Search

Terrestrial and ocean ecosystems are increasingly under threat from an array of anthropogenic pressures. And while threats mount, how people view and value nature is changing. In the United States (U.S.) in particular, there is a shift away from viewing nature as something to 'dominate,' as evidenced in the decline in hunting. However, it is unclear if or how opinions around environmental issues and conservation need might differ when comparing ocean versus terrestrial ecosystems, especially given the prevalence and continued importance of wild capture fishing in the U.S. We employed two national parallel surveys, one focused on oceans, the other land, receiving responses from nearly every state in the U.S. (N = 1,973). While we found only slight, but statistically significant more concern for ocean habitats and animals over terrestrial ecosystems, this did not translate to increased willingness to monetarily support more ocean conservation actions. Using Random Forest models, we also found the best predictor of conservation need was feeling most impacted by environmental issues personally (self and/or community), regardless of ecosystem type. In fact, land versus sea (survey) had the lowest rank in the models, underscoring the importance of general nature-based interactions. Instead, the number of outdoor recreational activities was a highly ranked variable explaining the level of reported impact to self/community, with people who participate in 2 or more activities scoring higher levels of impact, on average. Notably, people who hunt and fish, versus only do one or the other, reported higher levels of impact and participated in more activities overall, providing a more nuanced finding regarding the nature 'dominance hypothesis.' Voting, not political affiliation, was also important in explaining responses, and governmental mechanisms to fund conservation were favored over voluntary. Overall, our results add to the strong existing literature that access and connection to nature is key, but uniquely broad connection may "float all boats," especially when diversified. [ABSTRACT FROM AUTHOR]

Background: In randomized clinical trials, treatment effects may vary, and this possibility is referred to as heterogeneity of treatment effect (HTE). One way to quantify HTE is to partition participants into subgroups based on individual's risk of experiencing an outcome, then measuring treatment effect by subgroup. Given the limited availability of externally validated outcome risk prediction models, internal models (created using the same dataset in which heterogeneity of treatment analyses also will be performed) are commonly developed for subgroup identification. We aim to compare different methods for generating internally developed outcome risk prediction models for subject partitioning in HTE analysis. Methods: Three approaches were selected for generating subgroups for the 2,441 participants from the United States enrolled in the ASPirin in Reducing Events in the Elderly (ASPREE) randomized controlled trial. An extant proportional hazards-based outcomes predictive risk model developed on the overall ASPREE cohort of 19,114 participants was identified and was used to partition United States' participants by risk of experiencing a composite outcome of death, dementia, or persistent physical disability. Next, two supervised non-parametric machine learning outcome classifiers, decision trees and random forests, were used to develop multivariable risk prediction models and partition participants into subgroups with varied risks of experiencing the composite outcome. Then, we assessed how the partitioning from the proportional hazard model compared to those generated by the machine learning models in an HTE analysis of the 5-year absolute risk reduction (ARR) and hazard ratio for aspirin vs. placebo in each subgroup. Cochran's Q test was used to detect if ARR varied significantly by subgroup. Results: The proportional hazard model was used to generate 5 subgroups using the quintiles of the estimated risk scores; the decision tree model was used to generate 6 subgroups (6 automatically determined tree leaves); and the random forest model was used to generate 5 subgroups using the quintiles of the prediction probability as risk scores. Using the semi-parametric proportional hazards model, the ARR at 5 years was 15.1% (95% CI 4.0–26.3%) for participants with the highest 20% of predicted risk. Using the random forest model, the ARR at 5 years was 13.7% (95% CI 3.1–24.4%) for participants with the highest 20% of predicted risk. The highest outcome risk group in the decision tree model also exhibited a risk reduction, but the confidence interval was wider (5-year ARR = 17.0%, 95% CI= -5.4–39.4%). Cochran's Q test indicated ARR varied significantly only by subgroups created using the proportional hazards model. The hazard ratio for aspirin vs. placebo therapy did not significantly vary by subgroup in any of the models. The highest risk groups for the proportional hazards model and random forest model contained 230 participants each, while the highest risk group in the decision tree model contained 41 participants. Conclusions: The choice of technique for internally developed models for outcome risk subgroups influences HTE analyses. The rationale for the use of a particular subgroup determination model in HTE analyses needs to be explicitly defined based on desired levels of explainability (with features importance), uncertainty of prediction, chances of overfitting, and assumptions regarding the underlying data structure. Replication of these analyses using data from other mid-size clinical trials may help to establish guidance for selecting an outcomes risk prediction modelling technique for HTE analyses. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic prompted governments worldwide to implement a range of containment measures, including mass gathering restrictions, social distancing, and school closures. Despite these efforts, vaccines continue to be the safest and most effective means of combating such viruses. Yet, vaccine hesitancy persists, posing a significant public health concern, particularly with the emergence of new COVID-19 variants. To effectively address this issue, timely data is crucial for understanding the various factors contributing to vaccine hesitancy. While previous research has largely relied on traditional surveys for this information, recent sources of data, such as social media, have gained attention. However, the potential of social media data as a reliable proxy for information on population hesitancy, especially when compared with survey data, remains underexplored. This paper aims to bridge this gap. Our approach uses social, demographic, and economic data to predict vaccine hesitancy levels in the ten most populous US metropolitan areas. We employ machine learning algorithms to compare a set of baseline models that contain only these variables with models that incorporate survey data and social media data separately. Our results show that XGBoost algorithm consistently outperforms Random Forest and Linear Regression, with marginal differences between Random Forest and XGBoost. This was especially the case with models that incorporate survey or social media data, thus highlighting the promise of the latter data as a complementary information source. Results also reveal variations in influential variables across the five hesitancy classes, such as age, ethnicity, occupation, and political inclination. Further, the application of models to different MSAs yields mixed results, emphasizing the uniqueness of communities and the need for complementary data approaches. In summary, this study underscores social media data's potential for understanding vaccine hesitancy, emphasizes the importance of tailoring interventions to specific communities, and suggests the value of combining different data sources. [ABSTRACT FROM AUTHOR]

Skin cancer has a significant impact on the lives of many individuals annually and is recognized as the most prevalent type of cancer. In the United States, an estimated annual incidence of approximately 3.5 million people receiving a diagnosis of skin cancer underscores its widespread prevalence. Furthermore, the prognosis for individuals afflicted with advancing stages of skin cancer experiences a substantial decline in survival rates. This paper is dedicated to aiding healthcare experts in distinguishing between benign and malignant skin cancer cases by employing a range of machine learning and deep learning techniques and different feature extractors and feature selectors to enhance the evaluation metrics. In this paper, different transfer learning models are employed as feature extractors, and to enhance the evaluation metrics, a feature selection layer is designed, which includes diverse techniques such as Univariate, Mutual Information, ANOVA, PCA, XGB, Lasso, Random Forest, and Variance. Among transfer models, DenseNet-201 was selected as the primary feature extractor to identify features from data. Subsequently, the Lasso method was applied for feature selection, utilizing diverse machine learning approaches such as MLP, XGB, RF, and NB. To optimize accuracy and precision, ensemble methods were employed to identify and enhance the best-performing models. The study provides accuracy and sensitivity rates of 87.72% and 92.15%, respectively. [ABSTRACT FROM AUTHOR]

This research utilizes the Long-Term Pavement Performance database, focusing on devel-oping a predictive model for flexible pavement performance in the Southern United States. Analyzing 367 pavement sections, this study investigates crucial factors influencing asphaltic concrete (AC) pavement deterioration, such as structural and material components, air voids, compaction density, temperature at laydown, traffic load, precipitation, and freeze–thaw cycles. The objective of this study is to develop a predictive machine learning model for AC pavement wheel path cracking (WpCrAr) and the age at which cracking initiates (WpCrAr) as performance indicators. This study thoroughly investigated three ensemble machine learning models, including random forest, extremely randomized trees (ETR), and extreme gradient boosting (XGBoost). It was observed that XGBoost, optimized using Bayesian methods, emerged as the most effective among the evaluated models, demonstrating good predictive accuracy, with an R2 of 0.79 for WpCrAr and 0.92 for AgeCrack and mean absolute errors of 1.07 and 0.74, respectively. The most important features influencing crack initiation and progression were identified, including equivalent single axle load (ESAL), pavement age, number of layers, precipitation, and freeze–thaw cycles. This paper also showed the impact of pavement material combinations for base and subgrade layers on the delay of crack initiation. [ABSTRACT FROM AUTHOR]

It has become crucial to have an early prediction model that provides accurate assurance for users about the financial situation of consumers. Recent studies have focused on predicting corporate bankruptcies and credit defaults, not personal bankruptcies. Due to this situation, the present study fills the literature gap by comparing different machine learning algorithms to predict personal bankruptcy. The main objective of the study is to examine the usefulness of machine learning models such as SVM, random forest, AdaBoost, XGBoost, LightGBM, and CatBoost in forecasting personal bankruptcy. The study relies on two samples of households (learning and testing) from the Survey of Consumer Finances, which was conducted in the United States. Among the models estimated, LightGBM, CatBoost, and XGBoost showed the highest effectiveness. The most important variables used in the models are income, refusal to grant credit, delays in the repayment of liabilities, the revolving debt ratio, and the housing debt ratio. [ABSTRACT FROM AUTHOR]

The application of lidar data to assist with forest inventory is common around the world. However, the determination of tree species is still somewhat elusive. Lidar data collected using UAS (uncrewed aircraft systems) platforms offer high density point cloud data for areas from a few to several hundred hectares. General point cloud metrics computed using these data captured differences in the crown structure that proved useful for species classification. For our study, we manually adjusted plot and tree locations to align field trees and UAS lidar point data and computed common descriptive metrics using a small cylindrical sample of points designed to capture the top three meters and leader of each tree. These metrics were used to train a random forest classifier to differentiate between two conifer species, Douglas fir and western hemlock, common in the Pacific Northwest region of the United States. Our UAS lidar data had a single swath pulse density of 90 pulses/m2 and an aggregate pulse density of 556 pulses/m2. We trained classification models using both height and intensity metrics, height metrics alone, intensity metrics alone, and a small subset of five metrics, and achieved overall accuracies of 91.8%, 88.7%, 78.6%, and 91.5%, respectively. Overall, we showed that UAS lidar data captured morphological differences between the upper crowns of our two target species and produced a classification model that could be applied over large areas. [ABSTRACT FROM AUTHOR]

Accurately predicting the condition rating of a bridge deck is crucial for effective maintenance and repair planning. Despite significant research efforts to develop deterioration models, the efficacy of Random Forest, eXtreme Gradient Boosting (XGBoost), and Artificial Neural Network (ANN) in predicting the condition rating of the nation's bridge decks has remained unexplored. This study aims to assess the effectiveness of these algorithms for deck condition rating prediction at the national level. To achieve this, the study collected bridge data, which includes National Bridge Inventory (NBI), traffic, and climate regions gathered using Geospatial Information Science (GIS) and remote sensing techniques. Two datasets were collected: bridge data for a single year of 2020 and historical bridge data covering a five-year period from 2016 to 2020. Three models were trained using 319,404 and 1,246,261 bridge decks in the single-year bridge data and the five-year historical bridge data, respectively. Results show that the use of historical bridge data significantly improves the performance of the models compared to the single-year bridge data. Specifically, the Random Forest model achieved an overall accuracy of 83.4% and an average F1 score of 79.7%. In contrast, the XGBoost model achieved an overall accuracy of 79.4% and an average F1 score of 77.5%, while the ANN model obtained an overall accuracy of 79.7% and an average F1 score of 78.4%. Permutation-based variable importance reveals that NBI, traffic, and climate regions significantly contribute to model development. In conclusion, the Random Forest, XGBoost, and ANN models, trained using updated historical bridge data, provide useful tools for accurately predicting the condition rating of bridge decks in the United States, allowing infrastructure managers to efficiently schedule inspections and allocate maintenance resources. [ABSTRACT FROM AUTHOR]

Knowledge of the soil water content (SWC) is important for many aspects of agriculture and must be monitored to maximize crop yield, efficiently use limited supplies of irrigation water, and ensure optimal nutrient management with minimal environmental impact. Single-location sensors are often used to monitor SWC, but a limited number of point measurements is insufficient to measure SWC across most fields since SWC is typically very heterogeneous. To overcome this difficulty, several researchers have used data acquired from unmanned aerial vehicles (UAVs) to predict the SWC by using machine learning on a limited number of point measurements acquired across a field. While useful, these methods are limited by the relatively small number of SWC measurements that can be acquired with conventional measurement techniques. This study uses UAV-based data and thousands of SWC measurements acquired using geophysical methods at two different depths and before and after precipitation to predict the SWC using the random forest method across a vineyard in the central United States. Both multispectral data (five reflectance bands and eleven vegetation indices calculated from these bands) and thermal UAV-based data were acquired, and the importance of different reflectance data and vegetation indices in the prediction of SWC was analyzed. Results showed that when both thermal and multispectral data were used to estimate SWC, the thermal data contributed the most to prediction accuracy, although multispectral data were also important. Reflectance data contributed as much or more to prediction accuracy than most vegetation indices. SWC measurements that had a larger sample size and greater penetration depth (~30 cm sampling depth) were more accurately predicted than smaller and shallower SWC estimates (~18 cm sampling depth). The timing of SWC estimation was also important; higher accuracy predictions were achieved in wetter soils than in drier soils, and a light precipitation event also improved prediction accuracy. [ABSTRACT FROM AUTHOR]

Depth regulates many attributes of aquatic ecosystems, but relatively few lakes are measured, and existing datasets are biased toward large lakes. To address this, we used a large dataset of maximum (Zmax; n = 16,831) and mean (Zmean; n = 5,881) depth observations to create new depth models, focusing on lakes < 1,000 ha. We then used the models to characterize patterns in lake basin shape and volume. We included terrain metrics, water temperature and reflectance, polygon attributes, and other predictors in a random forest model. Our final models generally outperformed existing models (ZmaxR2=0.35; root mean square error [RMSE] = 8.0 m and ZmeanR2=0.36; RMSE = 3.0 m). Our models show that lake depth followed a Pareto distribution, with 2.8 orders of magnitude fewer lakes for an order of magnitude increase in depth. In addition, despite orders of magnitude variation in surface area, most size classes had a modal maximum depth of ~ 5 m. Concave (bowl‐shaped) lake basins represented 79% of all lakes, but lakes were more convex (funnel‐shaped) as surface area increased. Across the conterminous United States, 9.8% of all lake water was within the top meter of the water column, and 48% in the top 10 m. Excluding the Laurentian Great Lakes, we estimate the total volume in the conterminous United States is 1,057–1,294 km3, depending on whether Zmax or Zmean was modeled. Lake volume also exhibited substantial geographic variation, with high volumes in the upper Midwest, Northeast, and Florida and low volumes in the southwestern United States. [ABSTRACT FROM AUTHOR]

The serovars of Salmonella enterica display dramatic differences in pathogenesis and host preferences. We developed a process (patent pending) for grouping Salmonella isolates and serovars by their public health risk. We collated a curated set of 12,337 S. enterica isolate genomes from human, beef, and bovine sources in the US. After annotating a virulence gene catalog for each isolate, we used unsupervised random forest methods to estimate the proximity (similarity) between isolates based upon the genomic presentation of putative virulence traits We then grouped isolates (virulence clusters) using hierarchical clustering (Ward's method), used non-parametric bootstrapping to assess cluster stability, and externally validated the clusters against epidemiological virulence measures from FoodNet, the National Outbreak Reporting System (NORS), and US federal sampling of beef products. We identified five stable virulence clusters of S. enterica serovars. Cluster 1 (higher virulence) serovars yielded an annual incidence rate of domestically acquired sporadic cases roughly one and a half times higher than the other four clusters combined (Clusters 2–5, lower virulence). Compared to other clusters, cluster 1 also had a higher proportion of infections leading to hospitalization and was implicated in more foodborne and beef-associated outbreaks, despite being isolated at a similar frequency from beef products as other clusters. We also identified subpopulations within 11 serovars. Remarkably, we found S. Infantis and S. Typhimurium subpopulations that significantly differed in genome length and clinical case presentation. Further, we found that the presence of the pESI plasmid accounted for the genome length differences between the S. Infantis subpopulations. Our results show that S. enterica strains associated with highest incidence of human infections share a common virulence repertoire. This work could be updated regularly and used in combination with foodborne surveillance information to prioritize serovars of public health concern. [ABSTRACT FROM AUTHOR]

Ensuring food security in precision agriculture requires early prediction of soybean yield at various scales within the United States (U.S.), ranging from international to local levels. Accurate yield estimation is essential in preventing famine by providing insights into food availability during the growth season. Numerous deep learning (DL) algorithms have been developed to estimate soybean yield effectively using time-series remote sensing (RS) data to achieve this goal. However, the training data with short time spans can limit their ability to adapt to the dynamic and nuanced temporal changes in crop conditions. To address this challenge, we designed a 3D-ResNet-BiLSTM model to efficiently predict soybean yield at the county level across the U.S., even when using training data with shorter periods. We leveraged detailed Sentinel-2 imagery and Sentinel-1 SAR images to extract spectral bands, key vegetation indices (VIs), and VV and VH polarizations. Additionally, Daymet data was incorporated via Google Earth Engine (GEE) to enhance the model's input features. To process these inputs effectively, a dedicated 3D-ResNet architecture was designed to extract high-level features. These enriched features were then fed into a BiLSTM layer, enabling accurate prediction of soybean yield. To evaluate the efficacy of our model, its performance was compared with that of well-known models, including the Linear Regression (LR), Random Forest (RF), and 1D/2D/3D-ResNet models, as well as a 2D-CNN-LSTM model. The data from a short period (2019 to 2020) were used to train all models, while their accuracy was assessed using data from the year 2021. The experimental results showed that the proposed 3D-Resnet-BiLSTM model had a superior performance compared to the other models, achieving remarkable metrics (R2 = 0.791, RMSE = 5.56 Bu Ac−1, MAE = 4.35 Bu Ac−1, MAPE = 9%, and RRMSE = 10.49%). Furthermore, the 3D-ResNet-BiLSTM model showed a 7% higher R2 than the ResNet and RF models and an enhancement of 27% and 17% against the LR and 2D-CNN-LSTM models, respectively. The results highlighted our model's potential for accurate soybean yield predictions, supporting sustainable agriculture and food security. [ABSTRACT FROM AUTHOR]

