Search Results (443)

Coronary slow flow (CSF) is delayed distal contrast transit on angiography without flow-limiting stenosis; disturbed sleep may impair vascular control through autonomic, endothelial, inflammatory, and metabolic pathways. We evaluated Pittsburgh Sleep Quality Index (PSQI)-based sleep quality in relation to TIMI frame count (TFC)-defined CSF. This cross-sectional study enrolled 307 adults with nonobstructive coronary arteries undergoing angiography for chest pain; PSQI referenced the preceding month, and CSF was defined by corrected TFC > 27 frames in any major vessel; 132 participants had CSF and 175 normal flow. Global PSQI score was higher in CSF (median 7.0 vs. 5.0) and poor sleep quality (score > 5) was more frequent (75.8% vs. 49.7%; both p < 0.001). The global score correlated with mean TFC overall but not within flow groups. Each PSQI point independently raised CSF odds, including after STOP-Bang adjustment, although discrimination was modest. Sleep latency and short duration raised CSF odds; poor efficiency did not. Poorer sleep quality was independently linked to this angiographic slow-flow phenotype, mainly differentiating flow categories rather than tracking frame-count burden. Because coronary microvascular function was not measured directly, these hypothesis-generating findings should prompt systematic sleep assessment and prospective studies integrating objective sleep measures with invasive coronary physiology. Full article

Tool-level attacks on Large Language Model (LLM) agents—poisoned tool descriptions, prompt injection, and capability misrepresentation—are universally effective, yet no existing defense provides comprehensive protection. We propose Architectural Intent Collapse (AIC), a formal framework capturing the systematic loss of communicative intent when text from heterogeneous sources is flattened into a single context window. Grounded as a novel instantiation of the Confused Deputy Problem, AIC reveals that the missing boundary is not permission but intent: the architecture cannot distinguish descriptive statements from prescriptive commands. We formalize AIC via an architectural collapse operator, introduce Intent Separation Degree (ISD) as a measurable metric, and develop a mechanism-based taxonomy of five intent-disguise attack types, including two previously undescribed (Conditional Latency and Inference Inducement). Experiments across 25 framework–model combinations (employing GPT-4o, Claude-4-Sonnet, Gemini-2.5-Pro, DeepSeek-V3, and Qwen3-32B as LLM backends) confirm that ISD degrades with description verbosity, strongly predicts defense effectiveness ($r=-0.97$), and is uniformly low across all current frameworks. Three root-cause defense principles are derived; one retains substantial protection against adaptive attackers. This research is useful for agent framework designers, security practitioners, and researchers seeking a principled understanding of why tool-level attacks succeed and how architectural defenses can address their root cause. Full article

Recent progress in intelligent, adaptive, and security-aware Software-Defined Wireless Sensor Networks (SDWSNs) is driven by the integration of Machine Learning (ML) with Software-Defined Networking (SDN) and Wireless Sensor Networks (WSNs). This systematic survey analyzes 46 peer-reviewed articles published between 2024 and 2025, providing a problem-oriented synthesis of ML-SDWSN research. Emphasizing security, routing, and performance optimization, with a particular focus on deployment architectures, the survey identifies three major trends: increased adoption of ensemble and Reinforcement Learning (RL) methods for security and adaptive control; broader implementation of edge-based ML to minimize inference latency; and greater emphasis on privacy-preserving techniques, especially Federated Learning (FL). The survey presents a structured taxonomy encompassing seven thematic areas: Distributed Denial-of-Service (DDoS) mitigation, Intrusion Detection Systems (IDSs), routing optimization, Quality of Service (QoS) management, privacy preservation, data integrity, and network-efficiency optimization. Findings are synthesized from over 120 experimental configurations reported in the literature. Due to substantial differences among the reviewed studies in terms of datasets, network topologies, hardware platforms, measurement definitions, and validation methodologies, the reported values are presented as descriptive cross-study aggregates rather than direct comparative benchmarks or formal effect-size estimates. Within these constraints, the survey identifies recurring trade-offs among accuracy, latency, scalability, and privacy. It provides evidence-based design considerations for researchers and practitioners. The survey also highlights eight critical research gaps, including limited multi-dataset validation, a lack of real-world deployments, insufficient scalability analysis, and the need for rigorous evaluation of RL-based SDWSN control. Full article

Advanced wireless sensor networks powered by artificial intelligence are increasingly required for applications demanding continuous monitoring, autonomous operation, reliable communication, and fast decision support. This systematic review examines recent work from 2023 to 2026 on energy-efficient, AI-enabled wireless sensor networks (WSNs) in mission-critical environments, with particular focus on power electronics and urban infrastructure systems. The authors synthesize a corpus of 50 DOI-indexed studies satisfying inclusion criteria that received qualitative thematic coding and comparative analysis. Other references were only cited to provide historical, methodological, or technical context and were not included in the systematic review corpus. As such, our results show that AI can improve WSN energy behaviour through routing and clustering, edge AI, reinforcement learning, fuzzy logic, metaheuristic optimization, and AI-based security. At the same time, energy efficiency cannot be treated as an isolated performance target. In mission-critical systems, security, latency, and reliability are closely interlinked requirements. The review concludes that future work should move away from optimizing protocols in isolation, and instead focus on building lightweight, explainable, secure, and field-tested AI-driven WSN architectures suited to real operational environments. Full article

The development of Motor Imagery (MI) Brain–Computer Interfaces (BCIs) is systematically constrained by low signal-to-noise ratios (SNRs), signal non-stationarity, and acute data scarcity. While complex Blind Source Separation (BSS) methods optimize signal clarity, their computational overhead introduces propagation delays that challenge real-time constraints. This study addresses this engineering trade-off by introducing a localized architectural framework to evaluate whether a lightweight pipeline operating without BSS (No-BSS) is sufficiently efficient for real-time control when compared against two BSS-equipped pipelines utilizing Independent Component Analysis (ICA) and Empirical Mode Decomposition (EMD). Validated across the BCI Competition IV Dataset 2A and the PhysioNet MI dataset, all three pipelines share an identical processing chain designed to maximize efficiency. To mitigate low SNRs, an Adaptive Laplacian spatial filter isolates neural intent across target sensorimotor electrodes (C3, C4, and Cz). Data scarcity is countered via a Gaussian noise injection data augmentation strategy, while session-to-session variability is addressed during feature extraction using Wavelet Packet Decomposition (WPD) paired with a Fisher Score criterion to dynamically isolate subject-specific time-frequency nodes. Redundant features are subsequently eliminated using a Genetic Algorithm (GA) before classification. Experimental evaluation reveals a distinct performance stratification: while the ICA (92.80%) and EMD (92.69%) pipelines yield the highest average accuracy for the PhysioNet dataset by isolating non-stationary and physiological noise, the No-BSS baseline (90.28%) remains the superior framework for the BCI Dataset 2A. Across all pipelines across both datasets, a stable classification hierarchy emerges wherein the Support Vector Machine (SVM) leads performance due to its maximum-margin decision boundary, followed by k-Nearest Neighbors (kNN), a modified EEGNet, and Decision Trees. The No-BSS baseline achieves classification accuracies highly competitive with its BSS counterparts while entirely bypassing their algorithmic overhead. Given the strict latency constraints of live BCI control loops, these findings establish the optimized No-BSS pipeline as a highly viable alternative for low-latency, real-time implementations. Full article

Olfactory enrichment is recommended for captive felids, but systematic evidence on practical implementation in multiple species and institutions is limited. We evaluated a simple, keeper-safe scent diffuser delivering spice-based odours to captive lions (Panthera leo), tigers (Panthera tigris), jaguars (Panthera onca), leopards (Panthera pardus), cheetahs (Acinonyx jubatus) and pumas (Puma concolor) at two zoological institutions in Johannesburg, South Africa. Behaviour was recorded during baseline and scent-exposure conditions in 23 individuals. Behavioural diversity (number of distinct behaviours per 15-min session) was the primary outcome. Engagement duration, latency to approach and individual response consistency were also quantified. Animals were exposed to a compound spice blend and four component spices (coriander, paprika, black pepper and garlic) and subsequently to decreasing concentrations of their most preferred scent. Behavioural diversity increased significantly during scent exposure at both institutions (1.9 times greater at Lory Park and 1.8 times greater at Johannesburg Zoo), with large effect sizes at both sites (Cohen’s d ≥ 1.2). Spice type did not significantly affect behavioural diversity, but individual animals showed consistent engagement preferences, approach latencies and response profiles during repeated exposures. Full odour concentrations showed significantly longer engagement durations than low concentrations at both institutions (approximately twofold at Lory Park and threefold at Johannesburg Zoo), but behavioural diversity did not differ significantly between concentration levels. No significant species differences were detected, indicating that assessment by individuals is more informative than species level predictions for enrichment planning. Spice-based diffusers provide a practical, low-cost enrichment tool suitable for multi-institutional implementation. Moreover, enrichment programmes should be evaluated and tailored at the individual animal level. Full article

