Search Results (1,734)

Modern AI systems can strategically misreport information when incentives diverge from truthfulness, posing risks for oversight and deployment. Prior studies often examine this behavior within a single paradigm; systematic, cross-architecture evidence under a unified protocol has been limited. We introduce the Strategy Elicitation Battery (SEB), a standardized probe suite for measuring deceptive reporting across large language models (LLMs), reinforcement-learning agents, vision-only classifiers, multimodal encoders, state-space models, and diffusion models. SEB uses Bayesian inference tasks with persona-controlled instructions, schema-constrained outputs, deterministic decoding where supported, and a probe mix (near-threshold, repeats, neutralized, cross-checks). Estimates use clustered bootstrap intervals, and significance is assessed with a logistic regression by architecture; a mixed-effects analysis is planned once the per-round agent/episode traces are exported. On the latest pre-correction runs, SEB shows a consistent cross-architecture pattern in deception rates: ViT $80.0\%$, CLIP $15.0\%$, Mamba $10.0\%$, RL agents $10.0\%$, Stable Diffusion $10.0\%$, and LLMs $5.0\%$ (20 scenarios/architecture). A logistic regression on per-scenario flags finds a significant overall architecture effect (likelihood-ratio test vs. intercept-only: ${\chi }^2\left(5\right)=41.56$, $p=7.22\times {10}^{-8}$). Holm-adjusted contrasts indicate ViT is significantly higher than all other architectures in this snapshot; the remaining pairs are not significant. Post-correction acceptance decisions are evaluated separately using residual deception and override rates under SEB-Correct. Latency varies by architecture (sub-second to minutes), enabling pre-deployment screening broadly and real-time auditing for low-latency classes. Results indicate that SEB-Detect deception flags are not confined to any one paradigm, that distinct architectures can converge to similar levels under a common interface, and that reporting interfaces and incentive framing are central levers for mitigation. We operationalize “deception” as reward-sensitive misreport flags, and we separate detection from intervention via a correction wrapper (SEB-Correct), supporting principled acceptance decisions for deployment. Full article

Background: The current polio epidemiology in Pakistan poses a unique challenge for global eradication, with polio transmission dynamics influenced by regional variations in immunity and disparities in immunization coverage. This study assesses the immunity level for all three poliovirus types among children aged 6–11 months in polio high-risk regions of Pakistan. Methods: Four consecutive rounds of cross-sectional serological surveys were conducted in polio high-risk areas of Pakistan between November 2016 and October 2023. Twelve high-risk areas were covered in the first three rounds of the survey, while 44 high-risk areas were covered in the fourth round. 25 clusters from each geographical stratum were selected utilizing probability proportional to size. Results: Across the four rounds of the survey, 32,907 children aged 6–11 months from 2084 clusters and 32,371 households were covered. Comparative analysis across the survey rounds showed that seroprevalence of poliovirus type 1 was high in provinces (>95%), albeit consistently lower in Balochistan (going down to 89.7% in Round 4). Type 2 seroprevalence was significantly lower and more heterogeneous, from 34.6% in Sindh to 83.4% in Punjab, with sharp declines by round 4, particularly in Balochistan (40.4%). Type 3 seroprevalence was overall high (>94% in Punjab, Sindh, and KPK) but dropped in the last round, while Balochistan exhibited continually lower immunity (81.1%). Conclusions: The findings reflect the variations in population immunity to poliovirus in the country, with notable fluctuations over the years. The gaps in type 2 immunity over time and consistently lowest in Balochistan highlight the need for continued monitoring of immunity levels and adaptable vaccination strategies. Full article

