Dynamic variability of methane and controlling mechanisms in the northern North Yellow Sea.

Distributions, influencing factors and fluxes of dissolved methane in the North Yellow Sea, near the Yalu River estuary, China

Regional distribution and environmental regulation mechanism of nitrous oxide in the Bohai Sea and North Yellow Sea: A preliminary study

In the Netherlands, concerned citizens have proposed reducing train speed as an effective measure to mitigate annoyance caused by railway-induced vibrations. In the present study the relationship between train speed and other influencing parameters (e.g. axle load, wheel roughness), and ground vibrations was investigated using measurements, at different locations, of ground vibrations caused by the passage of regular freight trains and a test train at different speeds. Measurements have been analysed using multivariate regression models and a random decision forest model. The prevailing uncertainties have also been measured using normalized mean deviation between the model predicted value and the actual value. A comparison of results demonstrates that a 'trained and tested' random forest model has certain predictive advantages: i) mean deviation between predicted and actual value is found to be the lowest with random forest model; ii) the random forest model considers all available parameters in the dataset, thus simulating the real situation more closely. However, the model is very location-specific and must therefore be used with caution. In general it is observed that a decrease in train speed results in the reduction of measured vibration levels.

To predict the potential geographic distribution of Oncomelania hupensis in Yunnan Province using random forest (RF) and maximum entropy (MaxEnt) models, so as to provide insights into O. hupensis surveillance and control in Yunnan Province. The O. hupensis snail survey data in Yunnan Province from 2015 to 2016 were collected and converted into O. hupensis snail distribution site data. Data of 22 environmental variables in Yunnan Province were collected, including twelve climate variables (annual potential evapotranspiration, annual mean ground surface temperature, annual precipitation, annual mean air pressure, annual mean relative humidity, annual sunshine duration, annual mean air temperature, annual mean wind speed, ≥ 0 ℃ annual accumulated temperature, ≥ 10 ℃ annual accumulated temperature, aridity and index of moisture), eight geographical variables (normalized difference vegetation index, landform type, land use type, altitude, soil type, soil textureclay content, soil texture-sand content and soil texture-silt content) and two population and economic variables (gross domestic product and population). Variables were screened with Pearson correlation test and variance inflation factor (VIF) test. The RF and MaxEnt models and the ensemble model were created using the biomod2 package of the software R 4.2.1, and the potential distribution of O. hupensis snails after 2016 was predicted in Yunnan Province. The predictive effects of models were evaluated through cross-validation and independent tests, and the area under the receiver operating characteristic curve (AUC), true skill statistics (TSS) and Kappa statistics were used for model evaluation. In addition, the importance of environmental variables was analyzed, the contribution of environmental variables output by the models with AUC values of > 0.950 and TSS values of > 0.850 were selected for normalization processing, and the importance percentage of environmental variables was obtained to analyze the importance of environmental variables. Data of 148 O. hupensis snail distribution sites and 15 environmental variables were included in training sets of RF and MaxEnt models, and both RF and MaxEnt models had high predictive performance, with both mean AUC values of > 0.900 and all mean TSS values and Kappa values of > 0.800, and significant differences in the AUC (t = 19.862, P < 0.05), TSS (t = 10.140, P < 0.05) and Kappa values (t = 10.237, P < 0.05) between two models. The AUC, TSS and Kappa values of the ensemble model were 0.996, 0.954 and 0.920, respectively. Independent data verification showed that the AUC, TSS and Kappa values of the RF model and the ensemble model were all 1, which still showed high performance in unknown data modeling, and the MaxEnt model showed poor performance, with TSS and Kappa values of 0 for 24%(24/100) of the modeling results. The modeling results of 79 RF models, 38 MaxEnt models and their ensemble models with AUC values of > 0.950 and TSS values of > 0.850 were included in the evaluation of importance of environmental variables. The importance of annual sunshine duration (SSD) was 32.989%, 37.847% and 46.315% in the RF model, the MaxEnt model and their ensemble model, while the importance of annual mean relative humidity (RHU) was 30.947%, 15.921% and 28.121%, respectively. Important environment variables were concentrated in modeling results of the RF model, dispersed in modeling results of the MaxEnt model, and most concentrated in modeling results of the ensemble model. The potential distribution of O. hupensis snails after 2016 was predicted to be relatively concentrated in Yunnan Province by the RF model and relatively large by the MaxEnt model, and the distribution of O. hupensis snails predicted by the ensemble model was mostly the joint distribution of O. hupensis snails predicted by RF and MaxEnt models. Both RF and MaxEnt models are effective to predict the potential distribution of O. hupensis snails in Yunnan Province, which facilitates targeted O. hupensis snail control.