The exponential growth of the Internet of Things (IoT) has generated massive data streams traditionally processed by centralized cloud architectures, which increasingly face latency, bandwidth, and privacy limitations. Shifting artificial intelligence to resource-constrained edge nodes, known as TinyML, offers a robust decentralized alternative, though it introduces severe memory, compute, and energy bottlenecks. To map this transition, a systematic literature review was conducted following PRISMA guidelines, analyzing peer-reviewed studies published between 2021 and 2026 across major databases. The analysis identifies primary architectural paradigms and evaluates the efficacy of state-of-the-art model compression techniques, such as quantization, pruning, and knowledge distillation. Furthermore, the findings reveal that hardware–software co-design and custom neural accelerators are crucial for overcoming operational bottlenecks, while also highlighting persistent security and privacy challenges in on-device learning. Ultimately, while deploying complex models on microcontrollers is increasingly viable, achieving optimal performance demands holistic optimization strategies. This review synthesizes current research gaps and provides a strategic roadmap to guide future interdisciplinary efforts toward resilient, energy-efficient, and secure next-generation intelligent edge systems. Full article

Small Language Models (SLMs) enable efficient deployment, but their limited parameter count constrains factual knowledge, particularly in low-resource languages like Kazakh. Integrating live web search can address this limitation, though its effectiveness is difficult to measure due to sparse in-language web indices and answer leakage during benchmarking. In this study, we systematically compare zero-shot parametric generation, in-language retrieval, and cross-lingual (translate-then-retrieve) web search using three 4B-parameter SLMs in both reasoning and non-reasoning modes. To evaluate factuality without search-engine leakage, we introduce machine-translated Kazakh versions of the FreshQA and DefAn benchmarks, and use GPQA as a Google-proof adversarial control. We also assess robustness across three prompt complexities, from simple JSON constraints to adversarial warnings that instruct the model to treat potentially unreliable context with caution. Finally, we propose a training-free, self-aware router that uses majority voting over repeated self-verification decisions to determine when to answer parametrically, when to search the web, and when to escalate to a more capable cloud model. Our results show that cross-lingual retrieval substantially outperforms in-language search on global factuality tasks, nearly doubling accuracy on FreshQA, while direct in-language search remains preferable for localized cultural queries. The choice of retrieval language depends on the task and does not always favor English. Additionally, cross-lingual retrieval is not consistently superior, because the best option depends on where relevant information is indexed. Pareto analysis indicates that cross-lingual search is on or near the optimal accuracy–latency frontier, adding minimal overhead compared to direct search. The router identifies which query types warrant retrieval, and as a system it tracks or exceeds always-search accuracy while issuing fewer searches and approaching the always-cloud ceiling at a fraction of its cost; per-query discrimination within a task family is weaker, which we quantify explicitly. Overall, this work offers a framework for optimizing and accurately measuring cross-lingual RAG pipelines in low-resource settings. Full article

The rapid integration of Large Language Models into high-stakes domains has elevated inference-time attacks into a primary security concern for production deployments. These attacks are adversarial techniques that exploit models exclusively through their input–output interface. The existing survey literature lacks a dedicated and structured treatment that jointly maps the attack surface and systematically evaluates the mitigation strategies developed against it. This paper addresses this gap through two original taxonomic contributions. First, LLM vulnerabilities are organized into a three-layer attack surface taxonomy stratified by lifecycle stage, establishing the theoretical primacy of the inference time category. Second, to directly address how these attacks can be mitigated, a defense taxonomy spanning three axes, namely prompt-level, inference-time, and training-time interventions, is proposed, within which 30 mitigation mechanisms published from 2024 onwards are systematically analyzed. Building on this taxonomy, an intersectional comparative analysis is conducted across three dimensions: defense-attack coverage, security-utility-latency tradeoffs, and white-box versus black-box applicability, in order to evaluate how effectively current mitigation strategies neutralize each attack category. These dimensions are further synthesized into a practitioner decision framework that maps deployment constraints to concrete defense configurations and identifies two structural coverage gaps that persist regardless of access level or latency budget. The resulting Defense-Attack Coverage Matrix demonstrates that no single defense mechanism provides comprehensive protection, and that robust deployment mandates layered, complementary strategies. The analysis further reveals that the fundamental unresolved tension limiting effective mitigation is the trade-off between adversarial robustness and model utility, with over-refusal and capability degradation constituting the primary practical barriers to deploying these defenses. Finally, open challenges related to multimodal attack surfaces, agentic LLM security, and the absence of standardized evaluation frameworks are identified, together with concrete future research directions. The taxonomies and analyses presented are intended to serve as an actionable reference for both researchers and practitioners tasked with mitigating inference-time attacks in secure LLM deployments. Full article