The adoption of dicamba-tolerant (DT) soybean in the United States resulted in extensive off-target dicamba damage to non-DT vegetation across soybeanproducing states. Although soybeans are highly sensitive to dicamba, the intensity of observed symptoms and yield losses are affected by the genetic background of genotypes. Thus, the objective of this study was to detect novel marker-trait associations and expand on previously identified genomic regions related to soybean response to off-target dicamba. A total of 551 non-DT advanced breeding lines derived from 232 unique bi-parental populations were phenotyped for off-target dicamba across nine environments for three years. Breeding lines were genotyped using the Illumina Infinium BARCSoySNP6K BeadChip. Filtered SNPs were included as predictors in Random Forest (RF) and Support Vector Machine (SVM) models in a forward stepwise selection loop to identify the combination of SNPs yielding the highest classification accuracy. Both RF and SVM models yielded high classification accuracies (0.76 and 0.79, respectively) with minor extreme misclassifications (observed tolerant predicted as susceptible, and vice-versa). Eight genomic regions associated with off-target dicamba tolerance were identified on chromosomes 6 [Linkage Group (LG) C2], 8 (LG A2), 9 (LG K), 10 (LG O), and 19 (LG L). Although the genetic architecture of tolerance is complex, high classification accuracies were obtained when including the major effect SNP identified on chromosome 6 as the sole predictor. In addition, candidate genes with annotated functions associated with phases II (conjugation of hydroxylated herbicides to endogenous sugar molecules) and III (transportation of herbicide conjugates into the vacuole) of herbicide detoxification in plants were co-localized with significant markers within each genomic region. Genomic prediction models, as reported in this study, can greatly facilitate the identification of genotypes with superior tolerance to off-target dicamba. [ABSTRACT FROM AUTHOR]

Accurate estimation of evapotranspiration (ET) is a prerequisite for water management in arid regions. Field based methods estimate point-wise ET accurately, but the challenge is in estimating ET over a region with high accuracies. Machine learning based approaches were taken to estimate ET over a large spatial scale using the Bowen Ratio Energy Balance (BREB) technique. The BREB method depends on terrestrial energy balance equations to estimate ET. Thus, remote sensing-based parameters representing variables in the energy balance equation, and vegetation index representing plant health conditions were used in the model. The study was conducted in the arid areas of the southwestern United States, where dense patches of Salt cedar consume water from the primary water source. The preliminary model used enhanced vegetation index (EVI), global horizontal irradiance (GHI), surface temperature (TS), and relative humidity (RH) as parameters. The k-nearest neighbor method consistently generated poor accuracies. When all the parameters were used, accuracies of the other models varied within 90–94%. When one predictor parameter was dropped, the best model produced accuracies between 90 to 93%, which dropped to 87–92% when a second variable was dropped. Random forest and support vector machine with radial kernel consistently produced the best predictive accuracies. [ABSTRACT FROM AUTHOR]

Regression algorithms are regularly used for improving the accuracy of satellite precipitation products. In this context, satellite precipitation and topography data are the predictor variables, and gauged-measured precipitation data are the dependent variables. Alongside this, it is increasingly recognised in many fields that combinations of algorithms through ensemble learning can lead to substantial predictive performance improvements. Still, a sufficient number of ensemble learners for improving the accuracy of satellite precipitation products and their large-scale comparison are currently missing from the literature. In this study, we work towards filling in this specific gap by proposing 11 new ensemble learners in the field and by extensively comparing them. We apply the ensemble learners to monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and IMERG (Integrated Multi-satellitE Retrievals for GPM) gridded datasets that span over a 15-year period and over the entire contiguous United States (CONUS). We also use gauge-measured precipitation data from the Global Historical Climatology Network monthly database, version 2 (GHCNm). The ensemble learners combine the predictions of six machine learning regression algorithms (base learners), namely the multivariate adaptive regression splines (MARS), multivariate adaptive polynomial splines (poly-MARS), random forests (RF), gradient boosting machines (GBM), extreme gradient boosting (XGBoost) and Bayesian regularized neural networks (BRNN), and each of them is based on a different combiner. The combiners include the equal-weight combiner, the median combiner, two best learners and seven variants of a sophisticated stacking method. The latter stacks a regression algorithm on top of the base learners to combine their independent predictions. Its seven variants are defined by seven different regression algorithms, specifically the linear regression (LR) algorithm and the six algorithms also used as base learners. The results suggest that sophisticated stacking performs significantly better than the base learners, especially when applied using the LR algorithm. It also beats the simpler combination methods. [ABSTRACT FROM AUTHOR]

This article presents a methodology based on web mining techniques for estimating furniture prices using e-commerce data. Data on different public e-commerce sites in the United States were collected and analyzed using web mining methods. Deep learning and random forest algorithms were used to predict the prices of different types of furniture. Bookcase and dresser type furniture, which are widely used in price estimation, were selected. The inquiry identified a collection of eight distinctive attributes linked to furniture items, spanning measurements such as width, depth, and height, alongside features encompassing frame material, partition count, drawer count, color, and price. In preparation for constructing predictive models, a dataset comprising 300 instances was compiled for comprehensive analysis. Models developed based on web mining to predict furniture prices gave satisfactory results. During the testing phase, the random forest algorithm outperformed deep learning, achieving high goodness of fit values of 0.89 and 0.94 for bookcase and dresser furniture, respectively. The results indicate that price estimation for dresser furniture was more accurate than for bookcases in all models. The findings demonstrate that web mining techniques can be used effectively in competitive furniture pricing, with potential to save time and cost in pricing for furniture purchasing. [ABSTRACT FROM AUTHOR]

Improved understanding of characteristics related to weather forecast accuracy in the United States may help meteorologists develop more accurate predictions and may help Americans better interpret their daily weather forecasts. This article examines how spatio-temporal characteristics across the United States relate to forecast accuracy. We cluster the United States into six weather regions based on weather and geographic characteristics and analyze the patterns in forecast accuracy within each weather region. We then explore the relationship between climate characteristics and forecast accuracy within these weather regions. We conclude that patterns in forecast errors are closely related to the unique climates that characterize each region. [ABSTRACT FROM AUTHOR]

Introduction: Despite a recent global decrease in suicide rates, death by suicide has increased in the United States. It is therefore imperative to identify the risk factors associated with suicide attempts to combat this growing epidemic. In this study, we aim to identify potential risk factors of suicide attempt using geospatial features in an Artificial intelligence framework. Methods: We use iterative Random Forest, an explainable artificial intelligence method, to predict suicide attempts using data from the Million Veteran Program. This cohort incorporated 405,540 patients with 391,409 controls and 14,131 attempts. Our predictive model incorporates multiple climatic features at ZIPcode-level geospatial resolution. We additionally consider demographic features from the American Community Survey as well as the number of firearms and alcohol vendors per 10,000 people to assess the contributions of proximal environment, access to means, and restraint decrease to suicide attempts. In total 1,784 features were included in the predictive model. Results: Our results show that geographic areas with higher concentrations of married males living with spouses are predictive of lower rates of suicide attempts, whereas geographic areas where males are more likely to live alone and to rent housing are predictive of higher rates of suicide attempts. We also identified climatic features that were associated with suicide attempt risk by age group. Additionally, we observed that firearms and alcohol vendors were associated with increased risk for suicide attempts irrespective of the age group examined, but that their effects were small in comparison to the top features. Discussion: Taken together, our findings highlight the importance of social determinants and environmental factors in understanding suicide risk among veterans. [ABSTRACT FROM AUTHOR]