Machine Learning Offers Exciting Potential for Predicting Postprocedural Outcomes: A Framework for Developing Random Forest Models in IR

Existing risk prediction models for hospitalized heart failure patients are limited. We identified patients hospitalized with a diagnosis of heart failure between 7 May 2013 and 26 April 2022 from a large academic, quaternary care medical centre (training cohort). Demographics, medical comorbidities, vitals, and labs were collected and were used to construct random forest machine learning models to predict in‐hospital mortality. Models were compared with logistic regression, and to commonly used heart failure risk scores. The models were subsequently validated in patients hospitalized with a diagnosis of heart failure from a second academic, community medical centre (validation cohort). The entire cohort comprised 21 802 patients, of which 14 539 were in the training cohort and 7263 were in the validation cohort. The median age (25th–75th percentile) was 70 (58–82) for the entire cohort, 43.2% were female, and 6.7% experienced inpatient mortality. In the overall cohort, 7621 (35.0%) patients had heart failure with reduced ejection fraction (EF ≤ 40%), 1271 (5.8%) had heart failure with mildly reduced EF (EF 41–49%), and 12 910 (59.2%) had heart failure with preserved EF (EF ≥ 50%). Random forest models in the validation cohort demonstrated a c‐statistic (95% confidence interval) of 0.96 (0.95–0.97), sensitivity (SN) of 87.3%, and specificity (SP) of 90.6% for the prediction of in‐hospital mortality. Models for those with HFrEF demonstrated a c‐statistic of 0.96 (0.94–0.98), SN 88.2%, and SP 91.0%, and those for patients with HFpEF showed a c‐statistic of 0.95 (0.93–0.97), SN 87.4%, and SP 89.5% for predicting in‐hospital mortality. The random forest model significantly outperformed logistic regression (c‐statistic 0.87, SN 75.9%, and SP 86.9%), and current existing risk scores including the Acute Decompensated Heart Failure National Registry risk score (c‐statistic of 0.70, SN 69%, and SP 62%), and the Get With the Guidelines‐Heart Failure risk score (c‐statistic 0.69, SN 67%, and SP 63%); P < 0.001 for comparison. Machine learning models built from commonly recorded patient information can accurately predict in‐hospital mortality among patients hospitalized with a diagnosis of heart failure.

The purpose of this study is to map spatial metabolite differences across three molecular subgroups of glial tumors, defined by the IDH1/2 mutation and 1p19q-co-deletion, using magnetic resonance spectroscopy. This work reports a new MR spectroscopy based classification algorithm by applying a radiomics analytics pipeline. 65 patients received anatomical and chemical shift imaging (5 × 5 × 20mm voxel size). Tumor regions were segmented and registered to corresponding spectroscopic voxels. Spectroscopic features were computed (n = 860) in a radiomic approach and selected by a classification algorithm. Finally, a random forest machine-learning model was trained to predict the molecular subtypes. A cluster analysis identified three robust spectroscopic clusters based on the mean silhouette widths. Molecular subgroups were significantly associated with the computed spectroscopic clusters (Fisher's Exact test p < 0.01). A machine-learning model was trained and validated by public available MRS data (n = 19). The analysis showed an accuracy rate in the Random Forest model by 93.8%. MR spectroscopy is a robust tool for predicting the molecular subtype in gliomas and adds important diagnostic information to the preoperative diagnostic work-up of glial tumor patients. MR-spectroscopy could improve radiological diagnostics in the future and potentially influence clinical and surgical decisions to improve individual tumor treatment.

