Search Results (96)

Background/Objectives: Obesity is a major global public health and economic challenge. Governments worldwide have implemented nutrition-focused policies such as sugar-sweetened beverage taxes, front-of-pack labeling, food assistance reforms, and school nutrition standards to improve diet quality and reduce obesity. Because large-scale randomized controlled trials are often infeasible and conventional epidemiologic methods overlook population heterogeneity and behavioral feedback, microsimulation modeling has become a key tool for evaluating long-term and distributional policy impacts. This scoping review examined the application of microsimulation to obesity-related nutrition policies, focusing on model structure, behavioral parameterization, and integration of economic and equity analyses. Methods: Following PRISMA guidelines (PROSPERO CRD42024599769), five databases were searched for peer-reviewed studies. Data were extracted on policy mechanisms, model design, parameterization, and equity analysis. Study quality was assessed using a customized 21-item checklist adapted from CHEERS and NIH tools. Results: Twenty-nine studies met the inclusion criteria, with most policy settings based in the United States. Most employed dynamic, stochastic, individual-level microsimulation models with diverse behavioral assumptions, obesity equations, and calibration approaches. While most studies stratified outcomes by socioeconomic or demographic group, only one used a formal quantitative equity metric. Conclusions: Microsimulation modeling provides valuable evidence on the long-term health, economic, and distributional impacts of nutrition policies. Future work should strengthen methodological transparency, standardize equity assessment, and expand application beyond high-income settings to improve the comparability, credibility, and policy relevance of simulation-based nutrition policy research. Full article

Accurate and real-time state of charge (SOC) monitoring is critical for the safe, efficient, and stable long-term operation of vanadium redox flow batteries (VRFBs). Traditional monitoring methods are susceptible to errors arising from side reactions, cumulative drift, and electrolyte imbalance. This study develops a non-invasive optical sensor module for the negative electrolyte (anolyte), utilizing the favorable spectral properties of V(II)/V(III) ions at 850 nm for real-time SOC tracking. A fifth-order polynomial model was employed for calibration, successfully managing the non-linear optical response of highly concentrated electrolytes and achieving exceptional accuracy (adjusted R² > 0.9999). The optical sensor reliably tracked capacity degradation over 50 galvanostatic cycles, yielding a degradation curve that showed a high correlation with the conventional coulomb counting method, thus confirming its feasibility for assessing battery’s state of health. Contrary to initial expectations, operating at higher current densities resulted in a lower capacity degradation rate (CDR). This phenomenon is primarily attributed to the time-dependent nature of parasitic side reactions. Higher current densities reduce the cycle duration, thereby minimizing the temporal exposure of active species to degradation mechanisms and mitigating cumulative ion imbalance. This mechanism was corroborated by physicochemical analysis via UV-Vis spectroscopy, which revealed a strong correlation between the severity of spectral deviation and the CDR ranking. This non-invasive optical technology offers a low-cost and effective solution for precise VRFB management and preventative maintenance. Full article

Veteran suicide remains a critical public health crisis in the United States, with rates nearly twice those of the general population. Addressing this challenge requires multiple evidence-based interventions across settings. This paper presents a system dynamics model developed within the Face the Fight™ veteran suicide prevention initiative to evaluate and optimize strategies from 2022 to 2032. The model integrates peer-reviewed evidence on intervention effectiveness, subject-matter expert calibration, and annual updates from Veterans Affairs and grantee data to estimate the potential population-level impact of suicide prevention. The model organizes veterans by levels of suicide distress and estimates the impact of interventions in an initial three target areas aligned with a public health approach to suicide prevention: creating protective environments (e.g., secure firearm storage), strengthening access and delivery of suicide care (e.g., suicide-specific clinical programs), and identifying and supporting people at risk (e.g., suicide screening). Model results indicate that focusing solely on high-distress veterans is insufficient to reduce suicide rates to those of the general population, while balanced portfolios combining clinical, community, and firearm-safety approaches yield the greatest projected benefit. Sensitivity analyses demonstrate the model’s responsiveness to population distress distributions and intervention capacities, underscoring the need for a balanced, scalable strategy. Evaluating suicide-prevention impact is inherently challenging, but the model provides a dynamic and transparent framework for assessing investment effectiveness, refining strategies, and forecasting long-term outcomes. Its adaptability ensures ongoing insights to guide funding priorities, informs data-driven policy, and extends to other populations and public health challenges where multiple interventions interact to influence outcomes. Full article

