Search Results (1,665)

This comprehensive review maps the fast-evolving landscape in which artificial intelligence (AI) and deep-learning (DL) techniques converge with the Internet of Things (IoT) to manage energy, comfort, and sustainability across smart environments. A PRISMA-guided search of four databases retrieved 1358 records; after applying inclusion criteria, 143 peer-reviewed studies published between January 2019 and April 2025 were analyzed. This review shows that AI-driven controllers—especially deep-reinforcement-learning agents—deliver median energy savings of 18–35% for HVAC and other major loads, consistently outperforming rule-based and model-predictive baselines. The evidence further reveals a rapid diversification of methods: graph-neural-network models now capture spatial interdependencies in dense sensor grids, federated-learning pilots address data-privacy constraints, and early integrations of large language models hint at natural-language analytics and control interfaces for heterogeneous IoT devices. Yet large-scale deployment remains hindered by fragmented and proprietary datasets, unresolved privacy and cybersecurity risks associated with continuous IoT telemetry, the growing carbon and compute footprints of ever-larger models, and poor interoperability among legacy equipment and modern edge nodes. The authors of researches therefore converges on several priorities: open, high-fidelity benchmarks that marry multivariate IoT sensor data with standardized metadata and occupant feedback; energy-aware, edge-optimized architectures that lower latency and power draw; privacy-centric learning frameworks that satisfy tightening regulations; hybrid physics-informed and explainable models that shorten commissioning time; and digital-twin platforms enriched by language-model reasoning to translate raw telemetry into actionable insights for facility managers and end users. Addressing these gaps will be pivotal to transforming isolated pilots into ubiquitous, trustworthy, and human-centered IoT ecosystems capable of delivering measurable gains in efficiency, resilience, and occupant wellbeing at scale. Full article

Background and Objectives: General-purpose multimodal large language models (LLMs) are increasingly used for medical image interpretation despite lacking clinical validation. This study evaluates the diagnostic reliability of ChatGPT-4o and Claude 2 in photographic assessment of adolescent idiopathic scoliosis (AIS) against radiological standards. This study examines two critical questions: whether families can derive reliable preliminary assessments from LLMs through analysis of clinical photographs and whether LLMs exhibit cognitive fidelity in their visuospatial reasoning capabilities for AIS assessment. Materials and Methods: A prospective diagnostic accuracy study (STARD-compliant) analyzed 97 adolescents (74 with AIS and 23 with postural asymmetry). Standardized clinical photographs (nine views/patient) were assessed by two LLMs and two orthopedic residents against reference radiological measurements. Primary outcomes included diagnostic accuracy (sensitivity/specificity), Cobb angle concordance (Lin’s CCC), inter-rater reliability (Cohen’s κ), and measurement agreement (Bland–Altman LoA). Results: The LLMs exhibited hazardous diagnostic inaccuracy: ChatGPT misclassified all non-AIS cases (specificity 0% [95% CI: 0.0–14.8]), while Claude 2 generated 78.3% false positives. Systematic measurement errors exceeded clinical tolerance: ChatGPT overestimated thoracic curves by +10.74° (LoA: −21.45° to +42.92°), exceeding tolerance by >800%. Both LLMs showed inverse biomechanical concordance in thoracolumbar curves (CCC ≤ −0.106). Inter-rater reliability fell below random chance (ChatGPT κ = −0.039). Universal proportional bias (slopes ≈ −1.0) caused severe curve underestimation (e.g., 10–15° error for 50° deformities). Human evaluators demonstrated superior bias control (0.3–2.8° vs. 2.6–10.7°) but suboptimal specificity (21.7–26.1%) and hazardous lumbar concordance (CCC: −0.123). Conclusions: General-purpose LLMs demonstrate clinically unacceptable inaccuracy in photographic AIS assessment, contraindicating clinical deployment. Catastrophic false positives, systematic measurement errors exceeding tolerance by 480–1074%, and inverse diagnostic concordance necessitate urgent regulatory safeguards under frameworks like the EU AI Act. Neither LLMs nor photographic human assessment achieve reliability thresholds for standalone screening, mandating domain-specific algorithm development and integration of 3D modalities. Full article

