Search

Objective: To predict risk of pre-eclampsia (PE) in women using machine learning (ML) algorithms, based on electronic health records (EHR) collected at the early second trimester.Study Design: A total of 3759 cases of pregnancy who received antenatal care at Xinhua hospital Chongming branch Affiliated to Shanghai Jiaotong University were included in this retrospective EHR-based study. Thirty-eight candidate clinical parameters routinely available at the first visit in antenatal care were collected by manual chart review. Logistic regression (LR), random forest (RF), support vector machine (SVM) and extreme gradient boosting (XGBoost) were used to construct the prediction model. Features that contributed to the model predictions were identified using XGBoost.Outcome Measures: The performance of ML models to predict women at risk of PE was quantified in terms of accuracy, precision, recall, false negative score, f1_score, brier score and the area under the receiver operating curve (auROC).Results: The XGboost model had the best prediction performance (accuracy = 0.920, precision = 0.447, recall = 0.789, f1_score = 0.571, auROC = 0.955). The most predictive feature of PE development was fasting plasma glucose, followed by mean blood pressure and body mass index. An easy-to-use model that a patient could answer independently still enabled accurate prediction, with auROC of 0.83.Conclusion: risk of PE development can be predicted with excellent discriminative ability using ML algorithms based on EHR collected at the early second trimester. Future studies are needed to assess the real-world clinical utility of the model. [ABSTRACT FROM AUTHOR]

Landfills have been the most common methods of organized waste disposal in China posing an incredible groundwater pollution threat. Dissolved organic matter (DOM) can be used to trace the source, species and migration of contaminants in groundwater, and the investigations of its composition, structure, and distribution play a role in environmental protection. This study investigated the DOM source, composition, and molecular structure in groundwater at landfills for different years of operation, and explored the dynamics of groundwater DOM evolution over time, usingmodern spectroscopy in combination with multivariate statistical analysis. The results showed that DOM in landfill groundwater was initially dominated by outputs from microbial activities, and this was followed by autogenous terrigenous input. In the early stages of landfilling, the DOM of microbial origin was significant; however, towards the late stages of landfilling, the presence of microbial DOM has weakened. The groundwater DOM with short landfill times were mainly composed of newly produced tryptophan and tyrosine, which had low humification, aromaticity, and molecular weight. Microbial activity was strong, and while there were initial, significant differences between sampling points, evidence of its presence could be used for early warning of contamination and monitoring should be conducted for its presence. Microbial activity weakened with longer landfill operation time, landfill waste tended towards stability, and the DOM in groundwater with high humification, aromaticity and molecular weight, was able to reduce the landfill impact on groundwater.

This study explores the impact of facility farming on the origin and composition of groundwater and on the distribution characteristics of inorganic salts, heavy metals, and dissolved organic matter in groundwater in a typical greenhouse vegetable planting area in North China plains. The methods used include conventional analysis, UV-Visible, and fluorescence spectroscopy combined with parallel factor analysis, principle components analysis, and two-dimensional (2D) hetero-spectral correlation spectra techniques. The results showed that the hydrochemistry type of groundwater in the facility vegetable field was Cl-SO 4 , the nitrogen content was higher than the National Groundwater V Class Quality Standard (GB/T14848-93), and the heavy metal content was lower than the National Drinking Water Standard (GB5749-2006). The nitrification process was blocked due to a strongly reducing atmosphere and shallow depth in groundwater, which caused high concentrations of NH 4 + -N. Part of the toxic sulfur-containing metal content increased after the application of inorganic fertilizers. The sources of dissolved organic matter in groundwater were similar; the organic manure components which leach into groundwater were recently produced by microbial metabolism. The main components of dissolved organic matter were small-molecule-biodegradable protein-like substances and the fulvic-like content binding to them. The fulvic-like content that bound to protein-like content in vegetable planting groundwater was stable, while the protein-like content was greatly influenced by the cultivation process of the organic manure application. In the process of vegetable planting, organic manure should be applied scientifically and the supervision of the planting area is crucial.