Introduction: The lengthy and costly process of drug development can be expedited through drug repositioning (DR), a strategy that identifies new therapeutic targets using existing products. Supervised machine learning (SML) models, incorporating interaction networks, offer a promising approach for DR. This study aims to systematically review and meta-analyze SML models predicting DR, identifying key characteristics influencing their performance. Methodology: A systematic review was conducted to identify SML models that used networks to predict DR, which were evaluated by comparing their performance through a random-effects meta-analysis. Results: 19 studies were included in the qualitative synthesis and 17 in the quantitative evaluation, The Random Forest (RF) model emerged as the predominant classifier (63%), yielding the highest performance in AUC ROC comparisons (overall value: 0.91, 95% CI: 0.86 – 0.96). Validation efforts in 18 studies confirmed the predictions of the SML models, affirming the proposed drugs. The incorporation of chemical structure in model training was found to enhance performance by aiding in prediction discrimination. Conclusion: SML models can predict DR, the RF model was the most widely used SML model with the best performance results, which underscores the potential use of FR models for predicting DR using network form biomedical information.

Reliable estimates of crop nitrogen (N) uptake and offtake are critical in estimating N balances, N use efficiencies and potential losses to the environment. Calculation of crop N uptake and offtake requires estimates of yield of crop product (e.g. grain or beans) and crop residues (e.g. straw or stover) and the N concentration of both components. Yields of crop products are often reasonably well known, but those of crop residues are not. While the harvest index (HI) can be used to interpolate the quantity of crop residue from available data on crop product yields, harvest indices are known to vary across locations, as do N concentrations of residues and crop products. The increasing availability of crop data and advanced statistical and machine learning methods present us with an opportunity to move towards more locally relevant estimates of crop harvest index and N concentrations using more readily available data. The aim of this study was to investigate whether improved estimates of maize crop HI and N concentrations of crop products and crop residues can be based on crop data available at the global scale, such as crop yield, fertilizer application rates and estimates of yield potential. Experiments from 1487 different locations conducted across 31 countries were used to test various prediction models. Predictions from mixed-effects models and random forest machine learning models provided reasonable levels of prediction accuracy (R 2 of between 0.33 and 0.68), with the random forest method having greater accuracy. Although the mixed-effects prediction models had lower prediction accuracy than random forest, they did provide better interpretability. Selection of which method to use will depend on the objective of the user. Here, the random forest and mixed-effects methods were applied to N in maize, but could equally be applied to other crops and other nutrients, if data becomes available. This will enable obtaining more locally relevant estimates of crop nutrient offtake to improve estimates of nutrient balances and nutrient use efficiency at national, regional or global levels, as part of strategies towards more sustainable nutrient management. • Predictions of crop nitrogen (N) removal at national to sub-national scales will improve global nutrient budgeting. • Maize harvest index (HI), crop product N concentration (CPN) and crop residue N concentrations (CRN) were analyzed. • Predictor variables included crop product yield, fertilizer nitrogen application rates and yield potential. • Random forest (RF) models had greater prediction accuracy of HI, CPN and CRN compared with mixed-effects (ME) models. • Both methods could predict HI, CPN and CRN, but ME models were easier to interpret and extrapolate than RF models.

In cave mines, wet inrushes occur when there is an uncontrolled inflow of fine, wet material from drawpoints. Currently, uncertainty exists regarding the spatial-temporal pattern and severity of inrush incidents. This uncertainty arises from the limited understanding of wet inrush mechanisms within the complex conditions of a cave mine. In this study, the existing gaps in knowledge around the spatial and temporal patterns of inrush incidents were addressed using machine learning techniques. A random forest (RF) model was employed to analyse the inrush database collected at the Deep Ore Zone mine over several years. The conceptual understanding of inrush mechanisms and triggers, along with historical evidence, was employed to establish an initial set of key inrush variables to be used in the RF model. The developed RF model demonstrated promising performance with an accuracy of 85%. The feature importance results indicated that previous inrush history, fragment size, draw rate (short term and long term), differential draw index (short term and long term) and history of inrush at neighbouring drawpoints had the highest impact on inrush susceptibility. The insights gained provide an improved assessment of inrush susceptibility, thereby improving the strategies employed to mitigate inrush risk.