Background: Some studies have established associations between the prevalence of new-onset asthma and asthma exacerbation and socioeconomic and environmental determinants. However, research remains limited concerning the shape of these associations, the importance of the risk factors, and how these factors vary geographically. Objective: We aimed (1) to examine ecological associations between asthma prevalence and multiple socio-physical determinants in the United States; and (2) to assess geographic variations in their relative importance. Methods: Our study design is cross sectional based on county-level data for 2020 across the United States. We obtained self-reported asthma prevalence data of adults aged 18 years or older for each county. We applied conventional and geographically weighted random forest (GWRF) to investigate the associations between asthma prevalence and socioeconomic (e.g., poverty) and environmental determinants (e.g., air pollution and green space). To enhance the interpretability of the GWRF, we (1) assessed the shape of the associations through partial dependence plots, (2) ranked the determinants according to their global importance scores, and (3) mapped the local variable importance spatially. Results: Of the 3059 counties, the average asthma prevalence was 9.9 (standard deviation ± 0.99). The GWRF outperformed the conventional random forest. We found an indication, for example, that temperature was inversely associated with asthma prevalence, while poverty showed positive associations. The partial dependence plots showed that these associations had a non-linear shape. Ranking the socio-physical environmental factors concerning their global importance showed that smoking prevalence and depression prevalence were most relevant, while green space and limited language were of minor relevance. The local variable importance measures showed striking geographical differences. Conclusion: Our findings strengthen the evidence that socio-physical environments play a role in explaining asthma prevalence, but their relevance seems to vary geographically. The results are vital for implementing future asthma prevention programs that should be tailor-made for specific areas. Highlights: Asthma risk and protective factors were assessed with explainable geospatial machine learning. The geographically weighted random forest outperformed its conventional aspatial counterpart. We found striking non-linear environment-asthma associations. Smoking and depression were the most influential risk factors; green space was among the least. The relevance of socioeconomic and environmental factors varied geographically. [ABSTRACT FROM AUTHOR]

Constructing models that accurately predict Fusarium head blight (FHB) epidemics and are also amenable to large-scale deployment is a challenging task. In the United States, the emphasis has been on simple logistic regression (LR) models, which are easy to implement but may suffer from lower accuracies when compared with more complicated, harder-to-deploy (over large geographies) model frameworks such as functional or boosted regressions. This article examined the plausibility of random forests (RFs) for the binary prediction of FHB epidemics as a possible mediation between model simplicity and complexity without sacrificing accuracy. A minimalist set of predictors was also desirable rather than having the RF model use all 90 candidate variables as predictors. The input predictor set was filtered with the aid of three RF variable selection algorithms (Boruta, varSelRF, and VSURF), using resampling techniques to quantify the variability and stability of selected variable sets. Postselection filtering produced 58 competitive RF models with no more than 14 predictors each. One variable representing temperature stability in the 20 days before anthesis was the most frequently selected predictor. This was a departure from the prominence of relative humidity-based variables previously reported in LR models for FHB. The RF models had overall superior predictive performance over the LR models and may be suitable candidates for use by the Fusarium Head Blight Prediction Center. [ABSTRACT FROM AUTHOR]

Identifying determinants of smoking cessation is critical for developing optimal cessation treatments and interventions. Machine learning (ML) is becoming more prevalent for smoking cessation success prediction in treatment programs. However, only individuals with an intention to quit smoking cigarettes participate in such programs, which limits the generalizability of the results. This study applies data from the Population Assessment of Tobacco and Health (PATH), a United States longitudinal nationally representative survey, to select primary determinants of smoking cessation and to train ML classification models for predicting smoking cessation among the general population. An analytical sample of 9,281 adult current established smokers from the PATH survey wave 1 was used to develop classification models to predict smoking cessation by wave 2. Random forest and gradient boosting machines were applied for variable selection, and the SHapley Additive explanation method was used to show the effect direction of the top-ranked variables. The final model predicted wave 2 smoking cessation for current established smokers in wave 1 with an accuracy of 72% in the test dataset. The validation results showed that a similar model could predict wave 3 smoking cessation of wave 2 smokers with an accuracy of 70%. Our analysis indicated that more past 30 days e-cigarette use at the time of quitting, fewer past 30 days cigarette use before quitting, ages older than 18 at smoking initiation, fewer years of smoking, poly tobacco past 30-days use before quitting, and higher BMI resulted in higher chances of cigarette cessation for adult smokers in the US. [ABSTRACT FROM AUTHOR]

Background: Diphenhydramine (DPH) is an antihistamine medication that in overdose can result in anticholinergic symptoms and serious complications, including arrhythmia and coma. We aimed to compare the value of various machine learning (ML) models, including light gradient boosting machine (LGBM), logistic regression (LR), and random forest (RF), in the outcome prediction of DPH poisoning. Materials and Methods: We used the National Poison Data System database and included all of the human exposures of DPH from January 01, 2017 to December 31, 2017, and excluded those cases with missing information, duplicated cases, and those who reported co-ingestion. Data were split into training and test datasets, and three ML models were compared. We developed confusion matrices for each, and standard performance metrics were calculated. Results: Our study population included 53,761 patients with DPH exposure. The most common reasons for exposure, outcome, chronicity of exposure, and formulation were captured. Our results showed that the average precision-recall area under the curve (AUC) of 0.84. LGBM and RF had the highest performance (average AUC of 0.91), followed by LR (average AUC of 0.90). The specificity of the models was 87.0% in the testing groups. The precision of models was 75.0%. Recall (sensitivity) of models ranged between 73% and 75% with an F1 score of 75.0%. The overall accuracy of LGBM, LR, and RF models in the test dataset was 74.8%, 74.0%, and 75.1%, respectively. In total, just 1.1% of patients (mostly those with major outcomes) received physostigmine. Conclusion: Our study demonstrates the application of ML in the prediction of DPH poisoning. [ABSTRACT FROM AUTHOR]

BACKGROUND. West Nile virus (WNV) is the leading cause of mosquito-borne disease in humans in the United States. Since the introduction of the disease in 1999, incidence levels have stabilized in many regions, allowing for analysis of climate conditions that shape the spatial structure of disease incidence. OBJECTIVES. Our goal was to identify the seasonal climate variables that influence the spatial extent and magnitude of WNV incidence in humans. METHODS. We developed a predictive model of contemporary mean annual WNV incidence using U.S. county-level case reports from 2005 to 2019 and seasonally averaged climate variables. We used a random forest model that had an out-of-sample model performance of 혙² =0.61. RESULTS. Our model accurately captured the V-shaped area of higher WNV incidence that extends from states on the Canadian border south through the middle of the Great Plains. It also captured a region of moderate WNV incidence in the southern Mississippi Valley. The highest levels of WNV incidence were in regions with dry and cold winters and wet and mild summers. The random forest model classified counties with average winter precipitation levels <23.3 mm/month as having incidence levels over 11 times greater than those of counties that are wetter. Among the climate predictors, winter precipitation, fall precipitation, and winter temperature were the three most important predictive variables. DISCUSSION. We consider which aspects of the WNV transmission cycle climate conditions may benefit the most and argued that dry and cold winters are climate conditions optimal for the mosquito species key to amplifying WNV transmission. Our statistical model may be useful in projecting shifts in WNV risk in response to climate change. [ABSTRACT FROM AUTHOR]

Background: Trauma is a serious medical and economic burden worldwide, and patients with traumatic injuries have a poor survival rate after cardiac arrest. The authors developed a prediction model specific to prehospital trauma care and used machine learning techniques to increase its accuracy. Methods: This retrospective observational study analyzed data from patients with blunt trauma injuries due to traffic accidents and falls from January 1, 2018, to December 31, 2019. The data were collected from the National Emergency Medical Services Information System, which stores emergency medical service activity records nationwide in the United States. A random forest algorithm was used to develop a machine learning model. Results: The prediction model had an area under the curve of 0.95 and a negative predictive value of 0.99. The feature importance of the predictive model was highest for the AVPU (Alert, Verbal, Pain, Unresponsive) scale, followed by oxygen saturation (SpO2). Among patients who were progressing to cardiac arrest, the cutoff value was 89% for SpO2 in nonalert patients. Conclusions: The machine learning model was highly accurate in identifying patients who did not develop cardiac arrest. [ABSTRACT FROM AUTHOR]

Geospatial data and tools evolve as new technologies are developed and landscape change occurs over time. As a result, these data may become outdated and inadequate for supporting critical habitat-related work across the international boundary in the Sonoran and Mojave Deserts Bird Conservation Region (BCR 33) due to the area's complex vegetation communities and the discontinuity in data availability across the United States (US) and Mexico (MX) border. This research aimed to produce the first 30 m continuous land cover map of BCR 33 by prototyping new methods for desert vegetation classification using the Random Forest (RF) machine learning (ML) method. The developed RF classification model utilized multitemporal Landsat 8 Operational Land Imager spectral and vegetation index data from the period of 2013–2020, and phenology metrics tailored to capture the unique growing seasons of desert vegetation. Our RF model achieved an overall classification F-score of 0.80 and an overall accuracy of 91.68%. Our results portrayed the vegetation cover at a much finer resolution than existing land cover maps from the US and MX portions of the study area, allowing for the separation and identification of smaller habitat pockets, including riparian communities, which are critically important for desert wildlife and are often misclassified or nonexistent in current maps. This early prototyping effort serves as a proof of concept for the ML and data fusion methods that will be used to generate the final high-resolution land cover map of the entire BCR 33 region. [ABSTRACT FROM AUTHOR]

