Project description:There is increasing interest to control or eradicate the major neglected tropical diseases. Accurate modelling of the geographic distributions of parasitic infections will be crucial to this endeavour. We used 664 community level infection prevalence data collated from the published literature in conjunction with eight environmental variables, altitude and population density, and a multivariate Bayesian generalized linear spatial model that allows explicit accounting for spatial autocorrelation and incorporation of uncertainty in input data and model parameters, to construct the first spatially-explicit map describing LF prevalence distribution in Africa. We also ran the best-fit model against predictions made by the HADCM3 and CCCMA climate models for 2050 to predict the likely distributions of LF under future climate and population changes. We show that LF prevalence is strongly influenced by spatial autocorrelation between locations but is only weakly associated with environmental covariates. Infection prevalence, however, is found to be related to variations in population density. All associations with key environmental/demographic variables appear to be complex and non-linear. LF prevalence is predicted to be highly heterogenous across Africa, with high prevalences (>20%) estimated to occur primarily along coastal West and East Africa, and lowest prevalences predicted for the central part of the continent. Error maps, however, indicate a need for further surveys to overcome problems with data scarcity in the latter and other regions. Analysis of future changes in prevalence indicates that population growth rather than climate change per se will represent the dominant factor in the predicted increase/decrease and spread of LF on the continent. We indicate that these results could play an important role in aiding the development of strategies that are best able to achieve the goals of parasite elimination locally and globally in a manner that may also account for the effects of future climate change on parasitic infection.

Project description:Modeling ecological patterns and processes often involve large-scale and complex high-dimensional spatial data. Due to the nonlinearity and multicollinearity of ecological data, traditional geostatistical methods have faced great challenges in model accuracy. As machine learning has increased our ability to construct models on big data, the main focus of the study is to propose the use of statistical models that hybridize machine learning and spatial interpolation methods to cope with increasingly large-scale and complex ecological data. Here, two machine learning algorithms, boosted regression tree (BRT) and least absolute shrinkage and selection operator (LASSO), were combined with ordinary kriging (OK) to model plant invasions across the eastern United States. The accuracies of the hybrid models and conventional models were evaluated by 10-fold cross-validation. Based on an invasive plants dataset of 15 ecoregions across the eastern United States, the results showed that the hybrid algorithms were significantly better at predicting plant invasion when compared to commonly used algorithms in terms of RMSE and paired-samples t-test (with the p-value < .0001). Besides, the additional aspect of the combined algorithms is to have the ability to select influential variables associated with the establishment of invasive cover, which cannot be achieved by conventional geostatistics. Higher accuracy in the prediction of large-scale biological invasions improves our understanding of the ecological conditions that lead to the establishment and spread of plants into novel habitats across spatial scales. The results demonstrate the effectiveness and robustness of the hybrid BRTOK and LASOK that can be used to analyze large-scale and high-dimensional spatial datasets, and it has offered an optional source of models for spatial interpolation of ecology properties. It will also provide a better basis for management decisions in early-detection modeling of invasive species.

Project description:BackgroundMorbidity due to Schistosoma haematobium and hookworm infections is marked in those with intense co-infections by these parasites. The development of a spatial predictive decision-support tool is crucial for targeting the delivery of integrated mass drug administration (MDA) to those most in need. We investigated the co-distribution of S. haematobium and hookworm infection, plus the spatial overlap of infection intensity of both parasites, in Ghana. The aim was to produce maps to assist the planning and evaluation of national parasitic disease control programs.Methodology/principal findingsA national cross-sectional school-based parasitological survey was conducted in Ghana in 2008, using standardized sampling and parasitological methods. Bayesian geostatistical models were built, including a multinomial regression model for S. haematobium and hookworm mono- and co-infections and zero-inflated Poisson regression models for S. haematobium and hookworm infection intensity as measured by egg counts in urine and stool respectively. The resulting infection intensity maps were overlaid to determine the extent of geographical overlap of S. haematobium and hookworm infection intensity. In Ghana, prevalence of S. haematobium mono-infection was 14.4%, hookworm mono-infection was 3.2%, and S. haematobium and hookworm co-infection was 0.7%. Distance to water bodies was negatively associated with S. haematobium and hookworm co-infections, hookworm mono-infections and S. haematobium infection intensity. Land surface temperature was positively associated with hookworm mono-infections and S. haematobium infection intensity. While high-risk (prevalence >10-20%) of co-infection was predicted in an area around Lake Volta, co-intensity was predicted to be highest in foci within that area.Conclusions/significanceOur approach, based on the combination of co-infection and co-intensity maps allows the identification of communities at increased risk of severe morbidity and environmental contamination and provides a platform to evaluate progress of control efforts.

Project description:BackgroundThe highly infectious coronavirus disease (COVID-19) was first detected in Wuhan, China in December 2019 and subsequently spread to 212 countries and territories around the world, infecting millions of people. In India, a large country of about 1.3 billion people, the disease was first detected on January 30, 2020, in a student returning from Wuhan. The total number of confirmed infections in India as of May 3, 2020, is more than 37,000 and is currently growing fast.ObjectiveMost of the prior research and media coverage focused on the number of infections in the entire country. However, given the size and diversity of India, it is important to look at the spread of the disease in each state separately, wherein the situations are quite different. In this paper, we aim to analyze data on the number of infected people in each Indian state (restricted to only those states with enough data for prediction) and predict the number of infections for that state in the next 30 days. We hope that such statewise predictions would help the state governments better channelize their limited health care resources.MethodsSince predictions from any one model can potentially be misleading, we considered three growth models, namely, the logistic, the exponential, and the susceptible-infectious-susceptible models, and finally developed a data-driven ensemble of predictions from the logistic and the exponential models using functions of the model-free maximum daily infection rate (DIR) over the last 2 weeks (a measure of recent trend) as weights. The DIR is used to measure the success of the nationwide lockdown. We jointly interpreted the results from all models along with the recent DIR values for each state and categorized the states as severe, moderate, or controlled.ResultsWe found that 7 states, namely, Maharashtra, Delhi, Gujarat, Madhya Pradesh, Andhra Pradesh, Uttar Pradesh, and West Bengal are in the severe category. Among the remaining states, Tamil Nadu, Rajasthan, Punjab, and Bihar are in the moderate category, whereas Kerala, Haryana, Jammu and Kashmir, Karnataka, and Telangana are in the controlled category. We also tabulated actual predicted numbers from various models for each state. All the R2 values corresponding to the logistic and the exponential models are above 0.90, indicating a reasonable goodness of fit. We also provide a web application to see the forecast based on recent data that is updated regularly.ConclusionsStates with nondecreasing DIR values need to immediately ramp up the preventive measures to combat the COVID-19 pandemic. On the other hand, the states with decreasing DIR can maintain the same status to see the DIR slowly become zero or negative for a consecutive 14 days to be able to declare the end of the pandemic.

Dataset Information

Spatial prediction and mapping of the COVID-19 hotspot in India using geostatistical technique

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets