Search

Dinitrogen pentoxide (N 2 O 5 ) plays an essential role in tropospheric chemistry, serving as a nocturnal reservoir of reactive nitrogen and significantly promoting nitrate formations. However, identifying key environmental drivers of N 2 O 5 formation remains challenging using traditional statistical methods, impeding effective emission control measures to mitigate NO x -induced air pollution. Here, we adopted machine learning assisted by steady-state analysis to elucidate the driving factors of N 2 O 5 before and during the 2022 Winter Olympics (WO) in Beijing. Higher N 2 O 5 concentrations were observed during the WO period compared to the Pre-Winter-Olympics (Pre-WO) period. The machine learning model accurately reproduced ambient N 2 O 5 concentrations and showed that ozone (O 3 ), nitrogen dioxide (NO 2 ), and relative humidity (RH) were the most important driving factors of N 2 O 5 . Compared to the Pre-WO period, the variation in trace gases (i.e., NO 2 and O 3 ) along with the reduced N 2 O 5 uptake coefficient was the main reason for higher N 2 O 5 levels during the WO period. By predicting N 2 O 5 under various control scenarios of NO x and calculating the nitrate formation potential from N 2 O 5 uptake, we found that the progressive reduction of nitrogen oxides initially increases the nitrate formation potential before further decreasing it. The threshold of NO x was approximately 13 ppbv, below which NO x reduction effectively reduced the level of night-time nitrate formations. These results demonstrate the capacity of machine learning to provide insights into understanding atmospheric nitrogen chemistry and highlight the necessity of more stringent emission control of NO x to mitigate haze pollution.

Diabetic retinopathy (DR) is one of the leading causes of adult blindness in the United States. Although studies applying traditional statistical methods have revealed that heavy metals may be essential environmental risk factors for diabetic retinopathy, there is a lack of analyses based on machine learning (ML) methods to adequately explain the complex relationship between heavy metals and DR and the interactions between variables. Based on characteristic variables of participants with and without DR and heavy metal exposure data obtained from the NHANES database (2003-2010), a ML model was developed for effective prediction of DR. The best predictive model for DR was selected from 11 models by receiver operating characteristic curve (ROC) analysis. Further permutation feature importance (PFI) analysis, partial dependence plots (PDP) analysis, and SHapley Additive exPlanations (SHAP) analysis were used to assess the model capability and key influencing factors. A total of 1042 eligible individuals were randomly assigned to two groups for training and testing set of the prediction model. ROC analysis showed that the k-nearest neighbour (KNN) model had the highest prediction performance, achieving close to 100% accuracy in the testing set. Urinary Sb level was identified as the critical heavy metal affecting the predicted risk of DR, with a contribution weight of 1.730632 ± 1.791722, which was much higher than that of other heavy metals and baseline variables. The results of the PDP analysis and the SHAP analysis also indicated that antimony (Sb) had a more significant effect on DR. The interaction between age and Sb was more significant compared to other variables and metal pairs. We found that Sb could serve as a potential predictor of DR and that Sb may influence the development of DR by mediating cellular and systemic senescence. The study revealed that monitoring urinary Sb levels can be useful for early non-invasive screening and intervention in DR development, and also highlighted the important role of constructed ML models in explaining the effects of heavy metal exposure on DR., (© 2024. The Author(s).)

Purpose: Arterial spin labeling (ASL) perfusion MRI is a noninvasive technique for measuring cerebral blood flow (CBF) in a quantitative manner. A technical challenge in ASL MRI is data processing because of the inherently low signal-to-noise-ratio (SNR). Deep learning (DL) is an emerging machine learning technique that can learn a nonlinear transform from acquired data without using any explicit hypothesis. Such a high flexibility may be particularly beneficial for ASL denoising. In this paper, we proposed and validated a DL-based ASL MRI denoising algorithm (DL-ASL)., Methods: The DL-ASL network was constructed using convolutional neural networks (CNNs) with dilated convolution and wide activation residual blocks to explicitly take the inter-voxel correlations into account, and preserve spatial resolution of input image during model learning., Results: DL-ASL substantially improved the quality of ASL CBF in terms of SNR. Based on retrospective analyses, DL-ASL showed a high potential of reducing 75% of the original acquisition time without sacrificing CBF measurement quality., Conclusion: DL-ASL achieved improved denoising performance for ASL MRI as compared with current routine methods in terms of higher PSNR, SSIM and Radiologic scores. With the help of DL-ASL, much fewer repetitions may be prescribed in ASL MRI, resulting in a great reduction of the total acquisition time., (Copyright © 2020 Elsevier Inc. All rights reserved.)

Motivation Large-scale clinical proteomics datasets of infectious pathogens, combined with antimicrobial resistance outcomes, have recently opened the door for machine learning models which aim to improve clinical treatment by predicting resistance early. However, existing prediction frameworks typically train a separate model for each antimicrobial and species in order to predict a pathogen's resistance outcome, resulting in missed opportunities for chemical knowledge transfer and generalizability. Results We demonstrate the effectiveness of multimodal learning over proteomic and chemical features by exploring two clinically relevant tasks for our proposed deep learning models: drug recommendation and generalized resistance prediction. By adopting this multi-view representation of the pathogenic samples and leveraging the scale of the available datasets, our models outperformed the previous single-drug and single-species predictive models by statistically significant margins. We extensively validated the multi-drug setting, highlighting the challenges in generalizing beyond the training data distribution, and quantitatively demonstrate how suitable representations of antimicrobial drugs constitute a crucial tool in the development of clinically relevant predictive models. Availability and implementation The code used to produce the results presented in this article is available at https://github.com/BorgwardtLab/MultimodalAMR. [ABSTRACT FROM AUTHOR]

Forest biomass is a key index to comprehend the changes of ecosystem productivity and forest growth and development. Accurate acquisition of single tree scale biomass information is of great significance to the protection, management and monitoring of forest resources. LiDAR technology can penetrate the forest canopy and obtain information on the vertical structure of the forest. Aerial photography technology has the advantages of low cost and high speed, and can obtain information on the horizontal structure of the forest. Therefore, in this study, multispectral imagery and LiDAR data were integrated, and a part of the Zengcheng Forest Farm in Guangdong Province was selected as the study area. Large-scale and high-precision Eucalyptus biomass estimation research was gradually carried out by screening influencing factors and establishing models. This study compared and analysed the performance of multiple stepwise regression methods, random forest algorithms, support vector machine algorithms and decision tree algorithms for Eucalyptus biomass estimation to determine the best method for Eucalyptus biomass estimation. The results demonstrated that the accuracy of the model established by the machine learning method was higher than that of the linear regression model, and in the machine learning model, the random forest model had the best performance on both the training set (R2 = 0.9346, RMSE = 8.8399) and the test set (R2 = 0.8670, RMSE = 15.0377). RF was more suitable for the biomass estimation of Eucalyptus in this study. The spatial resolution of Eucalyptus biomass distribution was 0.05 m in this study, which had higher accuracy and was more accurate. It can provide data reference for the details about biomass distribution of Eucalyptus in the majority of provinces, and has certain practical reference significance. [ABSTRACT FROM AUTHOR]

Online reviews have become one of the most influential persuasive messages concerning decisions making. Researchers have explored different aspects of online reviews and the characteristics of reviewers using various methods; however, few studies have focused on consumers who do not post online reviews after purchasing, resulting in a gap in the research. This study aims to identify consumers' characteristics and analyze the importance of consumers in different online review scenarios. We employed a fusion analysis framework, which used machine learning to determine the value of different types of consumers; in addition, we quantitatively estimate the relationship between variables and consumer types using econometrics modelling. This research used real consumer data from three data resources, solving problems related to the use of limited questionnaire datasets and single-source data. Through rigorous analysis, we demonstrate that lurkers are more valuable than posters because they can be easily served and create more profit; thus, researchers should consider those who do not post online reviews rather than focusing only on the influence of online reviews and posters. Our findings also provide managerial implications for precision marketing by indicating that e-marketers should employ various marketing strategies and pay attention to different types of consumers, especially lurkers. [ABSTRACT FROM AUTHOR]