Simple Summary: While the incidence of late-onset colorectal cancer (LOCRC) has steadily decreased, the incidence of early-onset colorectal cancer (EOCRC) has continued to increase in the US. This study aims to uncover geographic disparities in EOCRC and understand how risk factors between EOCRC and LOCRC differ. The geographic analysis revealed regions with relatively low LOCRC rates and high EOCRC rates, identifying regions with a disproportionate burden of EOCRC. We then evaluated and compared community-level risk factors associated with incidence rates of EOCRC and LOCRC using the random forest machine learning method. The analysis identified a set of risk factors most predictive of EOCRC and LOCRC, such as diabetes prevalence and physical inactivity, but the importance of these risk factors varied between EOCRC and LORC. Collectively, these findings can help facilitate future studies that further uncover actionable interventions to reduce EOCRC and guide where targeted interventions to reduce EOCRC burden should be deployed. The proportion of patients diagnosed with colorectal cancer (CRC) at age < 50 (early-onset CRC, or EOCRC) has steadily increased over the past three decades relative to the proportion of patients diagnosed at age ≥ 50 (late-onset CRC, or LOCRC), despite the reduction in CRC incidence overall. An important gap in the literature is whether EOCRC shares the same community-level risk factors as LOCRC. Thus, we sought to (1) identify disparities in the incidence rates of EOCRC and LOCRC using geospatial analysis and (2) compare the importance of community-level risk factors (racial/ethnic, health status, behavioral, clinical care, physical environmental, and socioeconomic status risk factors) in the prediction of EOCRC and LOCRC incidence rates using a random forest machine learning approach. The incidence data came from the Surveillance, Epidemiology, and End Results program (years 2000–2019). The geospatial analysis revealed large geographic variations in EOCRC and LOCRC incidence rates. For example, some regions had relatively low LOCRC and high EOCRC rates (e.g., Georgia and eastern Texas) while others had relatively high LOCRC and low EOCRC rates (e.g., Iowa and New Jersey). The random forest analysis revealed that the importance of community-level risk factors most predictive of EOCRC versus LOCRC incidence rates differed meaningfully. For example, diabetes prevalence was the most important risk factor in predicting EOCRC incidence rate, but it was a less important risk factor of LOCRC incidence rate; physical inactivity was the most important risk factor in predicting LOCRC incidence rate, but it was the fourth most important predictor for EOCRC incidence rate. Thus, our community-level analysis demonstrates the geographic variation in EOCRC burden and the distinctive set of risk factors most predictive of EOCRC. [ABSTRACT FROM AUTHOR]

Gridded satellite precipitation datasets are useful in hydrological applications as they cover large regions with high density. However, they are not accurate in the sense that they do not agree with ground-based measurements. An established means for improving their accuracy is to correct them by adopting machine learning algorithms. This correction takes the form of a regression problem, in which the ground-based measurements have the role of the dependent variable and the satellite data are the predictor variables, together with topography factors (e.g., elevation). Most studies of this kind involve a limited number of machine learning algorithms and are conducted for a small region and for a limited time period. Thus, the results obtained through them are of local importance and do not provide more general guidance and best practices. To provide results that are generalizable and to contribute to the delivery of best practices, we here compare eight state-of-the-art machine learning algorithms in correcting satellite precipitation data for the entire contiguous United States and for a 15-year period. We use monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) gridded dataset, together with monthly earth-observed precipitation data from the Global Historical Climatology Network monthly database, version 2 (GHCNm). The results suggest that extreme gradient boosting (XGBoost) and random forests are the most accurate in terms of the squared error scoring function. The remaining algorithms can be ordered as follows, from the best to the worst: Bayesian regularized feed-forward neural networks, multivariate adaptive polynomial splines (poly-MARS), gradient boosting machines (gbm), multivariate adaptive regression splines (MARS), feed-forward neural networks and linear regression. [ABSTRACT FROM AUTHOR]

Merging satellite products and ground-based measurements is often required for obtaining precipitation datasets that simultaneously cover large regions with high density and are more accurate than pure satellite precipitation products. Machine and statistical learning regression algorithms are regularly utilized in this endeavor. At the same time, tree-based ensemble algorithms are adopted in various fields for solving regression problems with high accuracy and low computational costs. Still, information on which tree-based ensemble algorithm to select for correcting satellite precipitation products for the contiguous United States (US) at the daily time scale is missing from the literature. In this study, we worked towards filling this methodological gap by conducting an extensive comparison between three algorithms of the category of interest, specifically between random forests, gradient boosting machines (gbm) and extreme gradient boosting (XGBoost). We used daily data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and the IMERG (Integrated Multi-satellitE Retrievals for GPM) gridded datasets. We also used earth-observed precipitation data from the Global Historical Climatology Network daily (GHCNd) database. The experiments referred to the entire contiguous US and additionally included the application of the linear regression algorithm for benchmarking purposes. The results suggest that XGBoost is the best-performing tree-based ensemble algorithm among those compared. Indeed, the mean relative improvements that it provided with respect to linear regression (for the case that the latter algorithm was run with the same predictors as XGBoost) are equal to 52.66%, 56.26% and 64.55% (for three different predictor sets), while the respective values are 37.57%, 53.99% and 54.39% for random forests, and 34.72%, 47.99% and 62.61% for gbm. Lastly, the results suggest that IMERG is more useful than PERSIANN in the context investigated. [ABSTRACT FROM AUTHOR]

Anthropogenic activities that degrade natural ecosystems may also impact human health. However, research on the links between human and environmental health has most often been conducted at small scales (e.g. individual cities) and cannot easily be extrapolated to larger scales.We created a macrosystems ecology framework to identify associations between human and environmental health by combining human mortality and socioeconomic data for the conterminous United States with spatially aligned data on the physicochemical characteristics of river basins.Principal component analysis was first used to reduce a list of 596 environmental variables to a subset of 64 environmental covariates, representing six main environmental themes (climate, geology, hydrology, land use, river basin morphology and pollution). Independent, spatially aligned information was then obtained for 12 socioeconomic covariates.Random forest modelling was used to predict age‐adjusted mortality rate as a function of the environmental and socioeconomic covariates. An independent data subset (random 75:25 model building vs. testing split) was also used for model validation. The coefficient of determination between predicted and observed mortality rates was 0.76 for the validation data. Furthermore, model residuals (predicted − observed mortality) were centered near zero and normally distributed (1 SD = 62.26), suggesting high model accuracy and precision.Socioeconomic covariates were consistently the most influential predictors of mortality rate. Smoking, food insecurity, and lack of physical activity were particularly important. However, environmental covariates accounted for 5 of the 10 strongest predictors overall, with air temperature and precipitation being most influential among environmental variables.This proof‐of‐concept study demonstrates the utility of a modelling framework that combines environmental and human health data at macroscales. We suggest that further application of macrosystems ecology tools will improve the capacity to anticipate human health responses to ongoing environmental change. Read the free Plain Language Summary for this article on the Journal blog. Read the free Plain Language Summary for this article on the Journal blog. [ABSTRACT FROM AUTHOR]

The MODIS 8-day composite evapotranspiration (ET) product (MOD16A2) is widely used to study large-scale hydrological cycle and energy budgets. However, the MOD16A2 spatial resolution (500 m) is too coarse for local and regional water resource management in agricultural applications. In this study, we propose a Deep Neural Network (DNN)-based MOD16A2 downscaling approach to generate 30 m ET using Landsat 8 surface reflectance and temperature and AgERA5 meteorological variables. The model was trained at a 500 m resolution using the MOD16A2 ET as reference and applied to the Landsat 8 30 m resolution. The approach was tested on 15 Landsat 8 images over three agricultural study sites in the United States and compared with the classical random forest regression model that has been often used for ET downscaling. All evaluation sample sets applied to the DNN regression model had higher R2 and lower root-mean-square deviations (RMSD) and relative RMSD (rRMSD) (the average values: 0.67, 2.63 mm/8d and 14.25%, respectively) than the random forest model (0.64, 2.76 mm/8d and 14.92%, respectively). Spatial improvement was visually evident both in the DNN and the random forest downscaled 30 m ET maps compared with the 500 m MOD16A2, while the DNN-downscaled ET appeared more consistent with land surface cover variations. Comparison with the in situ ET measurements (AmeriFlux) showed that the DNN-downscaled ET had better accuracy, with R2 of 0.73, RMSD of 5.99 mm/8d and rRMSD of 48.65%, than the MOD16A2 ET (0.65, 7.18 and 50.42%, respectively). [ABSTRACT FROM AUTHOR]