Background/Objectives: This study aims to examine the relationships between 24-h movement guideline (24HMG) adherence and macronutrient intake, as well as assess dose–response relationships between 24-h movement behaviors and macronutrient intake among students aged 6–17 years. Methods: The study included 3624 participants aged 6 to 17 years from four rounds (2004–2011) of the Chinese Health and Nutrition Survey (CHNS). Participants’ 24-h movement behaviors and dietary intakes were evaluated. Results: Adherents to physical activity (PA) guideline had higher carbohydrate, fat, and protein intake (all p < 0.05). Those following the screen time (ST) guideline had a higher percentage of dietary energy intake (E%) from carbohydrates but a lower percentage from fat (all p < 0.05). Sleep (SLP) guideline adherents demonstrated lower protein intake and E% (all p < 0.05). PA guideline adherents were less likely to exceed carbohydrate Dietary Reference Intakes (DRIs) (OR = 0.83, 95% CI: 0.69–0.99), but more likely to surpass fat DRIs (OR = 1.20, 95% CI: 1.02–1.40). ST guideline adherents were more likely to exceed carbohydrate DRIs (OR = 1.32, 95% CI: 1.11–1.56) and less likely to surpass fat DRIs (OR = 0.78, 95% CI: 0.68–0.91). Dose–response analyses showed that moderate-to-vigorous physical activity (MVPA) and ST had positive linear associations with carbohydrate intake below DRIs. ST also showed positive linear associations with fat intake above DRIs. MVPA showed a nonlinear relationship with fat intake above DRIs. Conclusions: Among Chinese children and adolescents aged 6–17 years, those who meet the PA guideline should be cautious about the risk of excessive fat intake, while those adhering to the ST guideline should be aware of the risk of excessive carbohydrate intake in their daily diet. For promoting health and maintaining balanced macronutrient intake, MVPA should range from 60 to 90 min per day. This study underscores the importance of adjusting macronutrient intake according to levels of 24-h movement behaviors, especially MVPA and ST. Full article

Background/Objectives: Odontogenic keratocysts are benign cysts originating from remnants of the dental lamina, rarely showing peripheral (gingival) localization. In this study, we compiled data on the peripheral variant by reviewing the literature and presenting a new case to establish criteria for accurate differential diagnosis and treatment. Methods: A systematic literature review was conducted following the PRISMA flowchart, leading to the collection of existing data on peripheral odontogenic keratocyst. In addition, we present a new case of a 68-year-old female patient referred to our attention for an asymptomatic swelling of the mandible in the premolar area. Radiographic examination revealed a round radiolucency with well-defined borders located between teeth #4.3 and #4.4, surgically removed and diagnosed as a peripheral (gingival) keratocyst. Results: Including the herein described new case, 37 cases were reviewed from data literature showing occurrence in the mandible (43.2%) and maxilla (46%)—with 10.8% of cases not stated-, with an age range of 14–83 year old, recurrence rate of 12.5–13.6% (total recurrences/total cases) and median follow-up time of 19 months. Conclusions: Data from literature highlights the rarity of odontogenic keratocyst with peripheral (gingival) localization, which can be misleading for differential diagnosis, emphasizing the necessity of histopathological examination as the definitive diagnostic tool for all the cystic lesions of the jaws. The absence of pathognomonic clinical and radiological features, combined with the potential for extraosseous manifestation of odontogenic lesions with high recurrence rates, underscores the importance of complete excision to ensure proper healing and prevent recurrence of odontogenic keratocyst. Full article

The rapid growth of Internet of Medical Things (IoMT) devices offers promising avenues for real-time, personalized healthcare while also introducing critical challenges related to data privacy, device heterogeneity, and deployment scalability. This paper presents E-DMFL (Enhanced Distributed Multimodal Federated Learning), an Enhanced Distributed Multimodal Federated Learning framework designed to address these issues. Our approach combines systems analysis principles with intelligent model design, integrating PyTorch-based modular orchestration and TensorFlow-style data pipelines to enable multimodal edge-based training. E-DMFL incorporates gated attention fusion, differential privacy, Shapley-value-based modality selection, and peer-to-peer communication to facilitate secure and adaptive learning in non-IID environments. We evaluate the framework using the EarSAVAS dataset, which includes synchronized audio and motion signals from ear-worn sensors. E-DMFL achieves a test accuracy of 92.0% in just six communication rounds. The framework also supports energy-efficient and real-time deployment through quantization-aware training and battery-aware scheduling. These results demonstrate the potential of combining systems-level design with federated learning (FL) innovations to support practical, privacy-aware IoMT applications. Full article