[Display omitted] • The stratification of recurrence risk is critical for patients with ESCC. • Radiomics achieved an AUC of 0.840 in recurrence risk prediction for ESCC. • Nomogram achieved an AUC of 0.860 in recurrence risk prediction for ESCC. • Nomogram will help guide individualized treatment for ESCC. Adjuvant treatments are valuable to decrease the recurrence rate and improve survival for early-stage cervical cancer patients (ESCC), Therefore, recurrence risk evaluation is critical for the choice of postoperative treatment. A magnetic resonance imaging (MRI) based radiomics nomogram integrating postoperative adjuvant treatments was constructed and validated externally to improve the recurrence risk prediction for ESCC. 212 ESCC patients underwent surgery and adjuvant treatments from three centers were enrolled and divided into the training, internal validation, and external validation cohorts. Their clinical data, pretreatment T2-weighted images (T2WI) were retrieved and analyzed. Radiomics models were constructed using machine learning methods with features extracted and screen from sagittal and axial T2WI. A nomogram for recurrence prediction was build and evaluated using multivariable logistic regression analysis integrating radiomic signature and adjuvant treatments. A total of 8 radiomic features were screened out of 1020 extracted features. The extreme gradient boosting (XGboost) model based on MRI radiomic features performed best in recurrence prediction with an area under curve (AUC) of 0.833, 0.822 in the internal and external validation cohorts, respectively. The nomogram integrating radiomic signature and clinical factors achieved an AUC of 0.806, 0.718 in the internal and external validation cohorts, respectively, for recurrence risk prediction for ESCC. In this study, the nomogram integrating T2WI radiomic signature and clinical factors is valuable to predict the recurrence risk, thereby allowing timely planning for effective treatments for ESCC with high risk of recurrence. [ABSTRACT FROM AUTHOR]

• Predicted values calculated by different ML algorithms fit the measured values well. • Different ML models performed better in sediment discharge simulation. • XGB outperformed the other ML models on the training set. • RFR and SVR had best prediction ability in the AR and AS simulations, respectively. • Rainfall and runoff factor contributed most to the AR and AS changes, respectively. Predicting runoff and sediment discharge changes and understanding their drivers can provide insight into hydrological processes and aid in watershed management planning. In recent years, machine learning (ML) has become a popular tool for predicting runoff and sediment discharge in large rivers. In this study, six ML models, support vector regression (SVR), artificial neural network (ANN), random forest regression (RFR), extreme gradient boosting (XGB), k-nearest neighbor regression (kNNR), and bootstrap aggregating regression (BAR), were applied to simulate watershed runoff and sediment discharge. Notably, the Shapley additive explanation (SHAP) method was applied to identify the contribution of driving factors to runoff and sediment discharge generation. The results showed that the annual runoff and sediment discharge values predicted by the six ML algorithms fit well to the measured values. In particular, all models performed better in the sediment discharge simulation than in the runoff simulation. Combining the four evaluation indexes and their scores, the model performance hierarchy in runoff simulation was XGB > RFR > BAR > SVR > ANN > kNNR, while the model performance hierarchy in sediment discharge simulation was SVR > XGB > BAR > RFR > kNNR > ANN. After excluding the effects of sample order on the simulation results, XGB showed less empirical risk and better generalizability, while RFR and SVR showed better prediction ability in the annual runoff and annual sediment discharge simulations, respectively. Additionally, the rainfall factor contributed most to runoff changes, while the runoff factor contributed most to sediment discharge changes. Overall, the results of the present study can help assess river safety and promote sustainable river development. [ABSTRACT FROM AUTHOR]

This study utilizes transfer learning (TL) to enhance long-term atmospheric corrosion predictions. Using a Fe/Cu galvanic-type sensor, we gathered data in a controlled SAE J2334 salt spray setting and transferred this to an uncontrolled outdoor environment. Among TL methods tested, freezing the initial layer and fine-tuning others at a lower rate was most effective. The approach excelled at forecasting outdoor corrosion behaviour using a limited dataset. This approach could provide a solution to extrapolate results from controlled corrosion tests to unpredictable outdoor conditions and addressing data scarcity in machine learning modelling in the context of atmospheric corrosion. • TL to extrapolate SAE J2334 data to outdoor settings using Fe/Cu ACM corrosion sensor. • Performance assessed by accuracy in predicting split I ACM intervals. • Optimal TL: freeze initial layer; fine-tune at 1 × 10−5 learning rate. • TL superior to benchmark models in cases of scarce outdoor training data. • TL model excels in detecting I ACM peaks vs. base model for a 1-week outdoor training data case study. [ABSTRACT FROM AUTHOR]