Industrial sensor-based fault diagnosis often requires continuous data acquisition, local data processing, and timely model inference on edge devices. Although deep learning-based diagnostic methods have achieved promising performance, many existing approaches rely on cloud-centered processing pipelines that introduce communication overhead and potential data privacy concerns. This paper presents a CPU–NPU heterogeneous edge fault diagnosis framework for industrial sensor data. The framework runs on an RK3588 local edge device and includes SQLite- and RingBuffer-based data management, sliding window generation, micro-batch construction, and model inference. The CPU is responsible for data access and buffering, preprocessing, and micro-batch preparation, while the NPU executes fault diagnosis models using the RKNN runtime environment. By performing inference locally, the framework reduces the continuous transmission of raw sensor data and supports real-time fault diagnosis under resource-constrained edge devices. Experimental results demonstrate high consistency between ONNX-based CPU inference and RKNN-based NPU inference after model conversion. Furthermore, the effects of different data input paths and micro-batch configurations are systematically evaluated. A cross-platform comparison between server-class CPU/GPU execution and embedded NPU deployment is also conducted in terms of latency, throughput, and energy efficiency. The results show that RingBuffer-based streaming input significantly reduces data access overhead, while the effectiveness of NPU acceleration depends on both model structure and micro-batch size. The cross-platform results further demonstrate the energy efficiency advantages of the RK3588 platform, making it more suitable for practical deployment in resource-constrained edge scenarios. These findings provide practical insights for deploying fault diagnosis models on heterogeneous edge devices. Full article

Digital markets increasingly coordinate prices, matches, orders, and allocations through automated systems whose decision–execution loops can close faster than humans can intervene. This article develops a microfounded framework in which latency is a temporal feasibility constraint that complements adjustment costs, information delay, costly information acquisition, queueing, and technological execution costs. The model distinguishes common latency, human intervention latency, and relative latency. Common latency is produced by platform and participant investment and affects welfare through the freshness of the state on which decisions are executed. Human intervention is represented by a smooth, task- and organization-specific probability q(L,z,s), derived from a distribution of completion times and modified by interface and organizational support. Human, hybrid, and algorithmic decision technologies differ in speed, accuracy, cost, and systematic misspecification risk. Relative speed is modeled as a strategic priority contest in which each intermediary’s best response depends on rivals’ investments, while platform rules determine the sensitivity and value of being first. The framework derives conditions for human-algorithm substitution, welfare-improving common-speed investment, socially excessive strategic speed investment, and welfare-enhancing batching or latency floors. It separates temporal from structural market distortions, integrates decision-technology quality and state freshness in a total-welfare function, and develops an incidence model that traces gains across heterogeneous users, intermediaries, and infrastructure owners. Robustness results cover diffusion, mean-reverting, jump, stochastic-volatility, and regime-switching state processes. An illustrative dynamic simulation, explicit scope conditions, and an operational empirical agenda show how the theory can be tested without claiming empirical calibration. The central contribution is a non-equivalence result: when latency enters the probability of successful intervention, shortening the decision window can change the technology and locus of marginal choice even when information, objectives, adjustment costs, and the substantive decision rule are held fixed. Full article

Air-quality forecasting models are often compared by architecture, although reported skill also depends on the pollutant, monitoring density, forecast horizon, predictor latency, validation design, and deployment objective. We conducted a systematic mapping review of 533 unique records published between 2000 and 15 June 2026; 409 met the forecasting eligibility criteria. The evidence was analysed in two layers: a metadata-derived map of the full corpus and a targeted full-text synthesis of representative studies. The non-exclusive metadata categories show that general machine learning or benchmark studies were most common ($n=207$), followed by recurrent deep learning ($n=109$), hybrid or decomposition methods ($n=60$), Transformer or attention models ($n=51$), tree ensembles ($n=48$), CNN/ConvLSTM models ($n=43$), classical statistical methods ($n=26$), graph neural networks ($n=13$), and physics-informed or CTM-coupled methods ($n=7$). These counts describe topical prevalence, not comparative effectiveness. The main contribution is a decision framework that links the forecasting setting to a defensible starting model, the evidence available for that model family, and the minimum validation needed to support temporal, spatial, or external generalisation. The synthesis favours transparent statistical and tabular baselines for short or sparse single-station records; spatial deep models only when network geometry and leave-site-out testing support them; and CTM-coupled postprocessing when operational physical fields are available at issue time. Diffusion and foundation models remain promising but unevenly validated for pollutant forecasting. A leakage-safe daily PM_2.5 case study in Naples illustrates the practical consequence: model rankings change with the metric, and every fitted model underestimates the highest 5% of concentrations. Full article

