共找到 20 条结果
Leptospirosis remains a major public health concern in tropical regions, including Malaysia, where it is endemic. To date, evidence from remote and understudied areas remains limited. This study aims to assess the epidemiological evidence of leptospirosis among market traders in Sabah, addressing the limited available data for this potentially high-risk occupational group in this region. A cross-sectional study was conducted among 295 market traders in Kudat Town, Sabah. Sociodemographic, occupational, environmental and behavioural data were collected using a modified validated questionnaire. Venous blood samples were collected to test for Leptospira antibodies using the Microscopic Agglutination Test; a titre of 1:50 or higher was considered seropositive. Seroprevalence was estimated descriptively, while factors associated with leptospiral seropositivity were examined using univariable and multivariable logistic regression analyses. The seroprevalence of leptospirosis among market traders was 5.4% (95% CI: 3.1-8.7). Antibodies were detected against 11 Leptospira serovars/strains, mainly Lep175. In univariable analysis, predictors for seropositivity included trading frequency, rodent presence, and proximity to garbage piles. With only 16 cases, the initial multivariable model had a low EPV of 4, suggesting possible overfitting. After reducing the model (EPV = 8), proximity to garbage piles remained strongly associated (AOR = 52.61; 95% CI: 13.50-205.13; p < 0.001), while rodents at home also became significant (AOR = 5.70; 95% CI: 1.12-29.08; p = 0.037). This study provides the first empirical evidence of leptospirosis exposure among market traders in Sabah. Although the overall seroprevalence was low, the findings indicate a measurable risk and underscore the need for further large-scale studies and integrated One Health interventions to reduce exposure in this vulnerable population.
In vertebrate adaptive immune systems, somatically diversified antigen receptors assume a central role in self/nonself discrimination. Attesting to the presence of a unique but unknown selective environment at early stages of vertebrate evolution, this facility emerged twice, in the ancestors of jawless and jawed vertebrates. Thus, the molecular structure of incomplete antigen receptor genes and their mode of assembly into functional genes are different in the two sister groups of vertebrates. It appears that adaptive immunity evolved in steps, trading immunologically favorable diversity of antigen receptor repertoires against the inherent risks of potentially destructive self recognition. Initially, the associated quality control mechanisms were largely cell-autonomous and grounded in the evolutionarily selected sequence composition of individual components available for assembly. At later stages, diversity increased in lock-step with emerging cell-nonautonomous quality control strategies: primary lymphoid organs spatially and temporally coupled repertoire development and assessment for self reactivity; regulatory cell types emerged to keep self reactive clones in check in the periphery. In this review, we discuss how comparative studies of vertebrate species situated at key positions in the phylogenetic tree have revealed traces of the evolutionary past of adaptive immune systems.
Model-informed precision dosing is often constrained by the limited generalizability of traditional population pharmacokinetic models, especially in critically ill patients. A hybrid machine learning-population pharmacokinetic framework is proposed to improve a priori pharmacokinetic predictions by integrating real-world clinical data. This approach was applied to vancomycin trough concentration prediction. Two widely used two-compartment population pharmacokinetic models provided individual pharmacokinetic parameter estimates. Maximum a posteriori Bayesian estimation was used to adjust population parameters for individual patients based on drug administration records, therapeutic drug monitoring values, and patient-specific covariates from the MIMIC-IV database. The resulting clearance and central volume of distribution estimates served as training targets for XGBoost and symbolic regression models. Machine learning-predicted parameters were reinserted into the original pharmacokinetic equations to generate a priori vancomycin trough concentrations without reliance on therapeutic drug monitoring input. The hybrid models demonstrated improved prediction accuracy over traditional population pharmacokinetic covariate models and reduced vancomycin trough concentration prediction error by up to ~20%. XGBoost generally provided the highest predictive performance, while symbolic regression produced interpretable mathematical expressions revealing associations between non-traditional clinical predictors and pharmacokinetic parameters, highlighting a trade-off between accuracy and interpretability. This framework illustrates the potential of combining machine learning with population pharmacokinetic modeling to refine pharmacokinetic parameter estimation and support more precise, individualized dosing. The workflow is adaptable to other drugs and patient populations, offering a generalizable methodological strategy to identify non-traditional predictors and enhance existing pharmacometric model performance in real-world clinical settings.
