components provide confidentiality protection, record-level authentication, integrity verification, and tamper-evident logging for confirmed attack-event records. These results demonstrate the potential of WSHB as a reproducible framework for attack classification and secure event logging in cloud-integrated sensor environments. Full article

Traditional traffic safety analysis has long suffered from a reliance on retroactive, sparse crash logs, which mathematically struggle to capture the highly stochastic and non-linear dynamics of real-time traffic streams, rendering proactive safety prevention difficult. To bridge this gap, this study introduces an innovative, data-driven Traffic Operational Risk (TOR) assessment framework that integrates unsupervised deep learning with extreme-value statistics to achieve continuous, proactive risk monitoring. By mapping macroscopic traffic flow parameters and microscopic aggressive driving behaviors (ADBs) onto a unified spatiotemporal grid, we deploy a Variational Autoencoder (VAE) to learn the continuous latent “safe traffic manifold”. On this basis, Extreme Value Theory (EVT) is introduced to mathematically calibrate a dynamic, robust safety frontier on a unified scale (0–100), effectively suppressing sensor noise. The evaluation results demonstrate that the VAE-EVT framework robustly quantifies dynamic operational risks, effectively overcoming the linear limitations of traditional surrogate models. Furthermore, spatial frequency mapping reveals that elevated operational risks inherently cluster at geometric bottlenecks, such as merge/diverge zones and sharp curves. This spatial aggregation elucidates a typical “High Risk, Low Crash” phenomenon primarily driven by driver compensatory behaviors. Crucially, the integration of a novel multidimensional risk decoupling mechanism successfully isolates micro-behavioral volatility from macro-flow degradation. By tracing this causal progression, the framework captures the mechanistic evolution of traffic breakdowns, securing a critical 10 to 15 min proactive pre-warning window before systemic crashes or congestion materialize. Ultimately, this methodology liberates risk assessment from retroactive crash logs, providing a mathematically rigorous paradigm for precision-guided highway safety management. Full article

Wearable biosensors generate continuous physiological data in smart Internet of Disease (IoD) environments. These data can support early disease detection and remote patient monitoring. However, wearable data are often noisy, sensitive, and distributed across different devices. This paper proposes a multimodal neuro-symbolic model to overcome these limitations and incorporates it into a secure Edge–Fog–Cloud framework for anomaly detection in smart healthcare applications. The proposed system integrates the semantic analysis of clinical text using Bio-ClinicalBERT with temporal numerical data using an LSTM-based model, creating a unified neuro-symbolic artificial intelligence (AI) pipeline. Initial data processing is performed at the Edge, whereas inference is carried out at distributed Fog nodes for low-latency anomaly detection. Model training is handled in the Cloud, and privacy-preserving federated learning (FL) is supported through Homomorphic Encryption (HomEnc) to facilitate collaborative model training without sharing raw patient data. A sharded Tangle ledger is also used, with transactions broadcast by the Fog nodes and validated in the Cloud to create tamper-evident transaction logs. Furthermore, Honey Encryption (HoneyEnc) is integrated into the Fog layer to enhance security against brute-force attacks. Experimental results show that the proposed framework achieved 99.22% accuracy and a 99.31% F1-score on the held-out test set, with bootstrap 95% confidence intervals of 98.96–99.47% for accuracy and 99.08–99.53% for the F1-score. It also reduced detection latency from 185 ms in the baseline setting to approximately 50 ms in the Fog-inference setting. The blockchain layer achieved approximately 500 Transactions Per Second (TPS), while higher throughput was observed under increased transaction load and shard parallelism. Because the evaluation is based on synthetic multimodal EHR-like data and controlled simulations, the reported findings should be interpreted as proof-of-concept internal validation rather than evidence of deployment-ready clinical generalizability; external validation using real wearable biosensor data, hospital IoMT streams, or public clinical datasets such as MIMIC-III/MIMIC-IV is required before clinical deployment. These results highlight the potential of the proposed system for secure data processing and trustworthy anomaly detection in smart healthcare environments. Full article