The growing demand for accurate and efficient Chest X-Ray (CXR) interpretation has prompted the development of AI-driven systems to alleviate radiologist workload and reduce diagnostic variability. This paper introduces the Intelligent Humanized Radiology Analysis System (IHRAS), a modular framework that automates the end-to-end process of CXR analysis and report generation. IHRAS integrates four core components: (i) deep convolutional neural networks for multi-label classification of 14 thoracic conditions; (ii) Grad-CAM for spatial visualization of pathologies; (iii) SAR-Net for anatomical segmentation; and (iv) a large language model (DeepSeek-R1) guided by the CRISPE prompt engineering framework to generate structured diagnostic reports using SNOMED CT terminology. Evaluated on the NIH ChestX-ray dataset, IHRAS demonstrates consistent diagnostic performance across diverse demographic and clinical subgroups, and produces high-fidelity, clinically relevant radiological reports with strong faithfulness, relevancy, and alignment scores. The system offers a transparent and scalable solution to support radiological workflows while highlighting the importance of interpretability and standardization in clinical Artificial Intelligence applications. Full article

Network security courses rely on hands-on labs where students configure virtual Linux networks to practice attack and defense. Automated feedback is scarce because no standard exists for exchanging detailed configurations—interfaces, bridging, routing tables, iptables policies—between exercise software and large language models (LLMs) that could serve as tutors. We address this interoperability gap with an exercise-oriented YANG profile that augments the Internet Engineering Task Force (IETF) ietf-network module with a new network-devices module. The profile expresses Linux interface settings, routing, and firewall rules, and tags each node with roles such as linux-server or linux-firewall. Integrated into our LiNeS Cloud platform, it enables LLMs to both parse and generate machine-readable network states. We evaluated the profile on four topologies—from a simple client–server pair to multi-subnet scenarios with dedicated security devices—using ChatGPT-4o, Claude 3.7 Sonnet, and Gemini 2.0 Flash. Across 1050 evaluation tasks covering profile understanding (n = 180), instance analysis (n = 750), and instance generation (n = 120), the three LLMs answered correctly in 1028 cases, yielding an overall accuracy of 97.9%. Even with only minimal follow-up cues (≦3 turns) —rather than handcrafted prompt chains— analysis tasks reached 98.1% accuracy and generation tasks 93.3%. To our knowledge, this is the first exercise-focused YANG profile that simultaneously captures Linux/iptables semantics and is empirically validated across three proprietary LLMs, attaining 97.9% overall task accuracy. These results lay a practical foundation for artificial intelligence (AI)-assisted security labs where real-time feedback and scenario generation must scale beyond human instructor capacity. Full article

This paper presents an interpretable and modular framework for energy management in smart buildings based on fuzzy logic and the IEEE Std 1855-2016. The proposed system builds upon the JFML-IoT library, enabling the integration and execution of fuzzy rule-based systems on resource-constrained IoT devices using a lightweight and extensible architecture. Unlike conventional data-driven controllers, this approach emphasizes semantic transparency, expert-driven control logic, and compliance with fuzzy markup standards. The system is designed to enhance both operational efficiency and user comfort through transparent and explainable decision-making. A four-layer architecture structures the system into Perception, Communication, Processing, and Application layers, supporting real-time decisions based on environmental data. The fuzzy logic rules are defined collaboratively with domain experts and encoded in Fuzzy Markup Language to ensure interoperability and formalization of expert knowledge. While adherence to IEEE Std 1855-2016 facilitates system integration and standardization, the scientific contribution lies in the deployment of an interpretable, IoT-based control system validated in real conditions. A case study is conducted in a realistic indoor environment, using temperature, humidity, illuminance, occupancy, and CO₂ sensors, along with HVAC and lighting actuators. The results demonstrate that the fuzzy inference engine generates context-aware control actions aligned with expert expectations. The proposed framework also opens possibilities for incorporating user-specific preferences and adaptive comfort strategies in future developments. Full article

This paper introduces a unified and systematic design methodology for robotic systems that is generalizable across a wide range of applications. It integrates rigorous mathematical formalisms such as kinematics, dynamics, control theory, and optimization with advanced simulation tools, ensuring that each design decision is grounded in provable theory. The approach defines clear phases, including mathematical modeling, virtual prototyping, parameter optimization, and theoretical validation. Each phase builds on the previous one to reduce unforeseen integration issues. Spanning from conceptualization to deployment, it offers a blueprint for developing mathematically valid and robust robotic solutions while streamlining the transition from design intent to functional prototype. By standardizing the design workflow, this framework reduces development time and cost, improves reproducibility across projects, and enhances collaboration among multidisciplinary teams. Such a generalized approach is essential in today’s fast-evolving robotics landscape where rapid innovation and cross-domain applicability demand flexible yet reliable methodologies. Moreover, it provides a common language and set of benchmarks that both novice and experienced engineers can use to evaluate performance, facilitate knowledge transfer, and future-proof systems against emerging application requirements. Full article