Aflatoxin (AF) contamination in tree nuts poses a serious threat to global food safety, public health, and international trade due to the potent carcinogenicity of aflatoxin B1 (AFB1). Although traditional mitigation strategies exist, their industrial implementation is constrained by a strict "technological filter," in which high detoxification efficacy must be carefully balanced against the preservation of the nutritional, structural, and sensory quality of the food matrix. This scoping review systematically mapped and critically synthesized recent scientific advances (2005-2025) in physical, chemical, and biological decontamination methods, evaluating their operational effectiveness, underlying mechanisms, and qualitative impacts on tree nuts. Guided by the question, "What physical, chemical, and biological methods are most effective for AF decontamination in tree nuts, and to what extent are they feasible regarding quality preservation and industrial applicability?", the study strictly followed the Joanna Briggs Institute (JBI) Manual for Evidence Synthesis and was reported according to PRISMA-ScR guidelines. A total of 41 eligible original studies were selected after comprehensive screening of the PubMed, Scopus, Web of Science, and ScienceDirect databases. Physical approaches, particularly cold atmospheric plasma and UV-C radiation, achieved robust reduction rates ranging from 70% to 95%; however, these highly energetic treatments often induced lipid oxidation, evidenced by linear increases in malondialdehyde (MDA) levels in matrices such as pistachios. Among chemical methods, ozonation and the use of organic acids achieved degradation rates close to 100% for the most toxic forms, but significant technological trade-offs were identified, including up to a 29% loss of α-tocopherol in hazelnuts. Inorganic selenium emerged as a promising alternative by combining efficacy with matrix preservation. Biological strategies using microorganisms such as Bacillus subtilis, Bifidobacterium lactis, and Lactobacillus kefiri demonstrated substantial sustainable potential, with detoxification efficiencies exceeding 80% through dual mechanisms of active biosuppression and physical adsorption. No single universal method is currently sufficient to handle contamination safely. The future of commercial AF management depends on the development of "hurdle technology," integrating multi-stage synergistic interventions with automated optical sorting and intelligent packaging, supported by further technical advances, studies of practical applicability, and industrial-scale validation to ensure absolute consumer safety and commercial viability.
Proteolysis-targeting chimaeras (PROTACs) couple target recognition to ubiquitin-dependent degradation, but their translation requires coordinated optimisation of degradation efficiency and developability. This review frames PROTAC design as a context-dependent medicinal chemistry problem rather than modular assembly of a warhead, linker and ubiquitin ligase (E3) recruiter. Linker length, rigidity, and exit vectors, together with warhead recognition topology, determine whether binary binding can form a cooperative, ubiquitination-competent ternary complex. Warhead binding mode further affects catalytic turnover and cellular degradation. Linker-free designs and disclosed clinical PROTAC structures show that beyond Rule of Five property control remains central to exposure. Conditional linkers and E3 ligase choice add biological constraints through stimulus-responsive activation, recruiter tractability, E3 expression, localisation, pathway biology, and safety liabilities. This review integrates these principles into a framework for PROTAC design. The framework aligns productive ternary complex assembly, effective exposure and biological-context compatibility within the intended therapeutic context.
The optimal surgical approach for mandibular condyle fractures remains controversial, particularly regarding the trade-off between facial nerve safety and surgical efficiency. Our systematic review and frequentist NMA analyzed studies evaluating open reduction and internal fixation (ORIF) for mandibular condyle fractures. The primary safety outcome was transient or persistent (≥6 months) facial nerve weakness. The efficiency outcome was operative exposure time. Surgical approaches were categorized, and random-effects NMAs estimated odds ratios (ORs) and mean differences (MDs). Treatments were ranked using P-scores, and clustered rankings explored safety-efficiency relationships. Overall, 121 studies (n = 6659 patients) were included. For facial nerve safety, endoscope-assisted (EA), preauricular transmasseteric anteroparotid (PATMA), and high submandibular (HSMA) approaches ranked highest, while the retromandibular transparotid (RMTA) approach carried a higher risk. Analyses restricted to persistent weakness yielded consistent results. For exposure time, submandibular (SM), HSMA, and retromandibular transmasseteric anteroparotid (RMTMA) approaches were most efficient. Clustered ranking analysis identified HSMA and RMTMA as achieving the best overall balance between safety and efficiency. Subgroup analyses confirmed the overall hierarchy, though evidence for high-level and intracapsular fractures remains limited. HSMA and RMTMA offer an optimal compromise between safety and efficiency for extracapsular condylar fractures, while EA may minimize nerve injury risk where technical expertise allows. High-quality comparative trials - particularly for intracapsular fractures - are warranted.