BackgroundImmune checkpoint inhibitors (ICIs) have become a pillar of cancer therapy. However, they are associated with immune-related adverse events (irAEs), which pose significant barrier to ICI usage. ICI-induced inflammatory arthritis...

Prostate cancer is a common malignancy in men, and accurately distinguishing between benign and malignant nodules at an early stage is crucial for optimizing treatment. Multimodal imaging (such as ADC and T2) plays an important role in the diagnosis of prostate cancer, but effectively combining these imaging features for accurate classification remains a challenge. This retrospective study included MRI data from 199 prostate cancer patients. Radiomic features from both the tumor and peritumoral regions were extracted, and a random forest model was used to select the most contributive features for classification. Three machine learning models-Random Forest, XGBoost, and Extra Trees-were then constructed and trained on four different feature combinations (tumor ADC, tumor T2, tumor ADC+T2, and tumor + peritumoral ADC+T2). The model incorporating multimodal imaging features and peritumoral characteristics showed superior classification performance. The Extra Trees model outperformed the others across all feature combinations, particularly in the tumor + peritumoral ADC+T2 group, where the AUC reached 0.729. The AUC values for the other combinations also exceeded 0.65. While the Random Forest and XGBoost models performed slightly lower, they still demonstrated strong classification abilities, with AUCs ranging from 0.63 to 0.72. SHAP analysis revealed that key features, such as tumor texture and peritumoral gray-level features, significantly contributed to the model's classification decisions. The combination of multimodal imaging data with peritumoral features moderately improved the accuracy of prostate cancer classification. This model provides a non-invasive and effective diagnostic tool for clinical use and supports future personalized treatment decisions.

Land-terminating glaciers in Greenland and Iceland are sources of methane (CH4) to the atmosphere1,2,3. CH4 is produced through microbial methanogenesis underneath the ice, transported dissolved in subglacial meltwater to the margin where the gas is emitted to the atmosphere via degassing4. However, sparse empirical data exist about the spatial distribution of subglacial CH4 production and emission in other glaciated regions of the world, limiting our understanding of its regional and global importance in atmospheric carbon budgets and its possible role in the climate system.In August 2022, we conducted fieldwork at three outlet glaciers - Dusty, Kluane and Donjek glaciers - of the St. Elias Icefields in Yukon, Canada, to investigate if these alpine glaciers are also sources of CH4 emissions to the atmosphere. The glaciers were chosen due to the absence of proglacial lakes and the presence of meltwater upwellings at the glacier termini, which were accessed via helicopter.In-situ extracted dissolved CH4 and CO2 concentrations were measured in the field with a portable greenhouse gas analyzer. Additionally, extracted gas was collected in exetainers for concentration measurements via gas chromatography and in Tedlar gas bags for stable carbon and hydrogen isotope analyses of CH4 to decipher its origin. Further, water samples were collected for geochemical analyses. At Dusty glacier, we performed a high-intensity sampling campaign over 10 hours and continuous measurements of dissolved CH4 concentrations with a custom-made low-cost and low-power dissolved CH4 sensor5 to study changes in dissolved gas concentrations, stable isotopic signatures and water chemistry during the rising limb of the diurnal discharge curve.In-situ measured CH4 and CO2 concentrations yielded significantly elevated CH4 and depleted CO2 levels in the meltwater of all three glaciers. Discrete gas samples confirmed dissolved CH4 concentrations 45x, 135x and 250x above the atmospheric equilibrium concentration (3.6 nmol L-1) in the meltwater of Dusty, Kluane and Donjek glaciers, respectively. First measurements of stable carbon and hydrogen isotope values of CH4 showed enrichment in 13C, while 2H was depleted compared to atmospheric CH4, at all sites, likely originating from a thermogenic source or caused by bacterial CH4 oxidation. Water analyses showed an alkaline environment enriched in carbonates and DOC, in contrast to more acidic waters from glaciers in Greenland and Iceland.These first measurements demonstrate that the subglacial meltwaters from glaciers in the St. Elias Icefields are net sources of CH4 and net sinks of CO2 to the atmosphere. Our findings indicate that CH4 emissions from subglacial environments under alpine glaciers may be a more common phenomenon than previously thought, and a potential cause for remotely sensed CH4 concentrations anomalies over glaciated regions. However, more alpine glaciers and outlets from the Greenland Ice Sheet need to be studied to evaluate this link and provide the needed ground truthing for satellite sensors in high latitudes.&amp;#160;1. Christiansen &amp; J&amp;#248;rgensen (2018) DOI 10.1038/s41598-018-35054-72. Lamarche-Gagnon et al. (2019) DOI 10.1038/s41586-018-0800-03. Burns et al. (2018) DOI 10.1038/s41598-018-35253-24. Christiansen et al. (2021) DOI 10.1029/2021JG0063085. Sapper et al. (2022) DOI:10.5194/egusphere-egu22-9972