Simple Summary: Colorectal cancer is one of the most significant causes of cancer mortality worldwide. Patients stratification is central to improve clinical practice and the Consensus Molecular Subtypes (CMS) have been validated as a useful tool to predict both prognosis and treatment response. This is the first study describing that microRNA profiling can define colorectal cancer CMS subtypes as well as mRNA profiling. MicroRNAs small size facilitates its analysis in serum facilitating a real-time analysis of the disease course. Three microRNA subtypes are identified: miR-LS is associated with the low-stroma/CMS2-subtype; miR-MI with the mucinous-MSI/CMS1-subtype and miR-HS with the high-stroma/CMS4-subtype. MicroRNA novel subtypes and association to the CMS classification were externally validated using TGCA data. Analyzing both mRNAs and miRs in the same population enabled identification of miR target genes and altered biological pathways. A miR-mRNA interaction screening and regulatory network selected major miR targets and was functionally validated for the miR30b/SCL6A4 pair. Colorectal cancer consensus molecular subtypes (CMSs) are widely accepted and constitutes the basis for patient stratification to improve clinical practice. We aimed to find whether miRNAs could reproduce molecular subtypes, and to identify miRNA targets associated to the High-stroma/CMS4 subtype. The expression of 939 miRNAs was analyzed in tumors classified in CMS. TALASSO was used to find gene-miRNA interactions. A miR-mRNA regulatory network was constructed using Cytoscape. Candidate gene-miR interactions were validated in 293T cells. Hierarchical-Clustering identified three miRNA tumor subtypes (miR-LS; miR-MI; and miR-HS) which were significantly associated (p < 0.001) to the reported mRNA subtypes. miR-LS correlated with the low-stroma/CMS2; miR-MI with the mucinous-MSI/CMS1 and miR-HS with high-stroma/CMS4. MicroRNA tumor subtypes and association to CMSs were validated with TCGA datasets. TALASSO identified 1462 interactions (p < 0.05) out of 21,615 found between 176 miRs and 788 genes. Based on the regulatory network, 88 miR-mRNA interactions were selected as candidates. This network was functionally validated for the pair miR-30b/SLC6A6. We found that miR-30b overexpression silenced 3′-UTR-SLC6A6-driven luciferase expression in 293T-cells; mutation of the target sequence in the 3′-UTR-SLC6A6 prevented the miR-30b inhibitory effect. In conclusion CRC subtype classification using a miR-signature might facilitate a real-time analysis of the disease course and treatment response. [ABSTRACT FROM AUTHOR]

Understanding the phytochemical landscapes of essential and nonessential chemical elements to plants provides an opportunity to better link biogeochemical cycles to trophic ecology. We investigated the formation and regulation of the cationic phytochemical landscapes of four key elements for biota: Ca, Mg, K, and Na. We collected aboveground tissues of plants in Atriplex, Helianthus, and Opuntia and adjacent soils from 51, 131, and 83 sites, respectively, across the southern United States. We determined the spatial variability of these cations in plants and soils. Also, we quantified the homeostasis coefficient for each cation and genus combination, by using mixed‐effect models, with spatially correlated random effects. Additionally, using random forest models, we modeled the influence of bioclimatic, soil, and spatial variables on plant cationic concentrations. Sodium variability and spatial autocorrelation were considerably greater than for Ca, Mg, or K. Calcium, Mg, and K exhibited strongly homeostatic patterns, in striking contrast to non‐homeostatic Na. Even so, climatic and soil variables explained a large proportion of plants' cationic concentrations. Essential elements (Ca, Mg, and K) appeared to be homeostatically regulated, which contrasted sharply with Na, a nonessential element for most plants. In addition, we provide evidence for the No‐Escape‐from‐Sodium hypothesis in real‐world ecosystems, indicating that plant Na concentrations tend to increase as substrate Na levels increase. [ABSTRACT FROM AUTHOR]

The United States' blood supply chain is experiencing market decline due to recent innovations in surgical practice, transfusion management, and hospital policy. These innovations strain US blood centers, resulting in cuts to surge capacities, consolidation, and reduced funding for research and outreach programs. In this study, we use data from a regional blood center to explore the application of contemporary machine learning algorithms for modeling donor retention. Such predictive models of donor retention can be used to design more cost effective donor outreach programs. Using data from a large US blood center paired with random forest classifiers, we are able to build a model of donor retention with a Mathews correlation of coefficient of 0.851. [ABSTRACT FROM AUTHOR]

Early childhood is an important development period for establishing healthy physical activity (PA) habits. The objective of this study was to evaluate PA levels in a representative sample of U.S. preschool-aged children. The study sample included 301 participants (149 girls, 3–5 years of age) in the 2012 U.S. National Health and Examination Survey National Youth Fitness Survey. Participants were asked to wear an ActiGraph accelerometer on their wrist for 7 days. A machine learning random forest classification algorithm was applied to accelerometer data to estimate daily time spent in moderate- and vigorous-intensity PA (MVPA; the sum of minutes spent in running, walking, and other moderate- and vigorous-intensity PA) and total PA (the sum of MVPA and light-intensity PA). We estimated that U.S. preschool-aged children engaged in 28 min/day of MVPA and 361 min/day of total PA, on average. MVPA and total PA levels were not significantly different between males and females. This study revealed that U.S. preschool-aged children engage in lower levels of MVPA and higher levels of total PA than the minimum recommended by the World Health Organization. [ABSTRACT FROM AUTHOR]

Background: The proportion of patients with post-traumatic stress disorder (PTSD) that remain undiagnosed may be substantial. Without an accurate diagnosis, these patients may lack PTSD-targeted treatments and experience adverse health outcomes. This study used a machine learning approach to identify and describe civilian patients likely to have undiagnosed PTSD in the US commercial population. Methods: The IBM® MarketScan® Commercial Subset (10/01/2015–12/31/2018) was used. A random forest machine learning model was developed and trained to differentiate between patients with and without PTSD using non–trauma-based features. The model was applied to patients for whom PTSD status could not be confirmed to identify individuals likely and unlikely to have undiagnosed PTSD. Patient characteristics, symptoms and complications potentially related to PTSD, treatments received, healthcare costs, and healthcare resource utilization were described separately for patients with PTSD (Actual Positive PTSD cohort), patients likely to have PTSD (Likely PTSD cohort), and patients without PTSD (Without PTSD cohort). Results: A total of 44,342 patients were classified in the Actual Positive PTSD cohort, 5683 in the Likely PTSD cohort, and 2,074,471 in the Without PTSD cohort. While several symptoms/comorbidities were similar between the Actual Positive and Likely PTSD cohorts, others, including depression and anxiety disorders, suicidal thoughts/actions, and substance use, were more common in the Likely PTSD cohort, suggesting that certain symptoms may be exacerbated among those without a formal diagnosis. Mean per-patient-per-6-month healthcare costs were similar between the Actual Positive and Likely PTSD cohorts ($11,156 and $11,723) and were higher than those of the Without PTSD cohort ($3616); however, cost drivers differed between cohorts, with the Likely PTSD cohort experiencing more inpatient admissions and less outpatient visits than the Actual Positive PTSD cohort. Conclusions: These findings suggest that the lack of a PTSD diagnosis and targeted management of PTSD may result in a greater burden among undiagnosed patients and highlights the need for increased awareness of PTSD in clinical practice and among the civilian population. [ABSTRACT FROM AUTHOR]

Human activities alter ecosystems everywhere, causing rapid biodiversity loss and biotic homogenization. These losses necessitate coordinated conservation actions guided by biodiversity and species distribution spatial data that cover large areas yet have fine‐enough resolution to be management‐relevant (i.e., ≤5 km). However, most biodiversity products are too coarse for management or are only available for small areas. Furthermore, many maps generated for biodiversity assessment and conservation do not explicitly quantify the inherent tradeoff between resolution and accuracy when predicting biodiversity patterns. Our goals were to generate predictive models of overall breeding bird species richness and species richness of different guilds based on nine functional or life‐history‐based traits across the conterminous United States at three resolutions (0.5, 2.5, and 5 km) and quantify the tradeoff between resolution and accuracy and, hence, relevance for management of the resulting biodiversity maps. We summarized 18 years of North American Breeding Bird Survey data (1992–2019) and modeled species richness using random forests, including 66 predictor variables (describing climate, vegetation, geomorphology, and anthropogenic conditions), 20 of which we newly derived. Among the three spatial resolutions, the percentage variance explained ranged from 27% to 60% (median = 54%; mean = 57%) for overall species richness and 12% to 87% (median = 61%; mean = 58%) for our different guilds. Overall species richness and guild‐specific species richness were best explained at 5‐km resolution using ~24 predictor variables based on percentage variance explained, symmetric mean absolute percentage error, and root mean square error values. However, our 2.5‐km‐resolution maps were almost as accurate and provided more spatially detailed information, which is why we recommend them for most management applications. Our results represent the first consistent, occurrence‐based, and nationwide maps of breeding bird richness with a thorough accuracy assessment that are also spatially detailed enough to inform local management decisions. More broadly, our findings highlight the importance of explicitly considering tradeoffs between resolution and accuracy to create management‐relevant biodiversity products for large areas. [ABSTRACT FROM AUTHOR]