Maritime Autonomous Surface Ships (MASSs) raise safety and regulatory challenges that extend beyond technical reliability. This study builds on a published system-theoretic process analysis (STPA) of degraded operations that identified 92 loss scenarios. These scenarios were reformulated into a two-round Delphi survey with 20 experts from academic, industry, seafaring, and regulatory backgrounds. Panelists rated each scenario on severity, likelihood, and detectability. To avoid rank reversal, common in the Risk Priority Number, an adjusted index was applied. Initial concordance was low (Kendall’s W = 0.07), reflecting diverse perspectives. After feedback, Round 2 reached substantial agreement (W = 0.693, χ² = 3265.42, df = 91, p < 0.001) and produced a stable Top 10. High-priority items involved propulsion and machinery, communication links, sensing, integrated control, and human–machine interaction. These risks are further exacerbated by oceanographic conditions, such as strong currents, wave-induced motions, and biofouling, which can impair propulsion efficiency and sensor accuracy. This highlights the importance of environmental resilience in MASS safety. These clusters were translated into five action bundles that addressed fallback procedures, link assurance, sensor fusion, control chain verification, and alarm governance. The findings show that Remote Operator competence and oversight are central to MASS safety. At the same time, MASSs rely on artificial intelligence systems that can fail in degraded states, for example, through reduced explainability in decision making, vulnerabilities in sensor fusion, or adversarial conditions such as fog-obscured cameras. Recognizing these AI-specific challenges highlights the need for both human oversight and resilient algorithmic design. They support explicit inclusion of Remote Operators in the STCW convention, along with watchkeeping and fatigue rules for Remote Operation Centers. This study provides a consensus-based baseline for regulatory debate, while future work should extend these insights through quantitative system modeling. Full article

The accurate determination of the equilibrium moisture content (EMC) of gel-related powdery samples requires strictly controlled conditions and a long time period. In this study, the adsorption and desorption isotherms of two fructooligosaccharide (FOS) powders and three inulin powders were determined using a dynamic moisture sorption analyzer at 0.1–0.9 water activity (a_w) and 20–35 °C, respectively. The adsorption and desorption isotherms all exhibited type IIa sigmoidal curves; the desorptive isotherm was smooth, the FOS adsorption curves had three inflection points, and the inulin adsorption curves had five inflection points. Large hysteresis between the adsorption and desorption isotherms occurred at 0.1–0.7 a_w for FOS and 0.1–0.6 a_w for inulin. Seven equations, Boquet, Ferro–Fontan, Guggenheim–Anderson–de Boer (GAB), Generalized D’Arcy and Watt (GDW), modified GAB (MGAB), Peleg, and our developed Polynomial, were found to fit the isotherms of the FOS and inulin samples; for adsorption, the best equations were Ferro–Fontan and GDW, and for desorption, the best equations were Polynomial and MGAB. The GDW and MGAB equations could not distinguish the effect of temperature on the isotherms, while the Polynomial equation could. The mean adsorptive monolayer moisture content (M₀) values in FOS and inulin samples were predicted as 7.29% and 7.94% wet basis, respectively. The heat of moisture sorption of FOS and inulin approached that of pure water at about 32.5% and 22.5% wet basis (w.b.) moisture content (MC), respectively. Fourier Transform Infrared Spectroscopy (FTIR) showed that the peaks in inulin with absorbance values above 0.52 and in FOS with absorbance values above 0.35 were at 1020, 1084, and 337 cm⁻¹; these could represent the amorphous structure (primary alcohol C-OH), C-O group, and hydroxyl functional group, respectively. Microscopic structure analysis showed that inulin powder particles were more round-shaped and adhered together, resulting in hygroscopic and sticky characteristics, with a maximum equilibrium moisture content (EMC) of 34% w.b. In contrast, the FOS powders exhibited irregular amorphous particles and a maximum EMC of 60% w.b. As hydrogels, 3–10% FOS or inulin addition reduced the peak, trough, final, breakdown, and setback viscosities of rice starch pasting, but increased the peak time and pasting temperature. FOS addition gave stronger reduction in the setback viscosity and in amylose retrogradation of rice starch pasting than inulin addition. The differential scanning calorimeter (DSC) showed 3–10% FOS addition reduced the amylopectin aging of retrograded paste of rice starch, but 5–7% inulin addition tended to reduce. These results suggest that FOS and inulin have strong hygroscopic properties and can be used to maintain the freshness of starch-based foods. These data can be used for drying, storage, and functional food design of FOS and inulin products. Full article