The evaluation of open-ended questions is typically performed by human instructors using predefined criteria to uphold academic standards. However, manual grading presents challenges, including high costs, rater fatigue, and potential bias, prompting interest in automated essay scoring systems. While automated essay scoring tools can assess content, coherence, and grammar, discrepancies between human and automated scoring have raised concerns about their reliability as standalone evaluators. Large language models like ChatGPT offer new possibilities, but their consistency and fairness in feedback remain underexplored. This study investigates whether ChatGPT can provide stable and fair essay scoring—specifically, whether identical student responses receive consistent evaluations across multiple AI interactions using the same criteria. The study was conducted in two marketing courses at an engineering school in Chile, involving 40 students. Results showed that ChatGPT, when unprompted or using minimal guidance, produced volatile grades and shifting criteria. Incorporating the instructor’s rubric reduced this variability but did not eliminate it. Only after providing an example-rich rubric, a standardized output format, low temperature settings, and a normalization process based on decision tables did ChatGPT-4o demonstrate consistent and fair grading. Based on these findings, we developed a scalable algorithm that automatically generates effective grading rubrics and decision tables with minimal human input. The added value of this work lies in the development of a scalable algorithm capable of automatically generating normalized rubrics and decision tables for new questions, thereby extending the accessibility and reliability of automated assessment. Full article

The present study investigates how US Southerners perceive second language (L2) speech by recruiting 170 undergraduate students who spoke Southern American English to listen to recordings of four speakers (US, Bangladeshi, Chinese, and Saudi Arabian) and evaluate their attributes. The listeners were grouped based on their ethnic affiliation: African American, Anglo-American, and Asian/Hispanic/multi-racial. A random half were primed, being asked questions about whether/how other people had negatively commented on their accents. Results showed no effect of priming on speech ratings. Moreover, whilst African American and Anglo-American listeners rated L2 speakers lower than the L1 speaker in almost all aspects, Asian/Hispanic/multi-racial listeners did not. Full article

Microplastic (MP) pollution is an escalating global environmental concern, with a growing body of research addressing diverse dimensions of this issue. Despite this progress, the field remains hindered by generating large, heterogeneous datasets that follow inconsistent reporting standards, resulting in fragmented and often incompatible databases. While various databases on MPs have been developed, they primarily operate in isolation, limiting the accessibility and cross-comparison of data. This study presents a foundational approach to aggregating and accessing existing MP pollution datasets. A comprehensive review of the currently available databases was conducted to evaluate their integration potential. It revealed key challenges such as non-standardized data formats, limited accessibility, and difficulty performing comparative analyses across sources. To address these barriers, a prototype web-based platform was developed that enables unified access to MP datasets. The architecture includes a smart standardization layer that harmonizes inputs from disparate sources. The integration of Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) techniques was proposed to facilitate natural language querying. This enables researchers to interact with the platform intuitively and extract meaningful insights more efficiently. The proposed system aims to enhance data discoverability, promote interoperability, and support robust, data-driven environmental research, paving the way toward more informed policy-making and scientific collaboration in the fight against MP pollution. With this platform, there is a potential for new discoveries and a future in which the tools to effectively combat this global issue are available, making the audience realize the potential for new discoveries. Full article

This study investigates the cybersecurity landscape of Aotearoa-New Zealand through a culturally grounded lens, focusing on the integration of Indigenous Māori values into cybersecurity frameworks. In response to escalating cyber threats, the research adopts a mixed-methods and interdisciplinary approach—combining surveys, focus groups, and case studies—to explore how cultural principles such as whanaungatanga (collective responsibility) and manaakitanga (care and respect) influence digital safety practices. The findings demonstrate that culturally informed strategies enhance trust, resilience, and community engagement, particularly in rural and underserved Māori communities. Quantitative analysis revealed that 63% of urban participants correctly identified phishing attempts compared to 38% of rural participants, highlighting a significant urban–rural awareness gap. Additionally, over 72% of Māori respondents indicated that cybersecurity messaging was more effective when delivered through familiar cultural channels, such as marae networks or iwi-led training programmes. Focus groups reinforced this, with participants noting stronger retention and behavioural change when cyber risks were communicated using Māori metaphors, language, or values-based analogies. The study also confirms that culturally grounded interventions—such as incorporating Māori motifs (e.g., koru, poutama) into secure interface design and using iwi structures to disseminate best practices—can align with international standards like NIST CSF and ISO 27001. This compatibility enhances stakeholder buy-in and demonstrates universal applicability in multicultural contexts. Key challenges identified include a cybersecurity talent shortage in remote areas, difficulties integrating Indigenous perspectives into mainstream policy, and persistent barriers from the digital divide. The research advocates for cross-sector collaboration among government, private industry, and Indigenous communities to co-develop inclusive, resilient cybersecurity ecosystems. Based on the UTAUT and New Zealand’s cybersecurity vision “Secure Together—Tō Tātou Korowai Manaaki 2023–2028,” this study provides a model for small nations and multicultural societies to create robust, inclusive cybersecurity frameworks. Full article