Predicting wildfire spread is critical for land management and disaster preparedness. To this end, we present “Next Day Wildfire Spread,” a curated, large-scale, multivariate dataset of historical wildfires aggregating nearly a decade of remote-sensing data across the United States. In contrast to existing fire datasets based on Earth observation satellites, our dataset combines 2-D fire data with multiple explanatory variables (e.g., topography, vegetation, weather, drought index, and population density) aligned over 2-D regions, providing a feature-rich dataset for machine learning. To demonstrate the usefulness of this dataset, we implement a neural network that takes advantage of the spatial information of these data to predict wildfire spread. We compare the performance of the neural network with other machine learning models: logistic regression and random forest. This dataset can be used as a benchmark for developing wildfire propagation models based on remote-sensing data for a lead time of one day. [ABSTRACT FROM AUTHOR]

The 3D body scan technology has recently innovated the way of measuring human bodies and generated a large volume of body measurements. However, one inherent issue that plagues the use of the resultant database is the missing data usually caused by using automatic data extractions from the 3D body scans. Tedious extra efforts have to be made to manually fill the missing data for various applications. To tackle this problem, this paper proposes a machine learning (ML)-based approach for 3D body measurement estimation while considering the measurement (feature) importance. The proposed approach selects the most critical features to reduce the algorithm input and to improve the ML method performance. In addition, a Bayesian search is further used in fine-tuning the hyperparameters to minimize the mean square error. Two distinct ML methods, i.e., Random Forest and XGBoost, are used and tested on a real-world dataset that contains 3D body scans of 212 participants in the Kansas-Missouri area of the United States. The results show the effectiveness of the proposed methods with roughly 3% of Mean Absolute Percentage Errors in estimating the missing data. The two ML methods with the proposed hybrid feature selection and the Baysian search are comprehensively compared. The comparative results suggest that the Random Forest method performs better than the XGBoost counterpart in filling missing 3D body measurements. [ABSTRACT FROM AUTHOR]

Terrorism is a major problem worldwide, causing thousands of fatalities and billions of dollars in damage every year. To address this threat, we propose a novel feature representation method and evaluate machine learning models that learn from localized news data in order to predict whether a terrorist attack will occur on a given calendar date and in a given state. The best model (a Random Forest aided by a novel variable-length moving average method) achieved area under the receiver operating characteristic (AUROC) of ≥ 0.667 (statistically significant w.r.t. random guessing with p ≤.0001) on four of the five states that were impacted most by terrorism between 2015 and 2018. These results demonstrate that treating terrorism as a set of independent events, rather than as a continuous process, is a fruitful approach—especially when historical events are sparse and dissimilar—and that large-scale news data contains information that is useful for terrorism prediction. Our analysis also suggests that predictive models should be localized (i.e., state models should be independently designed, trained, and evaluated) and that the characteristics of individual attacks (e.g., responsible group or weapon type) were not correlated with prediction success. These contributions provide a foundation for the use of machine learning in efforts against terrorism in the United States and beyond. [ABSTRACT FROM AUTHOR]

Background: Super-utilizers represent approximately 5% of the population in the United States (U.S.) and yet they are responsible for over 50% of healthcare expenditures. Using characteristics of hospital service areas (HSAs) to predict utilization of resource intensive healthcare (RIHC) may offer a novel and actionable tool for identifying super-utilizer segments in the population. Consumer expenditures may offer additional value in predicting RIHC beyond typical population characteristics alone.Methods: Cross-sectional data from 2017 was extracted from 5 unique sources. The outcome was RIHC and included emergency room (ER) visits, inpatient days, and hospital expenditures, all expressed as log per capita. Candidate predictors from 4 broad groups were used, including demographics, adults and child health characteristics, community characteristics, and consumer expenditures. Candidate predictors were expressed as per capita or per capita percent and were aggregated from zip-codes to HSAs using weighed means. Machine learning approaches (Random Forrest, LASSO) selected important features from nearly 1,000 available candidate predictors and used them to generate 4 distinct models, including non-regularized and LASSO regression, random forest, and gradient boosting. Candidate predictors from the best performing models, for each outcome, were used as independent variables in multiple linear regression models. Relative contribution of variables from each candidate predictor group to regression model fit were calculated.Results: The median ER visits per capita was 0.482 [IQR:0.351-0.646], the median inpatient days per capita was 0.395 [IQR:0.214-0.806], and the median hospital expenditures per capita was $2,302 [1$,544.70-$3,469.80]. Using 1,106 variables, the test-set coefficient of determination (R2) from the best performing models ranged between 0.184-0.782. The adjusted R2 values from multiple linear regression models ranged from 0.311-0.8293. Relative contribution of consumer expenditures to model fit ranged from 23.4-33.6%.Discussion: Machine learning models predicted RIHC among HSAs using diverse population data, including novel consumer expenditures and provides an innovative tool to predict population-based healthcare utilization and expenditures. Geographic variation in utilization and spending were identified. [ABSTRACT FROM AUTHOR]

Historically, grasslands with limited tree presence were embedded in a matrix of predominantly open oak and pine forests in the eastern United States. These open ecosystems mostly have been lost to other land uses, particularly agriculture, and also to closed forests under fire exclusion because frequent surface fire prevents tree encroachment. We located the potential extent of eastern fire‐maintained grasslands by applying the random forests and C5.0 classifiers to determine the relationship between mapped areas of historical grasslands and topography and wind speed, which are proxies for surface fire frequency. A generalized ruleset was that fire‐maintained grasslands occurred at roughness values of less than 95, or flatter sites, and wind speeds ≥3.4 m s−1, which created large fire compartments. Potential grasslands covered 27 million ha, or 14% of the 200 million ha of the eastern United States, although these fire‐maintained locations also may have been savannas or open woodlands historically. Currently, potential grassland locations are 40% crops, 25% pasture, 18% forests, and 13% developed land, with about 1.5% each of herbaceous upland vegetation, herbaceous wetlands, and shrublands. According to historical accounts, fire‐maintained grasslands generally transitioned to dense young tree growth within a 20‐year interval after fire exclusion; in Kentucky, the transition transpired during the periods 1790–1810 or 1810–1830, but dates vary with Euro‐American settlement time. Finding the forgotten grasslands of the eastern United States, with this mechanistic approach to estimate fire disturbance, is an important first step for recovering and managing eastern grassland biodiversity. [ABSTRACT FROM AUTHOR]

Abrupt environmental changes can affect the population structures of living species and cause habitat loss and fragmentations in the ecosystem. During August–October 2020, remarkably high mortality events of avian species were reported across the western and central United States, likely resulting from winter storms and wildfires. However, the differences of mortality events among various species responding to the abrupt environmental changes remain poorly understood. In this study, we focused on three species, Wilson's Warbler, Barn Owl, and Common Murre, with the highest mortality events that had been recorded by citizen scientists. We leveraged the citizen science data and multiple remotely sensed earth observations and employed the ensemble random forest models to disentangle the species responses to winter storm and wildfire. We found that the mortality events of Wilson's Warbler were primarily impacted by early winter storms, with more deaths identified in areas with a higher average daily snow cover. The Barn Owl's mortalities were more identified in places with severe wildfire-induced air pollution. Both winter storms and wildfire had relatively mild effects on the mortality of Common Murre, which might be more related to anomalously warm water. Our findings highlight the species-specific responses to environmental changes, which can provide significant insights into the resilience of ecosystems to environmental change and avian conservations. Additionally, the study emphasized the efficiency and effectiveness of monitoring large-scale abrupt environmental changes and conservation using remotely sensed and citizen science data. [ABSTRACT FROM AUTHOR]

The United States NRCS has a soil database that has data collected from across the country over the last several decades. This also includes soil spectral scans. This data is available, but it may not be used to its full potential. For this study, pedon, horizon and spectral data was extracted from the database for samples collected from 2011 to 2015. Only sites that had been fully described and horizons that had been analyzed for the full suite of desired properties were used. This resulted in over 14,000 samples that were used for modeling and eight soil properties: soil organic carbon (SOC); total nitrogen (TN); total sulfur (TS); clay; sand; exchangeable calcium (Caex); cation exchange capacity (CEC); and pH. Four chemometric methods were employed for soil property prediction: partial least squares (PLSR); Random Forest (RF); Cubist; and multivariable adaptive regression splines (MARS). The dataset was sufficiently large that only random subsetting was used to create calibration (70%) and validation (30%) sets. SOC, TN, and TS had the strongest prediction results, with an R2 value of over 0.9. Caex, CEC and pH were predicted moderately well. Clay and sand models had slightly lower performance. Of the four methods, Cubist produced the strongest models, while PLSR produced the weakest. This may be due to the complex relationships between soil properties and spectra that PLSR could not capture. The only drawback of Cubist is the difficult interpretability of variable importance. Future research should include the use of environmental variables to improve prediction results. Future work may also avoid the use of PLSR when dealing with large datasets that cover large areas and have high degrees of variability. [ABSTRACT FROM AUTHOR]