Background/Objectives: Antimicrobial resistance (AMR) is a major global health threat, with children at higher risk due to developmental differences in drug metabolism, limited treatment options and inappropriate antibiotic use. Pediatric antimicrobial stewardship programs (ASPs) face implementation challenges, often relying on adult-based guidelines and limited pediatric-specific evidence. This study aimed to identify and prioritize the most critical areas for pediatric ASP intervention development through a structured, multi-round Delphi consensus process with experts in antimicrobial stewardship and infectious diseases. Method: A four-round modified Delphi process was conducted to identify and prioritize key pediatric ASP interventions. Experts in antimicrobial stewardship and infectious diseases were recruited through an existing clinical trial. Using an iterative survey and in-person discussions, experts provided input on priority areas, which were thematically grouped and refined across rounds. Structured feedback supported real-time refinement and consensus-building. Results: Twenty experts participated in the process, generating 25 priority items in Round 1 through open-ended responses. These were narrowed to seven key priorities through structured voting and discussion. The final items were clustered into three intersecting themes: Care Settings, Prescriptions, and Strategies. Care Settings focused on high-impact areas such as outpatient clinics and intensive care units, where misuse is common and/or care is complex. The prescriptions theme prioritized shorter durations and narrow-spectrum agents. The strategy theme highlighted the need for outcome-based metrics, improved diagnostic stewardship, and routine tracking of patient outcomes to guide and assess stewardship efforts. Conclusions: This expert consensus identified key priorities for pediatric ASPs, providing a foundation for future interventions. Findings can be used to inform policy and practice, improving the appropriate use of antimicrobials in pediatrics and combating AMR. Full article

Background/Objectives: Impaired balance and general mobility are common complications of Parkinson‘s disease (PD) and are largely caused by bradykinesia and hypokinesia. Although previous studies have shown that patients can increase the speed and amplitude of movement with training, apathy, which is also common among people with PD, reduces this prospect. Training with light pods was originally developed for athletes to enhance agility in a way that is motivating. However, this type of training could be ideal for individuals with PD and possibly reduce bradykinesia and its effects. This study used a longitudinal interventional design without a control group to explore the effects of a four-week agility training with light equipment on balance and general mobility in patients with PD, as well as to assess motivational properties. Methods: Seven individuals with PD of the motor subtype “akinetic–rigid” participated in this study. Each participant received training three times per week for four weeks. The training session consisted of five rounds; in each round, participants had to turn off 20 lights. Measurements were performed one and a half weeks before training, at the beginning of the program, and at the end of the program. Balance was assessed with Mini-BESTest, general mobility with Timed Up and Go (TUG), transfer skills with 5× Sit to Stand, walking speed with the 10 m walking test, and the ability to turn on a spot with the 360° Turn Test. Motivational aspects of training were assessed after each training session, with scoring on a scale of 0–10. Results: The training significantly improved overall balance (p < 0.001), especially reactive postural control, sensory orientation, and dynamic gait, while anticipatory balance remained unchanged. Turning ability improved, but mobility, transfer ability, and walking speed did not. Motivation remained consistently high across participants. Conclusions: A four-week light-based agility training program can improve balance and turning ability in people with PD and appears to be motivating. However, no clear effects were found for general mobility, transfer skills, or walking speed. Given the small sample size and absence of a control group, these findings should be interpreted with caution. Full article