To analyze the characteristics of shallow groundwater pollution in farmland ecosystem in south China and geochemical behavior of major pollutants in the environment, taking the typical farmland in Taihu River Basin as the object, combined with conventional analysis method of the water quality, fluorescence spectrum technology and multivariate statistical analysis method, the characteristics of the water chemistry in shadow groundwater and the distribution of the dissolved organic matter and the heavy metals were studied. The reasons for all the phenomena were researched as well. The result showed that, water body in the study area showed the reducing atmosphere, the main chemical types of shallow groundwater were HCO 3 -Ca·Na, and compared to the contents of NH 3 - -N and NH 2 - -N, the concentration of NH 4 + -N was higher. Dissolved organic matter of the shallow groundwater was composed of humic acid, tryptophan and tyrosine. And they were mainly derived from terrestrial and biological source. The distribution was affected by the flow direction of the surface water supplies and groundwater. In the detection of nine kinds of heavy metal elements, the average concentrations of Fe and Cr were higher than the national groundwater environmental quality in class Ⅲ standard. And the contents of Fe, Cr, Mn, Zn elements at some points exceeded the standard. The results showed that fluorescent organic matter characteristics could better reflect the status of shallow groundwater pollution, in which the kind of protein-like substances was closely related to the ammonia nitrogen. The heavy metal elements Cu and Ni were associated with the dissolved organic matter and they were mainly complexed in small molecular fluorescent organic matter.

The evolution of water DOC and COD, and the source, chemical structure, humification degree and redox of dissolved organic matter (DOM) in a constructed wetland of Xiao River, Hebei, was investigated by 3D excitation--emission matrix fluorescence spectroscopy coupled with ultraviolet spectroscopy and chemical reduction, in order to explore the geochemical processes and environmental effects of DOM. Although DOC contributes at least 60% to COD, its decrease in the constructed wetland is mainly caused by the more extensive degradation of elements N, H, S, and P than C in DOM, and 65% is contributed from the former. DOM is mainly consisted of microbial products based on proxies f470/520 and BIX, indicating that DOM in water is apparently affected by microbial degradation. The result based on PARAFAC model shows that DOM in the constructed wetland contains protein-like and humus-like components, and Fulvic- and humic-like components are relatively easier to degrade than protein-like components. Fulvic- and humic-like components undergo similar decomposition in the constructed wetland. A common source of chromophoric dissolved organic matter (CDOM) and fluorescent dissolved organic matter (FDOM) exists; both CDOM and FDOM are mainly composed of a humus-like material and do not exhibit selective degradation in the constructed wetland. The proxies E2 /E3, A240-400, r(A, C) and HIX in water have no changes after flowing into the constructed wetland, implying that the humification degree of DOM in water is hardly affected by wet constructed wetland. However, the constructed wetland environment is not only beneficial in forming the reduced state of DOM, but also facilitates the reduction of ferric. It can also improve the capability of DOM to function as an electron shuttle. This result may be related to the condition that the aromatic carbon of DOM can be stabilized well in the constructed wetland.

The spectral characteristic of dissolved organic matter (DOM) in Xiaohe River, Hebei, was investigated by fluorescence spectroscopy, ultraviolet-visible absorption spectroscopy, and basic chemical water quality indicators. The data was then statistical analyzed using principal component analysis and correlation analysis method. The result based on 3D excitation-emission matrix fluorescence spectroscopy showed that DOM in Xiaohe River contained both protein-like and humus-like components. DOM and N-containing compounds were obviously correlated with COD, especially between NH4+ -N and humic-like component, indicating that COD of water in Xiaohe River can be reduced by removing NH4+ -N and DOM, which could be good indicators for monitoring water quality in the future. The relative content of protein-like component reduces gradually along the downstream, while that of humic-like component showed an increasing trend. DOM in samples S1 and S2 was mainly consisted of humic-like components with larger molecular weight and higher aromaticity, while that in samples S3 and S6 was mainly consisted of protein-like components with smaller molecular weight, lower aromaticity, which are easier to be degraded. Therefore, in order to enhance the remove of refractory humic-like substances, sewage treatment plants of S1 and S2 or improved membrane treatment equipment with better removal effect of macromolecules should be provide. On the other hand, the anaerobic and aerobic biological treatment processes should be optimized in S3 and S6, so as to better remove these degradable protein-like substances.