Vehicle, industrial, and urban emissions remain major contributors to air quality degradation, affecting public health and the level of environmental cleanliness. Cost-effective specific pollutant estimation models, i.e., for carbon monoxide $\mathrm{CO}$, carbon dioxide ${\mathrm{CO}}_2$, and ammonia ${\mathrm{NH}}_3$, are essential to tackle the practical challenge of high-resolution monitoring for reducing vehicle emissions in traffic. Existing model design methods, however, may be insufficient, particularly for peak time estimations, since such models are typically designed using gridding-based vehicle-specific power polynomial and non-optimized artificial neural networks. In this paper, we propose vehicle emission models of pollutants based on a Bayesian Monte Carlo (MC) Dropout-based robust data-driven gated recurrent unit (BMC-GRU) method to enhance estimation robustness and mitigate the overfitting problem in the deep learning network. Bayesian optimization determines the optimal architecture by efficiently and probabilistically searching the hyperparameters of the network, while MC-Dropout quantifies epistemic uncertainty through multiple stochastic forward passes during testing. Therefore, the proposed method improves the models’ calibrations and robustness to distribution shifts. For benchmarking, least squares-based first- and fourth-order polynomials, conventional long-short term memory (LSTM), and bidirectional LSTM (BiLSTM)-based estimation models are designed. The proposed method outperforms the mentioned state-of-the-art methods with strong robust estimation performance. The experimental results on multiple real-world vehicle datasets demonstrate that the proposed method significantly outperforms state-of-the-art approaches. The method presents a promising solution for uncertainty-aware vehicle emission modeling that is applicable to transportation systems. Full article

Indoor Air Quality (IAQ) in educational environments is a critical determinant of students’ health, well-being, and learning performance, with inadequate ventilation and pollutant accumulation consistently associated with respiratory symptoms, fatigue, and impaired cognitive outcomes. Conventional monitoring approaches—based on periodic inspections or subjective perception—provide only fragmented insights and often underestimate exposure risks. Artificial intelligence (AI) offers a transformative framework to overcome these limitations through sensor calibration, anomaly detection, pollutant forecasting, and the adaptive control of ventilation systems. This review critically synthesizes the state of AI applications for IAQ management in educational environments, drawing on twenty real-world case studies from North America, Europe, Asia, and Oceania. The evidence highlights methodological innovations ranging from decision tree models integrated into large-scale sensor networks in Boston to hybrid deep learning architectures in New Zealand, and regression-based calibration techniques applied in Greece. Collectively, these studies demonstrate that AI can substantially improve predictive accuracy, reduce pollutant exposure, and enable proactive, data-driven ventilation management. At the same time, cross-case comparisons reveal systemic challenges—including sensor reliability and calibration drift, high installation and maintenance costs, limited interoperability with legacy building management systems, and enduring concerns over privacy and trust. Addressing these barriers will be essential for moving beyond localized pilots. The review concludes that AI holds transformative potential to shift school IAQ management from reactive practices toward continuous, adaptive, and health-oriented strategies. Realizing this potential will require transparent, equitable, and cost-effective deployment, positioning AI not only as a technological solution but also as a public health and educational priority. Full article

This paper presents a real-time, output-only structural health monitoring framework that integrates damage-sensitive cepstral features with a streaming Long Short-Term Memory (LSTM) network for automated damage detection. Acceleration time histories are segmented into overlapping windows, converted into cepstral coefficients, and processed sequentially by a stacked LSTM architecture with state carry-over. This design preserves temporal dependencies while enabling low-latency inference suitable for continuous monitoring. The framework was evaluated under a strict zero-shot setting on the full-scale Z24 Bridge benchmark, in which no training or calibration data from the bridge were used. Our results show that the proposed approach can reliably discriminate staged damage states and track their progression using only vibration measurements. By combining a well-established spectral feature representation with a streaming sequence model, the study demonstrates a practical pathway toward deployable, data-driven monitoring systems capable of operating without retraining on each individual asset. Full article