Background: Healthcare 5.0 represents a shift toward intelligent, human-centric care systems. Intensive care units generate vast amounts of data that require real-time decisions, but current decision support systems lack comprehensive frameworks for safe integration of artificial intelligence. Objective: We developed and validated the Learn–Predict–Monitor–Detect–Correct (LPMDC) framework as a methodology for systematic artificial intelligence integration across the critical care workflow. The framework improves predictive analytics, continuous patient monitoring, intelligent alerting, and therapeutic decision support while maintaining essential human clinical oversight. Methods: Framework development employed systematic theoretical modeling integrating Healthcare 5.0 principles, comprehensive literature synthesis covering 2020–2024, clinical workflow analysis across 15 international ICU sites, technology assessment of mature and emerging AI applications, and multi-round expert validation by 24 intensive care physicians and medical informaticists. Each LPMDC phase was designed with specific integration requirements, performance metrics, and safety protocols. Results: LPMDC implementation and aggregated evidence from prior studies demonstrated significant clinical improvements: 30% mortality reduction, 18% ICU length-of-stay decrease (7.5 to 6.1 days), 45% clinician cognitive load reduction, and 85% sepsis bundle compliance improvement. Machine learning algorithms achieved an 80% sensitivity for sepsis prediction three hours before clinical onset, with false-positive rates below 15%. Additional applications demonstrated effectiveness in predicting respiratory failure, preventing cardiovascular crises, and automating ventilator management. Digital twins technology enabled personalized treatment simulations, while the integration of the Internet of Medical Things provided comprehensive patient and environmental surveillance. Implementation challenges were systematically addressed through phased deployment strategies, staff training programs, and regulatory compliance frameworks. Conclusions: The Healthcare 5.0-enabled LPMDC framework provides the first comprehensive theoretical foundation for systematic AI integration in critical care while preserving human oversight and clinical safety. The cyclical five-phase architecture enables processing beyond traditional cognitive limits through continuous feedback loops and system optimization. Clinical validation demonstrates measurable improvements in patient outcomes, operational efficiency, and clinician satisfaction. Future developments incorporating quantum computing, federated learning, and explainable AI technologies offer additional advancement opportunities for next-generation critical care systems. Full article

Federated Learning (FL) has been proposed as a new machine learning paradigm to ensure data privacy by training the model in a decentralized manner. However, FL is challenged by device heterogeneity, asymmetric data contribution, and imbalanced datasets, which complicate system control and hinder performance due to long waiting times for aggregation. To tackle the FL challenges, we propose Adaptive Time Threshold Client Selection using DRL (ATCS-FL) to adjust the time threshold (α) in each communication round based on computing and resource capacity of each device and the volume of data updates. The Double Deep Q-Network (DDQN) model determines the appropriate α, according to the variations in local training time that achieves performance improvement alongside latency reduction. Based on the α, the server selects a subset of clients with adequate resources that can finish training within the α for participating in the training process. Our approach dynamically adjusts the α and adaptively selects the number of clients, effectively mitigates the impact of heterogeneous training speeds and significantly enhances communication efficiency. Our experiment utilizes CIFAR-10 and MNIST benchmarked datasets for image classification training with convolutional neural networks across non-IID distributed levels in FL. Specifically, ATCS-FL demonstrates performance improvement and latency reduction of 77% and 75%, respectively, compared to FedProx and FLASH-RL. Full article

Information announcement is the process of propagating and synchronizing the information of Computing Resource Nodes (CRNs) within the system of the Computing Networks. Accurate and timely acquisition of information is crucial to ensuring the efficiency and quality of subsequent task scheduling. However, existing announcement mechanisms primarily focus on reducing communication overhead, often neglecting the direct impact of information freshness on scheduling accuracy and service quality. To address this issue, this paper proposes a hierarchical and clustering-based announcement mechanism for the wide-area Computing Networks. The mechanism first categorizes the Computing Network Nodes (CNNs) into different layers based on the type of CRNs they interconnect to, and a top-down cross-layer announcement strategy is introduced during this process; within each layer, CNNs are further divided into several domains according to the round-trip time (RTT) to each other; and in each domain, inspired by the “Six Degrees of Separation” concept from social propagation, a RTT-aware fast clustering algorithm canopy is employed to partition CNNs into multiple overlap clusters. Intra-cluster announcements are modeled as a Traveling Salesman Problem (TSP) and optimized to accelerate updates, while inter-cluster propagation leverages overlapping nodes for global dissemination. Experimental results demonstrate that, by exploiting shortest path optimization within clusters and overlapping-node-based inter-cluster transmission, the mechanism is significantly superior to the comparison scheme in key indicators such as convergence time, Age of Information (AoI), and communication data volume per hop. The mechanism exhibits strong scalability and adaptability in large-scale network environments, providing robust support for efficient and rapid information synchronization in the Computing Networks. Full article

