Explore the vast range of titles available.

Water pollution greatly impacts humans and ecosystems, so a series of policies have been enacted to control it. The first step in performing pollution control is to detect contaminants in the water. Various methods have been proposed for water quality testing, such as spectroscopy, chromatography, and electrochemical techniques. However, traditional testing methods require the utilization of laboratory equipment, which is large and not suitable for real-time testing in the field. Microfluidic devices can overcome the limitations of traditional testing instruments and have become an efficient and convenient tool for water quality analysis. At the same time, artificial intelligence is an ideal means of recognizing, classifying, and predicting data obtained from microfluidic systems. Microfluidic devices based on artificial intelligence and machine learning are being developed with great significance for the next generation of water quality monitoring systems. This review begins with a brief introduction to the algorithms involved in artificial intelligence and the materials used in the fabrication and detection techniques of microfluidic platforms. Then, the latest research development of combining the two for pollutant detection in water bodies, including heavy metals, pesticides, micro- and nanoplastics, and microalgae, is mainly introduced. Finally, the challenges encountered and the future directions of detection methods based on industrial intelligence and microfluidic chips are discussed.

With the growth of people’s environmental awareness and the encouragement of government policies, the use of electric vehicles in logistics distribution is gradually increasing. In order to solve the dual demand of customers’ simultaneous pick-up and delivery in the “last kilometer logistics”, an electric vehicle routing problem with simultaneous pick-up and delivery and time window (EVRPSPDTW) is considered from the perspective of multi-objective distribution in this paper. Firstly, a decision-making model based on distribution cost and power consumption function is established. In this model, distribution cost includes transportation cost, vehicle use cost, penalty cost of not arriving on time and charging cost. Power consumption function is the energy loss caused by air resistance, tire rolling friction and transmission system. Secondly, a multi-objective genetic algorithm (NSGA-II) optimization solution with fast nondominated ranking and elite strategy is designed, and in view of the shortcomings of traditional NSGA-II, it is proposed to complete population initialization through greedy algorithm and random rules, introduce adaptive cross-mutation strategy in the chromosome crossing and mutation stage, and design three different neighborhood operators in mutation operation based on variant fitness function. Finally, the sensitivity analysis of traffic congestion coefficient further proves the effectiveness of the proposed model and the improved algorithm.

γ-Glutamyl transpeptidase (GGT) is overexpressed in a variety of diseases, making it an important diagnostic criterion for diseases. Herein, a new fluorescence probe based on naphthalimide (Glu-MDA) was developed and employed for the rapid detection of GGT in tumor cells or samples. Alkynylated naphthalimide is the fluorescent core for excellent fluorescence response. The covalent bridging of self-immolative short linkers reduces the steric hindrance between probes and enzyme cleavage sites, which leads to improved enzymatic reaction kinetics. Glu-MDA shows a rapid response and excellent selectivity with a detection limit of 0.044 U/L. This allows the efficient detection of GGT levels in solution and cells. Simultaneously, the construction of Glu-MDA pre-stained test strips provided an innovative strategy for the qualitative detection of GGT activity, helping to detect GGT faster, more portably, and cost-effectively in various scenarios.

Biochemical oxygen demand (BOD) is an important indicator of the degree of organic pollution in water bodies. Traditional methods for BOD5 determination, although widely used, are complicated and dependent on accurate chemical measurements of dissolved oxygen. The aim of this study was to propose a facile method for predicting biochemical oxygen demand by fluorescence signals using three-dimensional fluorescence spectroscopy and parallel factor analysis in combination with a machine learning algorithm. The water samples were incubated for five days using the national standard method, during which the dissolved oxygen contents and three-dimensional fluorescence spectroscopy data were measured at eight-hour intervals. The maximum fluorescence intensity of three fluorescence components was decomposed and extracted by parallel factor analysis. The relationship between the maximum fluorescence of the three fluorescence components and the BOD5 values was established using a random forest model. The results showed that there was a good correlation between the fluorescence components and BOD values. The BOD5 values were effectively predicted by the random forest model with a high goodness of fit (R2 = 0.878) and low mean square error (MSE = 0.28). Although this method did not shorten the incubation time, successful BOD5 prediction was realized by the non-contact measurement of fluorescence signals. This avoids the complicated operation of DO determination, improves detection efficiency, and provides a convenient solution for analyzing large quantities of water samples and monitoring facile water quality.

Named entity recognition (NER) in few-shot scenarios plays a critical role in entity annotation for low-resource domains. However, existing methods are often limited to learning semantic features and intermediate representations specific to the source domain, which restricts their generalization capability when applied to unseen target domains and leads to prominent performance degradation. To address this issue, we propose a novel few-shot NER model based on contrastive learning. Specifically, the model enhances token representations through Gaussian distribution embedding and a self-attention mechanism, while adaptively optimizing the weighting parameters of the contrastive loss to achieve performance improvement. This design effectively mitigates overfitting and enhances the model’s generalization ability. Experiments on multiple datasets (including CoNLL2003, GUM, and Few-NERD) demonstrate that our approach achieves performance gains of 2.05% to 15.89% compared to state-of-the-art methods. These results confirm the effectiveness of our model in few-shot NER tasks and suggest its potential for broader application in low-resource information extraction scenarios.