Accurate estimation of lithium-ion battery (LIB) state of health (SOH) underpins safe operation, predictive maintenance, and lifetime-aware energy management. Despite recent advances in machine learning (ML), systematic benchmarking across heterogeneous real-world cells remains limited, often confounded by data leakage and inconsistent validation. Here, we establish a leakage-averse, cross-battery evaluation framework encompassing 32 commercial LIBs (B5–B56) spanning diverse cycling histories and temperatures (≈4 °C, 24 °C, 43 °C). Models ranging from classical regressors to ensemble trees and deep sequence architectures were assessed under blocked 5-fold GroupKFold splits using RMSE, MAE, R² with confidence intervals, and inference latency. The results reveal distinct stratification among model families. Sequence-based architectures—CNN–LSTM, GRU, and LSTM—consistently achieved the highest accuracy (mean RMSE ≈ 0.006; per-cell R² up to 0.996), demonstrating strong generalization across regimes. Gradient-boosted ensembles such as LightGBM and CatBoost delivered competitive mid-tier accuracy (RMSE ≈ 0.012–0.015) yet unrivaled computational efficiency (≈0.001–0.003 ms), confirming their suitability for embedded applications. Transformer-based hybrids underperformed, while approximately one-third of cells exhibited elevated errors linked to noise or regime shifts, underscoring the necessity of rigorous evaluation design. Collectively, these findings establish clear deployment guidelines: CNN–LSTM and GRU are recommended where robustness and accuracy are paramount (cloud and edge analytics), while LightGBM and CatBoost offer optimal latency–efficiency trade-offs for embedded controllers. Beyond model choice, the study highlights data curation and leakage-averse validation as critical enablers for transferable and reliable SOH estimation. This benchmarking framework provides a robust foundation for future integration of ML models into real-world battery management systems. Full article

Heart rate variability (HRV), which quantitatively characterizes fluctuations in beat-to-beat intervals, serves as a critical indicator of cardiovascular and autonomic nervous system health. The inherent ability of non-contact methods to eliminate the need for subject contact effectively mitigates user burden and facilitates scalable long-term monitoring, thus attracting considerable research interest in non-contact HRV sensing. In this study, we propose a novel algorithm for HRV extraction utilizing FMCW millimeter-wave radar. First, we developed a calibration-free 3D target positioning module that captures subjects’ micro-motion signals through the integration of digital beamforming, moving target indication filtering, and DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering techniques. Second, we established separate phase-based mathematical models for respiratory and cardiac vibrations to enable systematic signal separation. Third, we implemented the Second Order Spectral Sparse Separation Algorithm Using Lagrangian Multipliers, thereby achieving robust heartbeat extraction in the presence of respiratory movements and noise. Heartbeat events are identified via peak detection on the recovered cardiac signal, from which inter-beat intervals and HRV metrics are subsequently derived. Compared to state-of-the-art algorithms and traditional filter bank approaches, the proposed method demonstrated an over 50% reduction in average IBI (Inter-Beat Interval) estimation error, while maintaining consistent accuracy across all test scenarios. However, it should be noted that the method is currently applicable only to scenarios with limited subject movement and has been validated in offline mode, but a discussion addressing these two issues is provided at the end. Full article

Anxiety and depression affect millions worldwide, yet stigma and long wait times often delay access to care. Mobile mental health apps can decrease these barriers by offering on-demand screening and support. Nevertheless, many machine and deep learning methods used in such tools perform poorly under severe class imbalance, yielding biased, poorly calibrated predictions. To address this challenge, this study proposes MCoG-LDPSNet, a brain-inspired model that combines dual, orthogonal encoding pathways with a novel Loss-Driven Parametric Swish (LDPS) activation. LDPS implements a neurobiologically motivated adaptive-gain mechanism via a learnable β parameter driven by calibration and confidence-aware loss signals that amplifies minority-class patterns while preserving overall reliability, enabling robust predictions under severe data imbalance. On a benchmark mental health corpus, MCoG-LDPSNet achieved AUROC = 0.9920 and G-mean = 0.9451, outperforming traditional baselines like GLMs, XGBoost, state-of-the-art deep models (CNN-BiLSTM-ATTN), and transformer-based approaches. After transfer learning to social media text, the MCoG-LDPSNet maintained a near-perfect AUROC of 0.9937. Integrated into the EmotiZen App with enhanced app features, MCoG-LDPSNet was associated with substantial symptom reductions (anxiety 28.2%; depression 42.1%). These findings indicate that MCoG-LDPSNet is an accurate, imbalance-aware solution suitable for scalable mobile screening of individuals for anxiety and depression. Full article