The ability to walk is essential in daily life, making walking outcomes key measures in clinical practice. This study aims to develop and validate two novel stride length model-based approaches for total distance estimation using a smart insole. Eight participants wore a pair of smart insoles. For a period of six minutes, each participant walked back and forth on a predefined 20 m pathway, and the numbers of round trips and strides taken were counted. Two stride length estimation approaches based on the director coefficients of acceleration data (Approach 1) and dynamic time warping (Approach 2) using smart insoles were used. The median accuracies of the total distance using Approach 1 are 98.92% [1.24%] (ICC = 0.992) and 98.69% [2.44%] (ICC = 0.994) for the right and left sides, respectively. For Approach 2, the average accuracies are 98.95% [0.18%] (ICC = 0.996) for the right side and 99.03% [0.14%] (ICC = 0.991) for the left side. The Mann–Whitney U test shows no statistically significant difference between the actual distance and smart insole for the total distance covered. Furthermore, there is no statistically significant difference between Approach 1 and Approach 2 for stride length. Although the sample size was small, the estimated total distance using the novel model-based algorithms appears to be accurate in comparison to the actual total distance. Full article

Wireless Sensor Networks (WSNs) consist of numerous battery-powered sensor nodes that operate with limited energy, computation, and communication capabilities. Designing routing strategies that are both energy-efficient and attack-resilient is essential for extending network lifetime and ensuring secure data delivery. This paper proposes Adaptive Federated Reinforcement Learning-Hunger Games Search (AFRL-HGS), a Hybrid Routing framework that integrates multiple advanced techniques. At the node level, tabular Q-learning enables each sensor node to act as a reinforcement learning agent, making next-hop decisions based on discretized state features such as residual energy, distance to sink, congestion, path quality, and security. At the network level, Federated Reinforcement Learning (FRL) allows the sink node to aggregate local Q-tables using adaptive, energy- and performance-weighted contributions, with Polyak-based blending to preserve stability. The binary Hunger Games Search (HGS) metaheuristic initializes Cluster Head (CH) selection and routing, providing a well-structured topology that accelerates convergence. Security is enforced as a constraint through a lightweight trust and anomaly detection module, which fuses reliability estimates with residual-based anomaly detection using Exponentially Weighted Moving Average (EWMA) on Round-Trip Time (RTT) and loss metrics. The framework further incorporates energy-accounted control plane operations with dual-format HELLO and hierarchical ADVERTISE/Service-ADVERTISE (SrvADVERTISE) messages to maintain the routing tables. Evaluation is performed in a hybrid testbed using the Graphical Network Simulator-3 (GNS3) for large-scale simulation and Kali Linux for live adversarial traffic injection, ensuring both reproducibility and realism. The proposed AFRL-HGS framework offers a scalable, secure, and energy-efficient routing solution for next-generation WSN deployments. Full article

Competitive programming has emerged as a critical benchmark for evaluating large language models (LLMs) in solving algorithmic problems under competitive conditions. Existing methods, such as the Sequential One-Agent Pipeline (SOP) approach, suffer from significant limitations, including the inability to effectively manage semantic drift across multiple stages, a lack of coordinated adversarial testing, and suboptimal final solutions. These issues lead to high rates of wrong answer (WA) and time-limit exceeded (TLE) errors, especially on complex problems. In this paper, we propose the Cross-Verification Collaboration Protocol (CVCP), a multi-agent framework that integrates symmetry detection, symmetry-guided adversarial testing, Round-Trip Review Protocol (RTRP), and Asynchronous Voting Resolution (AVR) to address these shortcomings. We evaluate our method on the CodeELO dataset, showing significant improvements in performance, with Elo Ratings increasing by up to 7.1% and Pass Rates for hard problems improving by as much as 1.8 times compared to the SOP baseline. Full article