The escalating global demand for clean water necessitates the development of advanced membrane technologies capable of addressing complex contamination challenges and increasing the desalination efficiency. Polymeric membranes are commonly used due to their scalability and processability; however, their efficiency is limited by permeability-selectivity trade-offs and fouling. Polymer nanocomposite membranes (PNCMs) provide a promising substitute for the incorporation of functional nanofillers into polymer matrices for improving the separation efficiency. This review focuses on PNCMs for water purification and desalination, highlighting the importance of nanofillers like carbon nanotubes, graphene, and graphene oxide, metal and metal oxide nanoparticles (TiO2, ZnO, Ag), zeolites, and metal-organic frameworks. The influence of fabrication strategies, including phase inversion, electrospinning, and interfacial polymerization, on membrane structure-property relationships is systematically examined. While PNCMs demonstrate enhanced water permeability, selectivity, antifouling characteristics, and mechanical robustness compared to pristine membranes, their performance remains highly sensitive to nanofiller dispersion, interfacial compatibility, and structural stability. Key challenges, including nanoparticle agglomeration, long-term durability, and scalability constraints, are highlighted. Finally, future perspectives emphasize rational nanofiller design, controlled interface engineering, and scalable manufacturing approaches to enable the development of robust, high-performance membranes for sustainable water treatment.
Over the past decade, wearable activity monitoring (WAM) devices have become prevalent to improve quality of life and support the prevention and management of a wide range of health disorders. WAM devices based on the Internet of Medical Things (IoMT) paradigm provide a practical means of tracking physical activity, but their widespread adoption raises sustainability concerns. Meanwhile, health technology assessment and life cycle assessment are typically applied at advanced development stages, when empirical certainty about the final design and operating conditions is available, leaving little room for further improvements. This exploratory work provides the empirical foundations for an information-driven approach addressing this paradox in the early evaluation and conception of wrist-worn step counters, for which evidence suggests overdimensioned and unsustainable electronic designs. Specifically, we identify optimal resource-performance trade-offs in critical electronic components of frugal smartbands based on the data they collect and the information they preserve for step counting. This approach accounts for uncertainties in device reliability, users' gait speeds, and material and energy consumption in final products. We conducted a secondary analysis of an accelerometer dataset characterizing wrist motion in healthy individuals walking at different speeds. The original x-, y-, and z-axis signals were progressively downsampled using cubic spline interpolation and discretized to quantify the information preserved, first across sampling frequencies and then with respect to changes in motion velocity. We also preliminarily assessed the viability of the downsampled signals for step detection by estimating percentage errors in peak and valley counts. Based on this, we constructed and evaluated 4 frugal smartband design archetypes, linking energy consumption and sampling frequency for 4 widely used accelerometers and combining essential electronic components, implementing a suboptimal asynchronous first-in-first-out (FIFO) algorithm at different sampling rates. Finally, we evaluated the environmental impact and circularity of these components through a streamlined analysis focused on raw materials. About 70%-90% of the information is lost in signals downsampled at very low frequencies (2-5 Hz), whereas losses remain below 24% from 20 Hz onward. Substantial information loss also occurs when individuals walk briskly or jog (≥8 km/h), or walk below 8 km/h with sampling frequencies below 7 Hz or above 25 Hz. Step-counting accuracy is expected to be acceptable from approximately 11 Hz onward. Conversely, higher sampling rates rapidly saturate FIFO buffers and increase energy overhead, particularly when implemented in memory-dense components handling both processing and data transfer. Finally, gold and silver in transceivers and microcontrollers contribute substantially to resource depletion, while copper remains relevant for material recovery. These findings provide preliminary insights into the concurrent assessment and development of frugal WAM devices. This work extends the understanding of step detection under knowledge-constrained conditions and provides mechanisms to reduce uncertainty during early health technology assessment and eco-design.