Depression remains a leading cause of global disability, yet scalable and objective diagnostic tools are still lacking. Speech has emerged as a promising non-invasive modality for automated depression detection, due to its strong correlation with emotional state and ease of acquisition. While convolutional neural networks (CNNs) have achieved state-of-the-art performance in this domain, their high computational demands limit deployment in low-resource or real-time settings. Spiking neural networks (SNNs), by contrast, offer energy-efficient, event-driven computation inspired by biological neurons, but they are difficult to train directly and often exhibit degraded performance on complex tasks. This study investigates whether CNNs trained on audio data from the clinically annotated DAIC-WOZ dataset can be effectively converted into SNNs while preserving diagnostic accuracy. We evaluate multiple conversion thresholds using the SpikingJelly framework and find that the 99.9% mode yields an SNN that matches the original CNN in both accuracy (82.5%) and macro F1 score (0.8254). Lower threshold settings offer increased sensitivity to depressive speech at the cost of overall accuracy, while naïve conversion strategies result in significant performance loss. These findings support the feasibility of CNN-to-SNN conversion for real-world mental health applications and underscore the importance of precise calibration in achieving clinically meaningful results. Full article

University classrooms, core to higher education, have indoor light environments that directly affect students’ learning efficiency, visual health, and psychological states. This study integrates parametric optical simulation and virtual reality (VR) to explore light environment perception in ordinary university classrooms. Forty college students (18–25 years, ~1:1 gender ratio) participated in real virtual comparative experiments. VR scenarios were optimized via real-time rendering and physical calibration. The results showed no significant differences in subjects’ perception evaluations between environments (p > 0.05), verifying virtual environments as effective experimental carriers. The analysis of eight virtual conditions (varying window-to-wall ratios and lighting methods) revealed that mixed lighting performed best in light perception, spatial perception, and overall evaluation. Light perception had the greatest influence on overall evaluation (0.905), with glare as the core factor (0.68); closure sense contributed most to spatial perception (0.45). Structural equation modeling showed that window-to-wall ratio and lighting power density positively correlated with subjective evaluations. Window-to-wall ratio had a 0.412 direct effect on spatial perception and a 0.84 total mediating effect (67.1% of total effect), exceeding the lighting power density’s 0.57 mediating effect sum. This study confirms mixed lighting and window-to-wall ratio optimization as keys to improving classroom light quality, providing an experimental paradigm and parameter basis for user-perception-oriented design. Full article

Food quality and safety are essential for ensuring public health, preventing foodborne illness, reducing food waste, maintaining consumer confidence, and supporting regulatory compliance and international trade. This has led to the emergence of many research works that focus on automating and streamlining the assessment of food quality. Electronic noses have become of paramount importance in this context. We analyze the current state of research in the development of electronic noses for food quality and safety. We examined research papers published in three different scientific databases in the last decade, leading to a comprehensive review of the field. Our review found that most of the efforts use portable, low-cost electronic noses, coupled with pattern recognition algorithms, for evaluating the quality levels in certain well-defined food classes, reaching accuracies exceeding 90% in most cases. Despite these encouraging results, key challenges remain, particularly in diversifying the sensor response across complex substances, improving odor differentiation, compensating for sensor drift, and ensuring real-world reliability. These limitations indicate that a complete device mimicking the flexibility and selectivity of the human olfactory system is not yet available. To address these gaps, our review recommends solutions such as the adoption of adaptive machine learning models to reduce calibration needs and enhance drift resilience and the implementation of standardized protocols for data acquisition and model validation. We introduce benchmark comparisons and a future roadmap for electronic noses that demonstrate their potential to evolve from controlled studies to scalable industrial applications. In doing so, this review aims not only to assess the state of the field but also to support its transition toward more robust, interpretable, and field-ready electronic nose technologies. Full article