Loss-of-function (LoF) variants in the PRPS1 gene, encoding the phosphoribosyl pyrophosphate (PRPP) synthetase 1 enzyme, cause rare neurometabolic disorders historically viewed as discrete entities: nonsyndromic deafness (DFNX1), Charcot–Marie–Tooth disease type X5 (CMTX5), and Arts syndrome. A major clinical challenge is the temporal dissociation between early auditory failure and subsequent neurodegeneration, causing fragmented diagnostics. We systematically quantified this diagnostic latency and reconceptualized the disease spectrum through a molecular lens. A PRISMA-compliant systematic review identified 19 patients with genetically confirmed PRPS1 LoF variants, including our index case (c.362C>G) presenting a 15-year diagnostic delay. Kaplan–Meier analysis revealed sensorineural hearing loss manifested acutely (median 0 years; 95% CI: 0–1). In contrast, neurological deficits demonstrated a prolonged latency (median 3 years; 95% CI: 1–8), followed by ophthalmological signs (median 11.5 years). The median symptomatic delay was 3 years (range up to 19). We posit that this temporal dissociation reflects differential tissue vulnerability to intracellular ATP/GTP and NAD+ depletion caused by the primary enzymatic defect. Ultimately, DFNX1, CMTX5, and Arts syndrome represent a continuous PRPS1-related neurometabolic spectrum. Because targeted metabolic interventions (such as S-adenosylmethionine or nicotinamide riboside) have limited efficacy on advanced structural nerve damage, recognizing this early diagnostic window to initiate biochemical rescue prior to irreversible axonal degeneration is critical. Full article

With the rapid expansion of the Internet of Medical Things (IoMT), the risks of cybersecurity have increased exponentially in healthcare settings, exposing patients’ safety. Three fundamental issues that existing intrusion detection systems (IDS) are challenged by are: (1) limited cross-domain generalization, (2) high computation requirements not suitable for edge deployment, and (3) absence of systematic comparison between classical machine learning (ML) and deep learning (DL) approaches on IoMT-specific data. In this paper, we propose a multi-dataset evaluation framework that covers six models (Random Forest, XGBoost, DNN, CNN, LSTM, and CNN-LSTM) across three different datasets: WUSTL-EHMS-2020, Edge-IIoTset, and UNSW-NB15. We show that there is a scale-dependent pattern: classical ensemble methods work best when the data is small (F1 = 0.914 ± 0.013 on WUSTL-EHMS-2020); the proposed hybrid CNN–Ensemble framework performs best when the data is large (F1 = 0.968 ± 0.002 on UNSW-NB15 with 62.8% fewer features). The proposed framework achieves a total model size of 2.11 MB and an inference latency of 111.6 ms, with seven out of the top 15 discriminative features being patient vital signs, giving the first quantitative evidence that physiological data systematically contributes to IoMT attack detection, which is demonstrated through an explainability analysis using the SHAP approach. Cross-dataset generalization experiments across six transfer scenarios expose fundamental limitations in domain transfer, establishing an important baseline for future research. Full article

Background: Mesothelioma is a rare but highly aggressive malignancy strongly associated with asbestos exposure. Owing to its long latency and poor prognosis, its burden requires systematic evaluation. Methods: Data on prevalence, incidence, deaths, disability-adjusted life years (DALYs), and age-standardized rates were extracted from the Global Burden of Disease Study 2023. The estimated annual percentage change, Joinpoint regression, Das Gupta decomposition, and Nordpred forecasting were used to assess temporal trends, identify turning points, quantify demographic and epidemiological contributions, and project future burden through 2045. Results: From 1990 to 2023, the absolute burden of mesothelioma in China increased substantially. Prevalent cases rose by 189%, incident cases by 150%, DALYs by 98%, and deaths by 142%. Males consistently showed a higher burden than females, and the burden was concentrated mainly among middle-aged and older adults. The age-standardized prevalence rate and age-standardized incidence rate increased, whereas the age-standardized DALY rate and age-standardized mortality rate remained stable or declined slightly. Decomposition analysis indicated that population growth and aging were the principal drivers of increased DALYs and deaths, while epidemiological change contributed negatively. Projections suggested that deaths may continue to increase through 2045, despite declining age-standardized fatal burden. Conclusions: This is the first update of the burden of mesothelioma in China over the past thirty-four years. The absolute burden of mesothelioma in China, as estimated by the GBD study, increased markedly, largely driven by demographic changes. Strengthening asbestos exposure surveillance, diagnostic standardization, and cancer registration systems would enable burden estimates to be derived from directly observed and certified data rather than relying primarily on model-based assumptions, while potentially identifying previously unrecognized sources of asbestos exposure. Full article