Redox-active quinones have emerged as promising organic cathode materials (OCMs) for next-generation lithium-ion batteries (LIBs). However, their practical application is hindered by rapid dissolution in organic electrolytes and the common molecular design trade-off where the introduction of non-active structural motifs diminishes the specific capacity. To address these challenges, we propose a full-active-unit molecular design strategy. This approach connects two quinone (9,10-anthraquinone or 9,10-phenanthrenequinone) units via C─C single bond to a high-capacity pyrene-4,5,9,10-tetraone core, aiming for both low solubility and high specific capacity. Accordingly, we synthesized 2,7-bis(9,10-anthraquinonyl)pyrene-4,5,9,10-tetraone (BAPO) and 2,7-bis(9,10-phenanthraquinonyl)pyrene-4,5,9,10-tetraone (BPPO), both exhibiting low solubility. Electrochemical tests revealed excellent cell performance, particularly for the BAPO cathode, which delivered a high capacity of 317.5 mAh g- 1 at 0.2 C and demonstrated exceptional long-term cycling stability with 70.2% capacity retention after 9000 cycles at 5 C. This work provides a new molecular design concept for developing quinone cathode materials that simultaneously achieve high capacity and long cycle life.
Achieving a high plateau capacity in hard carbon (HC) anodes is one of the most critical prerequisites for high-energy-density sodium-ion batteries (SIBs), yet it is fundamentally limited by inaccessible closed pores formed during conventional high-temperature annealing. Here, we propose a potentially scalable oxidation-reconfiguration strategy to unlock its latent capacity. By coupling controlled oxidative etching with subsequent thermal reconstruction, an interconnected closed-pore network is constructed. Oxidative pretreatment opens blocked channels and interconnects isolated voids, while reconstruction promotes void fusion, generating accessible internal reservoirs for the nucleation and storage of quasi-metallic sodium clusters. This structural evolution and storage mechanism are elucidated by total neutron scattering, SAXS, in situ techniques, and simulations. As a result, the optimized ICP-HC anode delivers reversible capacity 437 mAh g-1 with an initial Coulombic efficiency of 91.5%. It achieves an initial discharge plateau capacity of 399 mAh g-1 with fast kinetics (300 mAh g-1 at 2C), overcoming the capacity-rate trade-off. Furthermore, an NVP//ICP-HC full cell shows excellent rate capability (96 mAh g-1 at 5C) and stable cycling (95% retention over 200 cycles at 1C). This work provides a scalable strategy for advanced carbon anodes in high-energy-density SIBs.
Unlocking the commercial potential of MXene electrodes in high-energy supercapacitors requires overcoming a fundamental limitation: severe performance decay at increased electrode thickness and mass loading. This work resolves the intrinsic trade-off between surface functionalization and oxidation in MXene chemistry regulation by constructing a controlled redox environment. A urea-assisted hydrothermal process effectively removes inert -F terminations while inducing the formation of a 3D MXene hydrogel enriched with -N active sites. Concurrently, L-ascorbic acid is introduced as an antioxidant to suppress structural oxidation and enhance stability. As a result, the optimized MXene exhibits superior rate capability and long-term cycling stability under high mass loadings. Specifically, at 10.97 mg cm-2, a high specific capacitance of 597 F g-1 is achieved at 1 A g-1, with 69.66% retention at 50 A g-1, significantly outperforming that of pristine MXene (23.44%) and N-doped MXene (59.03%). Even at an ultrahigh loading of 108.64 mg cm-2, 44.50% of the capacitance is retained (from 564 F g-1 to 251 F g-1) over the current density range of 1 to 20 A g-1. This work establishes an effective strategy for designing high-performance MXene-based electrodes via reaction pathway regulation, enabling practical operation at commercially relevant mass loadings.