The present study compared the performance of logistic regression models with that of machine learning classification models (classification trees and random forests) in the context of predicting training attrition from the Delayed Enlistment Program in the United States Marine Corps (USMC) with scores from the Tailored Adaptive Personality Assessment System (TAPAS). Performance was assessed according to the type of misclassification error and across a variety of different reasons for attrition. The base rate of attrition was low, which impeded the training process, but the machine learning models outperformed logistic regression in predicting voluntary attrition in a stratified 50% attrition sample. [ABSTRACT FROM AUTHOR]

Predictors of opioid overdose death in neighborhoods are important to identify, both to understand characteristics of high-risk areas and to prioritize limited prevention and intervention resources. Machine learning methods could serve as a valuable tool for identifying neighborhood-level predictors. We examined statewide data on opioid overdose death from Rhode Island (log-transformed rates for 2016–2019) and 203 covariates from the American Community Survey for 742 US Census block groups. The analysis included a least absolute shrinkage and selection operator (LASSO) algorithm followed by variable importance rankings from a random forest algorithm. We employed double cross-validation, with 10 folds in the inner loop to train the model and 4 outer folds to assess predictive performance. The ranked variables included a range of dimensions of socioeconomic status, including education, income and wealth, residential stability, race/ethnicity, social isolation, and occupational status. The R 2 value of the model on testing data was 0.17. While many predictors of overdose death were in established domains (education, income, occupation), we also identified novel domains (residential stability, racial/ethnic distribution, and social isolation). Predictive modeling with machine learning can identify new neighborhood-level predictors of overdose in the continually evolving opioid epidemic and anticipate the neighborhoods at high risk of overdose mortality. [ABSTRACT FROM AUTHOR]

Objective: The primary objectives of this study were to examine referral sources and demographic, clinical, and socioenvironmental characteristics of Hispanics referred to and enrolled in a program of coordinated specialty care (Early CSC program) for first-episode psychosis, to compare them with characteristics of other referred and enrolled racial-ethnic groups, and to identify factors associated with enrollment in the program.Methods: A retrospective review was conducted for all individuals referred to and enrolled in the Early CSC program over a 2-year period. Extracted data included referral sources and demographic and clinical characteristics. Zip code-level data from publicly available sources were cross-referenced with individual records. Nonparametric tests and appropriate secondary analysis were used to determine significant differences across racial-ethnic groups referred to (N=180) or enrolled in (N=75) the Early CSC program. A random forest model was used to determine which factors or interacting factors were associated with enrollment among the eligible referrals (N=114).Results: Hispanic individuals were more likely to be referred from inpatient or outpatient mental health providers and not from other community sources. Among eligible Hispanic referrals, those who lived in areas with a lower percentage of Spanish speaking in the home were more likely to enroll in services, compared with those who lived in areas with a higher percentage of Spanish speaking.Conclusions: Continued exploration of factors associated with referral and enrollment in CSC programs for the growing Hispanic ethnic group in the United States can help determine best steps for developing these programs. [ABSTRACT FROM AUTHOR]

[The Geostationary Lightning Mapper (GLM) detects total lightning continuously from space, and does not distinguish intra‐cloud (IC) from cloud‐to‐ground (CG) lightning. This research focuses on differentiating CG and IC lightning detected by GLM using a random forests (RF) model. GLM flash and group characteristics are implemented into the RF model and used to predict flash type. The most important flash characteristic for distinguishing flash type is the maximum group area. Other features with high feature importance include time‐of‐day, elongation, propagation, footprint, slope, maximum distance between groups and events, and mean energy. Skill scores showcase the model's ability to distinguish flash type with moderate skill, with 81% probability of detection, 71% percent correct, 36% false alarm rate, 36% false alarm ratio, and 56% critical success index. These scores improve further when study area is limited to CONUS and the Southeastern United States. These results can be used to aid in future climatological analysis of flash type.] Plain Language Summary: The Geostationary Lightning Mapper (GLM) is a satellite‐based lightning sensor that allows for continuous detection of light emitted from the cloud tops due to lightning. GLM does not distinguish if the lightning is connecting to ground (CG) or remaining in the cloud (IC). In order to distinguish CG and IC flashes, this research uses a machine learning method called random forests (RF). The RF model attempts to classify lightning flashes based on their size, duration, and intensity. The most important flash characteristics for distinguishing flash type are the features related to the areal size of the lightning and the time of day the lightning occurs. Overall, moderate success is shown when attempting to divide total lightning into CG and IC over the 2018 period. This information can be used by researchers to improve the use of GLM in the study of different storm types as well as aiding in wildfire forecasting. Key Points: Although there is substantial overlap in the distributions for cloud‐to‐ground (CG) and intra‐cloud (IC) flash characteristics, the mean values differMost important features for flash distinction are maximum group area, time‐of‐day, elongation, propagation, footprint, slope, and max group/event distanceResults show skill in the random forests model's ability to distinguish CG and IC flashes in the Geostationary Lightning Mapper data [ABSTRACT FROM AUTHOR]

Foliar fungicide usage in soybeans in the north-central United States increased steadily over the past two decades. An agronomically-interpretable machine learning framework was used to understand the importance of foliar fungicides relative to other factors associated with realized soybean yields, as reported by growers surveyed from 2014 to 2016. A database of 2738 spatially referenced fields (of which 30% had been sprayed with foliar fungicides) was fit to a random forest model explaining soybean yield. Latitude (a proxy for unmeasured agronomic factors) and sowing date were the two most important factors associated with yield. Foliar fungicides ranked 7th out of 20 factors in terms of relative importance. Pairwise interactions between latitude, sowing date and foliar fungicide use indicated more yield benefit to using foliar fungicides in late-planted fields and in lower latitudes. There was a greater yield response to foliar fungicides in higher-yield environments, but less than a 100 kg/ha yield penalty for not using foliar fungicides in such environments. Except in a few production environments, yield gains due to foliar fungicides sufficiently offset the associated costs of the intervention when soybean prices are near-to-above average but do not negate the importance of disease scouting and fungicide resistance management. [ABSTRACT FROM AUTHOR]

Background: Cardiac dysrhythmias (CD) affect millions of Americans in the United States (US), and are associated with considerable morbidity and mortality. New strategies to combat this growing problem are urgently needed. Objectives: Predicting CD using electronic health record (EHR) data would allow for earlier diagnosis and treatment of the condition, thus improving overall cardiovascular outcomes. The Guideline Advantage (TGA) is an American Heart Association ambulatory quality clinical data registry of EHR data representing 70 clinics distributed throughout the US, and has been used to monitor outpatient prevention and disease management outcome measures across populations and for longitudinal research on the impact of preventative care. Methods: For this study, we represented all time-series cardiovascular health (CVH) measures and the corresponding data collection time points for each patient by numerical embedding vectors. We then employed a deep learning technique–long-short term memory (LSTM) model–to predict CD from the vector of time-series CVH measures by 5-fold cross validation and compared the performance of this model to the results of deep neural networks, logistic regression, random forest, and Naïve Bayes models. Results: We demonstrated that the LSTM model outperformed other traditional machine learning models and achieved the best prediction performance as measured by the average area under the receiver operator curve (AUROC): 0.76 for LSTM, 0.71 for deep neural networks, 0.66 for logistic regression, 0.67 for random forest, and 0.59 for Naïve Bayes. The most influential feature from the LSTM model were blood pressure. Conclusions: These findings may be used to prevent CD in the outpatient setting by encouraging appropriate surveillance and management of CVH. [ABSTRACT FROM AUTHOR]

Excessive rainfall is difficult to forecast, and there is a need for tools to aid Weather Prediction Center (WPC) forecasters when generating Excessive Rainfall Outlooks (EROs), which are issued for the contiguous United States at lead times of 1–3 days. To address this need, a probabilistic forecast system for excessive rainfall, known as the Colorado State University Machine Learning Probabilities (CSU-MLP) system, was developed based on ensemble reforecasts, precipitation observations, and machine-learning algorithms, specifically random forests. The CSU-MLP forecasts were designed to emulate the EROs, with the goal being a tool that forecasters can use as a "first guess" in the ERO forecast process. Resulting from close collaboration between CSU and WPC and evaluation at the Flash Flood and Intense Rainfall Experiment, iterative improvements were made to the forecast system and it was transitioned into operational use at WPC. Quantitative evaluation shows that the CSU-MLP forecasts are skillful and reliable, and they are now being used as a part of the WPC forecast process. This project represents an example of a successful research-to-operations transition, and highlights the potential for machine learning and other postprocessing techniques to improve operational predictions. [ABSTRACT FROM AUTHOR]