To solve the problem of a large amount of fly ash accumulation and study the axial compression and bearing capacity prediction of the self-compacting fly ash concrete filled circle steel tube (SCCFST) columns, eight specimens are designed to explore the impact of concrete strength grade, internal structural measures, and additional parameters. The stress, progression of deformation, and failure mode of each specimen are observed during the loading process. The load–displacement curves, load-strain curves, characteristic load and displacement, ductility, and stiffness degradation are analyzed. The findings revealed that shear deformation occurred predominantly in the middle and upper portions of the steel tubes. Enhancing the strength of the concrete or adopting internal structural measures could increase the bearing capacity and ductility of the specimens. The peak load and ductility could be increased by up to 17.6 and 53.6%, respectively. The proposed unified calculation equation for the axial compression bearing capacity of SCCFST columns demonstrates notable reliability and precision. Furthermore, these tests offer valuable references for the engineering application of various forms of SCCFST columns, which are of significant importance in practical engineering.

Rapid urbanization in China profoundly impacts terrestrial carbon stocks, necessitating robust assessment and prediction frameworks. This study employed a coupled InVEST-PLUS model to analyse carbon stock dynamics and the economic value of carbon sinks in Jiangsu Province, a rapidly urbanizing region. We evaluated historical changes (2000–2020) and projected future impacts (2020–2040) under various land-use planning scenarios and climate emission pathways (SSP1-2.6, SSP2-4.5, SSP5-8.5). Our findings reveal a significant historical carbon stock decrease of 14.34 Tg in Jiangsu Province between 2000 and 2020, with 86.11% of this reduction occurring from 2000 to 2010. The conversion of cropland to built-up land emerged as a critical driver of carbon loss and diminished carbon sink economic value. Projections indicate that Cropland Protection (CP) and Ecological Protection (EP) scenarios are crucial for mitigating these declines. Integrating climate change, we found carbon sink losses escalated stepwise with increasing emission intensity, with soil carbon losses consistently exceeding those from vegetation. Policy effectiveness varied, with the EP scenario performing optimally under low emission conditions. This study underscores the urgent need for strategic land-use planning to safeguard carbon stocks and economic stability, thereby contributing to carbon neutrality goals.

The high exciton binding energy and lack of a positive oxidation band potential restrict the photocatalytic CO 2 reduction efficiency of lead-free Bi-based halide perovskites Cs 3 Bi 2 X 9 (X = Br, I). In this study, a sequential growth method is presented to prepare a visible-light-driven ( λ  > 420 nm) Z-scheme heterojunction photocatalyst composed of BiVO 4 nanocrystals decorated on a Cs 3 Bi 2 I 9 nanosheet for photocatalytic CO 2 reduction coupled with water oxidation. The Cs 3 Bi 2 I 9 /BiVO 4 Z-scheme heterojunction photocatalyst is stable in the gas–solid photocatalytic CO 2 reduction system, demonstrating a high visible-light-driven photocatalytic CO 2 -to-CO production rate of 17.5 μmol/(g·h), which is approximately three times that of pristine Cs 3 Bi 2 I 9 . The high efficiency of the Cs 3 Bi 2 I 9 /BiVO 4 heterojunction was attributed to the improved charge separation in Cs 3 Bi 2 I 9 . Moreover, the Z-scheme charge-transfer pathway preserves the negative reduction potential of Cs 3 Bi 2 I 9 and the positive oxidation potential of BiVO 4 . This study offers solid evidence of constructing Z-scheme heterojunctions to improve the photocatalytic performance of lead-free halide perovskites and would inspire more ideas for developing lead-free halide perovskite photocatalysts.

Traditional fall risk assessment in older adults relies on intrinsic indicators derived from demographic information, clinical assessments, and mobility tests. These measures encompass variables of mixed types, including continuous, categorical, and ordinal, and are predominantly skill-oriented assessments. Wearable sensors can provide objective gait descriptors, yet how to best integrate wearable data with intrinsic indicators for fall-risk prediction remains insufficiently studied. This study systematically compares intrinsic risk indicators with wearable-derived gait parameters for predicting fall risk in older adults. We analysed 163 participants (86 fallers, 77 non-fallers; mean age 82.6 ± 6.2 years) and evaluated thirteen feature-set combinations, namely intrinsic-only, wearable-only, and hybrid, using four classifiers [logistic regression (LR), support vector machine (SVM), random forest (RF), and artificial neural network (ANN)] under systematic cross-validation. Genetic algorithms were employed for feature selection. Mobility tests were the strongest intrinsic indicators (AUC 0.87 to 0.90, 95% CI [0.81; 0.94]). Wearable-derived gait features alone yielded moderate discrimination (LR AUC 0.83, 95% CI [0.76; 0.88]). Combining wearable features with intrinsic indicators consistently improved performance (AUC 0.87 to 0.93 across combinations). The optimal combined signature achieved an LR AUC of 0.94 (95% CI [0.91; 0.97]; F1 = 0.871). These findings indicate that wearable-derived gait features complement, rather than replace, intrinsic risk indicators, and that their combination provides the most effective fall risk assessment.

To control COVID-19 pandemic, complete lockdown was initiated in 2020. We investigated the impact of lockdown on tertiary-level academic performance, by comparing educational outcomes amongst first-year students during second semester of their medical course prior to and during lockdown. Evidence : The demographics, including educational outcomes of the two groups were not significantly different during semester one (prior to the lockdown). The academic performance amongst women was better than men prior to lockdown. However, the scores were improved significantly for both sexes during lockdown in 2020, following the complete online teaching, compared to that in 2019, showing no significant difference between men and women in 2020, for English and Chinese History. There were significant different scores between men and women in lab-based Histology Practice in 2019 (in-person tuition) and 2020 (online digital tuition), although only a significant improvement in women was observed between 2019 and 2020. Implication : the forced change to online delivery of the second semester of the first-year medical program in 2020 due to the COVID-19 pandemic did not result in any decline in assessment outcomes in any of the subjects undertaken. We believe extensive online digital media should continue to be available to students in future.