It is important to accurately characterize the plane-stress state of pipe walls for evaluating the bearing capacity of the pipe and ensuring the structural safety. This paper describes a novel ultrasonic technique for evaluating anisotropic pipe-wall plane stresses using three-directional longitudinal critical refracted (LCR) wave time-of-flight (TOF) measurements. The connection between plane stress and ultrasonic TOF is confirmed by examining how the anisotropy of rolled steel plates affects the speed of ultrasonic wave propagation, which is a finding not previously documented in spiral-welded pipes. Then based on this relationship, an ultrasonic stress coefficient calibration experiment for spiral-welded pipes is designed. The results show that the principal stress obtained by the ultrasonic method is closer to the engineering stress than that obtained from the coercivity method. And, as a nondestructive testing technique, the ultrasonic method is more suitable for in-service pipelines. It also elucidates the effects of probe pressure and steel plate surface roughness on the ultrasonic TOF, obtains a threshold for probe pressure, and reveals a linear relationship between roughness and TOF. This study provides a feasible technique for nondestructive measurement of plane stress in anisotropic spiral-welded pipelines, which has potential application prospects in the health monitoring of in-service pipelines. Full article

Background/Objectives: The COVID-19 pandemic highlighted the critical need for accurate predictive models to guide public health interventions and optimize healthcare resource allocation. This study evaluates how the complexity of compartmental infectious disease models influences their forecasting accuracy and utility for pandemic resource planning. Methods: We analyzed a range of compartmental models, including simple susceptible-infected-recovered (SIR) models and more complex frameworks incorporating asymptomatic carriers and deaths. These models were calibrated and tested using real-world COVID-19 data from the United States to assess their performance in predicting symptomatic and asymptomatic infection counts, peak infection timing, and resource demands. Both adaptive models (updating parameters with real-time data) and non-adaptive models were evaluated. Results: Numerical results show that while more complex models capture detailed disease dynamics, simpler models often yield better forecast accuracy, especially during early pandemic stages or when predicting peak infection periods. Adaptive models provided the most accurate short-term forecasts but required substantial computational resources, making them less practical for long-term planning. Non-adaptive models produced stable long-term forecasts useful for strategic resource allocation, such as hospital bed and ICU planning. Conclusions: Model selection should align with the pandemic stage and decision-making horizon. Simpler models are effective for rapid early-stage interventions, adaptive models excel in short-term operational forecasting, and non-adaptive models remain valuable for long-term resource planning. These findings can inform policymakers on selecting appropriate modeling approaches to improve pandemic response effectiveness. Full article

The gut microbiota consists of trillions of microorganisms, mostly bacteria, which establish a symbiotic relationship with the host. The host provides a favourable environment and the essential nutrients for their proliferation, while the gut microbiota plays a key role in maintaining the host’s health. Therefore, imbalances in its composition, a state known as dysbiosis, can contribute to the onset or progression of various pathological conditions, including atherosclerosis. Atherosclerosis is a chronic, slow-progressing inflammatory disease characterised by the formation and potential rupture of atheromatous plaques in medium- and large-calibre arteries. It underlies major cardiovascular events, such as stroke and myocardial infarction, and remains a leading cause of global morbidity and mortality. The modulation of the gut microbiota using prebiotics, probiotics, and faecal microbiota transplantation (FMT) has emerged as a promising approach for preventing and managing atherosclerosis. Although numerous studies have explored these strategies, further research is needed to establish their efficacy and mechanisms. This review explores the pathophysiology of atherosclerosis, its main risk factors, and the interplay between the gut microbiota and atherosclerosis, with a particular focus on the mechanisms by which microbiota-targeted interventions, including prebiotics, probiotics, and FMT, may serve as therapeutic adjuvants in the prevention and treatment of atherosclerosis. Full article