Small Language Models (SLMs) have emerged as a critical area of study within natural language processing, attracting growing attention from both academia and industry. This systematic literature review provides a comprehensive and reproducible analysis of recent developments and advancements in SLMs post-2023. Drawing on 70 English-language studies published between January 2023 and January 2025, identified through Scopus, IEEE Xplore, Web of Science, and ACM Digital Library, and focusing primarily on SLMs (including those with up to 7 billion parameters), this review offers a structured overview of the current state of the art and potential future directions. Designed as a resource for researchers seeking an in-depth global synthesis, the review examines key dimensions such as publication trends, visual data representations, contributing institutions, and the availability of public datasets. It highlights prevailing research challenges and outlines proposed solutions, with a particular focus on widely adopted model architectures, as well as common compression and optimization techniques. This study also evaluates the criteria used to assess the effectiveness of SLMs and discusses emerging de facto standards for industry. The curated data and insights aim to support and inform ongoing and future research in this rapidly evolving field. Full article

Background/Objectives: Atypical sensory processing is increasingly recognized as a transdiagnostic dimension of neurodevelopmental disorders (NDDs), with critical implications for emotional and behavioral regulation. This study aimed to identify distinct sensory profiles in preschool children with NDDs and to examine their associations with emotional–behavioral and cognitive/developmental functioning. Methods: A total of 263 children (aged 21–71 months) diagnosed with autism spectrum disorder (ASD), language disorder (LD), or other NDDs (ONDD) were recruited. Sensory processing was assessed using the SPM-P, emotional–behavioral functioning was assessed via the CBCL 1½–5, and cognitive/developmental levels were assessed through standardized instruments. Latent profile analysis (LPA) was conducted to identify sensory subtypes. Group comparisons and multinomial logistic regression were used to examine profile characteristics and predictors of profile membership. Results: Three sensory profiles emerged: (1) Multisystemic Sensory Dysfunction (20.1%), characterized by pervasive sensory and emotional difficulties, primarily observed in ASD; (2) Typical Sensory Processing (44.9%), showing normative sensory and emotional functioning, predominantly LD; and (3) Mixed Subclinical Sensory Processing (35%), with subclinical-range scores across multiple sensory and emotional domains, spanning all diagnoses. Higher cognitive functioning and fewer internalizing symptoms significantly predicted membership in the typical profile. A gradient of symptom severity was observed across profiles, with the Multisystemic group showing the most pronounced emotional–behavioral impairments. Conclusions: Distinct sensory–emotional phenotypes were identified across diagnostic categories, supporting a dimensional model of neurodevelopment. Sensory profiles were strongly associated with emotional functioning, independently of diagnostic status. Early sensory assessment may therefore offer clinically meaningful insights into emotional vulnerability and inform targeted interventions in preschool populations with NDDs. Full article

Background: Greater tuberosity fracture–dislocations (GTFDs) represent a distinct subset of proximal humerus fractures, occurring in up to 57% of anterior glenohumeral dislocations. Malreduction may result in impingement, instability, and functional limitation. Treatment is influenced by the displacement magnitude and direction, bone quality, and patient activity level. Methods: This narrative review was based on a comprehensive search of PubMed, Scopus, and Web of Science for English-language articles published between January 2000 and March 2025. Studies on pathomechanics, classification, diagnosis, treatment, and outcomes of GTFDs in adult and pediatric populations were included. Data were analyzed to summarize the current evidence and identify clinical trends. Results: A displacement ≥ 5 mm is the standard surgical threshold, though superior or posterosuperior displacement ≥ 3 mm—and ≥2 mm in overhead athletes—may justify surgery. Conservative treatment remains appropriate for minimally displaced fractures but is associated with up to 48% subacromial impingement and 11% delayed surgery. Surgical options include arthroscopic repair for small or comminuted fragments and open reduction and internal fixation (ORIF) with screws or plates for larger, split-type fractures. Locking plates and double-row suture constructs demonstrate superior biomechanical performance compared with transosseous sutures. Reverse shoulder arthroplasty (RSA) is reserved for elderly patients with poor bone stock, cuff insufficiency, or severe comminution. Pediatric cases require physeal-sparing strategies. Conclusions: GTFDs management demands an individualized approach based on fragment displacement and direction, patient age and activity level, and bone quality. While 5 mm remains the common threshold, lower cutoffs are increasingly adopted in active patients. A tiered treatment algorithm integrating displacement thresholds, fracture morphology, and patient factors is proposed to support surgical decision making. The incorporation of fracture morphologic classifications further refines fixation strategy. Further prospective and pediatric-specific studies are needed to refine treatment algorithms and validate outcomes. Full article