The mutually reinforcing and coordinated development of Agricultural Digitalization (AD) and Agricultural New Quality Productive Forces (ANQPF) is an important path to achieving the modernization of agriculture and rural areas. Utilizing panel data from 31 provinces spanning 2012 to 2023, this study employs an Improved Coupling Coordination Degree Model (ICCDM), an Obstacle Degree Model (ODM), and a Random Effects Panel Tobit Model to investigate the spatiotemporal evolution patterns and the internal and external determinants of the Coupling Coordination Degree (CCD) between AD and ANQPF. The results indicate that: (1) From 2012 to 2023, both AD and ANQPF exhibited a steady upward trend, yet their overall development levels remained relatively low. (2) During this period, the CCD between the two systems evolved from moderately uncoordinated development to barely coordinated development, forming a distinct spatial pattern characterized by "the Eastern region leading, the Central region tackling key challenges, and the Western region catching up." (3) Internet penetration rate, leisure agriculture, and other factors are key obstacles affecting the coupling coordination, displaying heterogeneous characteristics across different regions. (4) Economic Development Level, Agricultural Development Level, Rural Consumption Level, Fiscal Support Intensity, Openness To Foreign Trade, and Agricultural Security Level significantly improve the CCD between AD and ANQPF, whereas Industrial Development Level has a significant negative effect on the CCD. Furthermore, external influencing factors demonstrated notable regional variations. Consequently, implementing targeted policies and region-targeted guidance represents an imperative pathway for fostering the coordinated development of AD and ANQPF.
Biodiversity science is increasingly data-rich and computationally sophisticated, yet humans are still often treated as absent, implicit or overly simplified drivers of ecological and evolutionary change. This limits how well models represent the real-world processes that generate biodiversity and practical impact at a time when human activities are rapidly reshaping biodiversity worldwide. Here, we describe ethnobiodiversity informatics as a necessary expansion of biodiversity informatics - the mobilization, integration and analysis of digital biodiversity data - that represents the human systems shaping biodiversity outcomes, from behavior, trade and governance and migration history to cultural knowledge, infrastructure, ownership and management intensity, as explicit components of biodiversity analysis. Drawing on examples from food systems, human resource provisioning, medicinal plant biogeography, corporate accountability and biodiversity-health linkages, we demonstrate that ethnobiodiversity informatics is already emerging across multiple scholarly domains. We argue that ethnobiodiversity informatics offers three major benefits: more accurate biodiversity models, stronger inference about human processes that most directly shape nature and more informed decision-making. As biodiversity loss intensifies, conservation ambitions expand and large-scale data infrastructures mature, explicitly integrating human systems into biodiversity science has become both necessary and feasible.
There is a notable lack of studies investigating the mental health associations of the COVID-19 pandemic in the Global South, particularly with regard to the use of digital technologies. We aimed to investigate the cross-sectional associations of digital technology use with mental health indicators during the pandemic in 12 countries in the Global South. We used data from the UNICEF Innocenti Disrupting Harm survey. The survey included a representative sample of 11,912 internet-using children aged 12-17 in Ethiopia, Kenya, Mozambique, Namibia, Tanzania, Uganda, Cambodia, Indonesia, Malaysia, Philippines, Thailand and Vietnam. We modelled the associations of social connection during lockdown and internet use during the pandemic with six different indicators of wellbeing in each country (excluding Tanzania, where there had been no lockdowns) and in the overall sample using robust linear and logistic regression. We controlled for a range of putative socioeconomic and demographic confounders and handled missing data using multiple imputation. We did not find clear evidence for any general associations of social connection or internet use with mental health indicators during lockdown across countries. Rather, our results are complex and demonstrate that the relationship depends heavily on an individual's context; not just on the country they are living in but also on their gender, urbanicity and other factors. Nonetheless, our results highlight key focus areas for further research. Extensive further research is needed to identify ways in which technology use, including internet use, might act as a risk or protective factor relevant to mental ill health among young people in the countries of the Global South.