Groundwater storage is influenced by many geo-environmental factors. Most of these factors are prepared in the form of categorical data. The present study utilized raster satellite data instead of categorical data and a Random Forest machine learning model to identify groundwater potential zones at the downstream parts of Wadi Yalamlam, western Saudi Arabia. Eighteen groundwater-influenced variables are prepared in continuous raster format from ASTER GDEM, TRMM, and SPOT-5 satellite data. The Random Forest (RF) model is trained using (70%) of the target variable and validated using the rest (30%). The accuracy, sensitivity, and F1-score are all generated to evaluate the model performance. SPOT band 3, band 4, and the rainfall variables are the most important for groundwater potential mapping contributing 11%, 7%, and 8% during the prediction stage. The GDEM elevation variable contributed 6% and the slope variable scored 1%. The main conclusions of the study are: (1) The RF machine learning algorithm successfully identified three groundwater potential zones with an accuracy of 96%. (2) The high, moderate, and low potential groundwater zones covered 11.5%, 59.9%, and 28.6% of the study area respectively. (3) Majority of high and moderate zones lie within the pumping rate range between 10 and 20 m3/day. (4) The approach developed in this study can be applied to any other wadis having the same conditions to help authorities and decision-makers in planning and development projects.

Study DesignRetrospective Cohort Study.ObjectivesUsing natural language processing (NLP) in combination with machine learning on standard operative notes may allow for efficient billing, maximization of collections, and minimization of coder error. This study was conducted as a pilot study to determine if a machine learning algorithm can accurately identify billing Current Procedural Terminology (CPT) codes on patient operative notes.MethodsThis was a retrospective analysis of operative notes from patients who underwent elective spine surgery by a single senior surgeon from 9/2015 to 1/2020. Algorithm performance was measured by performing receiver operating characteristic (ROC) analysis, calculating the area under the ROC curve (AUC) and the area under the precision-recall curve (AUPRC). A deep learning NLP algorithm and a Random Forest algorithm were both trained and tested on operative notes to predict CPT codes. CPT codes generated by the billing department were compared to those generated by our model.ResultsThe random forest machine learning model had an AUC of .94 and an AUPRC of .85. The deep learning model had a final AUC of .72 and an AUPRC of .44. The random forest model had a weighted average, class-by-class accuracy of 87%. The LSTM deep learning model had a weighted average, class-by-class accuracy 0f 59%.ConclusionsCombining natural language processing with machine learning is a valid approach for automatic generation of CPT billing codes. The random forest machine learning model outperformed the LSTM deep learning model in this case. These models can be used by orthopedic or neurosurgery departments to allow for efficient billing.