Background and Objectives: Congenital heart defects (CHD) affect around 1% of the population, making them the most common congenital disease worldwide. Thanks to advances in treatment, over 90% of affected children are able to reach adulthood, shifting focus to long-term outcomes such as disease-specific quality of life (DsQoL). To date, there has been no validated, standardized instrument for assessing DsQoL in young German CHD patients. This study introduces the Congenital Heart Disease Specific Inventory (CHDSI), the first freely available German-language instrument for measuring DsQoL in children and adolescents with CHD. Materials and Methods: The CHDSI was developed at the German Heart Center Munich in collaboration with affected children and adolescents and validated nationwide via the National Register for Congenital Heart Defects (NRCHD) with 1201 participants (46 kindergarten children, 530 children, 625 adolescents). Two age-specific versions (36/37 items) and a 31-item preschool version were created, alongside a 6-item short form (CHDSI-SF) for rapid screening. Reliability was assessed using Cronbach’s alpha and split-half methods; construct validity via confirmatory factor analysis (CFA) using DWLS; and score interpretation through standardized stanine scales. The small sample size of kindergarten children precluded a model test for this group. The standard values given for this subsample should therefore be interpreted with caution. Results: The CHDSI showed excellent internal consistency (Cronbach’s α = 0.856 to 0.900) and high split-half reliability (>0.95). CFA confirmed a robust six-factor structure with excellent model fit (CFI and TLI ≥ 0.991, RMSEA ≤ 0.05). Subscales showed strong discriminant validity, and significant differences were found by CHD severity and sex. Conclusions: The CHDSI is a psychometrically valid, age-appropriate, and freely available instrument for assessing DsQoL in children and adolescents with CHD. It provides valuable support for clinical decision-making and research. Further studies should explore international validation and cultural adaptation. Full article

Objective: This cross-sectional content analysis aimed to investigate how oil pulling is portrayed on YouTube Shorts, focusing on the types of speakers, claims made, and alignment with scientific evidence. The study further explored how the content may influence viewer perception, health behaviors, and the potential spread of misinformation. Methods: On 28 January 2025, a systematic search of YouTube Shorts was performed using the term “oil pulling” in incognito mode to reduce algorithmic bias. English language videos with at least 1000 views were included through purposive sampling. A total of 47 Shorts met the inclusion criteria. Data were extracted using a structured coding framework that recorded speaker type (e.g., dentist, hygienist, influencer), engagement metrics, stated benefits, oil type and regimen, the use of disclaimers or citations, and stance toward oil pulling rated on a 5-point Likert scale. Speaker background and nationality were determined through publicly available channel descriptions or linked websites, with user identities anonymized and ethical approval deemed unnecessary due to the use of publicly available content. In total, 47 videos met the inclusion criteria. Results: Of the 47 YouTube Shorts that met the inclusion criteria, most were posted by influencers rather than dental professionals. These videos predominantly encouraged oil pulling, often recommending coconut oil for 10–15 min daily and citing benefits such as reduced halitosis and improved gum health. However, a smaller subset advanced more extreme claims, including reversing cavities and remineralizing enamel. Notably, US-licensed dentists and dental hygienists tended to discourage or express skepticism toward oil pulling, assigning lower Likert scores (1 or 2) to influencers and alternative health practitioners (often 4 or 5). Conclusions: YouTube Shorts largely promote oil pulling through anecdotal and testimonial-driven content, often diverging from evidence-based dental recommendations. The findings reveal a disconnect between professional dental guidance and popular social media narratives. While some benefits like halitosis reduction may have limited support, exaggerated or misleading claims may result in improper oral hygiene practices. Greater engagement from dental professionals and improved health communication strategies are needed to counteract misinformation and reinforce oil pulling’s role, if any, as an adjunct—not a replacement—for standard oral care. Future studies should explore viewer interpretation, behavioral influence, and cross-platform content patterns to better understand the impact of short-form health videos. Full article