Power transmission insulators develop defects from sustained high-voltage stress, thermal cycling, and environmental contamination, with undetected degradation progressively leading to self-explosion, flashover, and mechanical failure. Unmanned aerial vehicle (UAV)-based defect detection faces persistent challenges from small object scales, complex background clutter, and multi-scale feature misalignment in lightweight deployable architectures. This paper proposes MLE-YOLOv11n (Multi-scale Layer aggregation and context-aware feature fusion Enhanced YOLOv11n), an enhanced detection framework built upon YOLOv11n that introduces three targeted module substitutions at distinct architectural stages. The Spatial Pyramid Pooling with Efficient Layer Aggregation Network (SPPELAN) replaces the standard Spatial Pyramid Pooling Fast (SPPF) module at the backbone terminus, preserving intermediate pooling representations at each aggregation depth to enrich multi-scale semantic features entering the neck. The Multi-branch Feature Context-Aware (MFCA) attention module replaces equal-weight concatenation at each Path Aggregation Network (PANet) neck fusion node, integrating parallel multi-branch feature enhancement with asymmetric kernels, learnable per-channel adaptive weighting, and cross-region spatial context modeling to improve neck feature discriminability. The Mamba-based Local-Long Attention (MLLA) block replaces the standard convolutional prediction unit at the P4-scale detection head, achieving linear-complexity 𝒪(N) global structural consistency verification through window-partitioned attention with an Irregular Serpentine Scan, suppressing false positives from background structures visually similar to genuine defects. Evaluated on two public benchmarks, MLE-YOLOv11n attains 92.1% mean Average Precision at IoU threshold 0.50 (mAP@50) on the China Power Line Insulator Dataset (CPLID) and 95.3% mAP@50 on the Insulator Defect Image Dataset (IDID), representing gains of 4.4 and 5.2 percentage points over baseline YOLOv11n respectively, with the largest improvements concentrated in rare and structurally complex defect categories. The framework introduces only 10.8% additional parameters (2.87 M total) at 7.2 GFLOPs while maintaining 85 frames per second (FPS) inference speed, demonstrating a competitive accuracy-efficiency trade-off for UAV-based power grid inspection, with inference speed and parameter count suitable for further evaluation on resource-constrained edge platforms in future deployment studies.
Tetracycline (TC) in wastewater impairs methanogenesis during anaerobic digestion. Although granular activated carbon (GAC) and electrical stimulation alone have been shown to mitigate TC inhibition, their combined effects remain unclear. Four up-flow anaerobic sludge blanket (UASB) reactors were operated at 5 mg/L TC, designated UASBCON (control), UASBGAC (5 g/L GAC dosed), UASBELE (0.6 V voltage applied), and UASBGAC/ELE (5 g/L GAC dosed and 0.6 V applied). Voltage was applied (Stage Ⅰ), terminated (Stage Ⅱ), and resumed (Stage Ⅲ) in UASBELE and UASBGAC/ELE. During the stable period of Stage I, UASBGAC enhanced methane production by 24.8% relative to UASBCON (p < 0.001), while UASBELE remained comparable to UASBCON. TC removal efficiency during Stage I followed the order UASBGAC (64.0%) > UASBGAC/ELE (57.5%) > UASBCON (34.5%) > UASBELE (30.1%). Neither single-amendment reactor showed a TC removal-methanogenesis trade-off. In contrast, UASBGAC/ELE suppressed methane production by 71.0% relative to UASBCON (p < 0.001), with a persistent trade-off (R2= 0.492, p < 0.001). In UASBGAC, GAC adsorption initially reduced bioavailable TC and sustained methane production. Meanwhile, UASBELE exhibited a zone-dependent community structure: TC-degrading genera were enriched at the electrode-proximal zone (cathode and anode biomass combined; 3.4 times that of bulk sludge), while methanogens predominantly resided in the bulk sludge. However, this zone-dependent pattern was attenuated in UASBGAC/ELE, where the electrode-to-bulk ratio of TC degraders declined to 2.2. GAC and electrical stimulation thus acted antagonistically rather than synergistically under TC stress. Mechanistic compatibility should guide hybrid amendment design in anaerobic systems treating antibiotic-containing wastewater.
Trade-offs between soil persistence and the provision of certain soil services and activity/functions.
Recent advances in multiomics technologies, continuous glucose monitoring, and machine learning (ML) enable precision nutrition by prediction of glycemic responses to foods. To date, no narrative reviews have focused on advances, challenges, and future directions in ML-based prediction of glycemic responses. This narrative review aimed to summarize the existing research on ML-based prediction of postprandial glycemic responses, highlight the challenges in prediction of glycemic responses, and offer directions for future research. We searched PubMed, Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Cochrane Library, Institute of Electrical and Electronics Engineers (IEEE) Xplore, and Association for Computing Machinery (ACM) Digital Library, from inception to 21 November, 2025. Studies that developed/updated/conducted external validation of an ML model and predicted the outcomes of interest (i.e., glycemic responses, peak glucose, and glucose excursions) were included. A total of 13,471 records and 2 reports identified 26 studies for inclusion. Although some studies used traditional linear or deep learning models, most existing prediction models were tree-based ML models and developed for healthy people and those with prediabetes or diabetes. Most existing models included anthropometry, biomarkers, and microbiota as personal features and nutrients/food groups as food features. However, some key food features such as food processing were overlooked. Some clinical trials reported that ML-based personalized dietary recommendations failed to yield favorable and clinically significant glycemic outcomes compared with dietary guidelines. There are challenges in ethics and compliance of algorithms, reliability of measurement instruments, prediction of multiple metabolic outcomes, trade-offs in raw data collection and feature selection, and the balance of performance metrics and clinical benefits. Most existing ML-based prediction models for postprandial glycemic responses are tree-based and incorporate personal and food features. Research on prediction of glycemic responses faces challenges that require careful consideration during study design. To improve transparency, reproducibility, and quality of future research, this review provides a minimum reporting framework with checklist items and recommendations.
Given the increasing demand for sustainable energy and self-powered devices, energy-harvesting technologies, such as triboelectric nanogenerators (TENGs), are drawing attention. To address the short charge retention in conventional polymer materials, 2D materials with high surface areas and intrinsic charge-trapping capabilities, such as MoS2, are being utilized as friction layers in TENGs. However, their power generation is too low for practical applications owing to their atomically thin nature, limiting their use as fillers in polymer-based systems. We report a 2D-material-based high-power TENG using 3D hierarchical MoS2 (3DH-MoS2) as a primary friction material. The 3DH-MoS2 is synthesized via low-temperature metal-organic chemical vapor deposition, enabling the direct growth of a uniform, large-area, 3D-nanostructured friction layer on a polymer substrate without additional processes. This 3D nanostructure increases the amount of charge-trapping sites and significantly enhances the durability of the device. The 4 × 4 cm2 3DH-MoS2-based TENG produces a maximum output voltage of 320.1 V and a power density of 0.841 mW cm-2, proving it can effectively power light-emitting diodes and a calculator, maintaining its performance over 10 000 cycles. In addition, the device can generate electricity from gas and water flow and human motion, highlighting the potential and versatility of 2D-material-based energy-harvesting systems.
Food safety monitoring increasingly requires analytical tools that are rapid, portable, and accessible beyond centralized laboratories. In this context, smartphone-based optical and spectroscopic sensors have emerged as highly promising platforms because they combine image acquisition, signal processing, data transmission, and user-friendly interfaces within a compact and widely available device. This review critically summarizes recent advances in smartphone-enabled food safety sensing, with emphasis on the underlying optical principles, hardware integration strategies, analytical workflows, and application scenarios. We discuss the major sensing modalities currently adapted to smartphone platforms, including colorimetric, fluorescence, chemiluminescence, and Raman/surface-enhanced Raman scattering approaches, and examine how camera modules, illumination control, add-on optics, 3D-printed attachments, and app-based processing pipelines influence analytical performance. Recent applications are organized by target class, covering pesticide residues, heavy metals, mycotoxins, adulterants, microbial pathogens, and spoilage-related markers. Across these categories, smartphone-assisted systems have demonstrated strong potential for decentralized screening, field deployment, and point-of-need decision support. However, the transition from laboratory demonstrations to practical deployment remains constrained by unresolved issues related to analytical standardization, validation under realistic food conditions, and integration of smartphone hardware with reliable data-processing workflows. Emerging trends indicate that future progress will be driven by artificial intelligence-assisted analysis, multimodal sensing, sustainable substrates, smart packaging integration, and connectivity with cloud- and traceability-based digital food systems. Overall, smartphone-based optical and spectroscopic sensors are evolving from low-cost imaging tools into increasingly sophisticated analytical platforms with the potential to bridge laboratory-grade detection and real-world food safety surveillance.