Geostatistical software for space-time interpolation and uncertainty modeling

描述（由申请人提供）：在对环境（例如空气污染、热浪）和健康结果（例如哮喘、心脏病、癌症）之间的关联和/或因果关系进行任何调查时，一个关键组成部分是在与健康结果相同的地理尺度和时间分辨率下获得准确的暴露模型。计算人类接触量对癌症来说尤其具有挑战性，因为癌症可能需要数年或数十年的时间才能发展，特别是在低水平污染物的情况下。在这种情况下，污染物水平很少可用于每个地点和时间间隔访问的 因此，需要通过时空插值填补数据空白。令人惊讶的是，目前还没有用于时空数据地质统计处理的商业软件，包括在未监测的时间和地点进行插值。该SBIR项目正在开发第一个商业软件，为地质统计ST插值和不确定性建模提供工具。该研究产品将是由Esri合作伙伴BioMedware开发的桌面时空可视化核心的独立模块。该软件包将提供一个全面的套件，用于：1）时空协方差函数的计算和基于时间序列的建模，2）与健康结果相同尺度的暴露数据的ST插值和随机建模（即个人层面或汇总），并使用任何可用的辅助信息（例如遥感、土地利用回归模型、空气扩散模型）; 3）通过暴露重建对估计值所附的不确定性进行量化和基于蒙特-卡罗的传播。这些工具将适用于分析健康科学以外的数据，如遥感、核环境工程或气候变化，从而大大拓宽最终产品的商业市场。该项目将实现三个目标：比较两类ST协方差模型的性能（即预测准确性，对确定性-响应评估的影响）和用户友好性（即易于推断，在软件中自动实现的潜力），包括平稳性，完全对称性，可分性和支持的紧凑性的主要假设。开发和测试一个原型模块，该模块将指导非专家用户选择和优化时空协方差模型，然后基于BioMedware的时空可视化和分析技术进行时空数据的插值。进行可用性研究，并确定在第二阶段考虑的其他方法和工具。这些技术、科学和商业创新将彻底改变我们对地理统计学时空现象进行建模的能力，并在与环境流行病学研究最相关的尺度（例如点位置、普查区一级）上计算估计值和相关的不确定性。

DESCRIPTION (provided by applicant): A key component in any investigation of association and/or cause-effect relationships between the environment (e.g. air pollution, heat waves) and health outcomes (e.g. asthma, heart disease, cancer) is the availability of accurate models of exposure at the same geographical scale and temporal resolution as the health outcomes. The computation of human exposure is particularly challenging for cancers since they may take years or decades to develop, especially in presence of low level of contaminants. In this situation pollutant levels are rarely available for every location and time interval visited by the subjects; therefore data gaps need to be filled-in through space-time (ST) interpolation. Surprisingly, there is currently no commercial software for the geostatistical treatment of space-time data, including the interpolation at unmonitored times and locations. This SBIR project is developing the first commercial software to offer tools for geostatistical ST interpolation and modeling of uncertainty. The research product will be a stand-alone module into the desktop space-time visualization core developed by BioMedware, an Esri partner. This software package will provide a comprehensive suite for: 1) the computation and advisor-guided modeling of space-time covariance functions, 2) the ST interpolation and stochastic modeling of exposure data at the same scale as health outcomes (i.e. individual-level or aggregated) and using any secondary information available (e.g. remote sensing, land-use regression model, air dispersion model), and 3) the quantification and Monte-Carlo based propagation of uncertainty attached to estimates through exposure reconstruction. These tools will be suited for the analysis of data outside health sciences, such as in remote sensing, nuclear environmental engineering or climate change, broadening significantly the commercial market for the end product. This project will accomplish three aims: ¿ Compare the performance (i.e. prediction accuracy, impact on exposure-response assessment) and user- friendliness (i.e. ease of inference, potential for automatic implementation in software) of two classes of ST covariance models that encompass the main hypotheses of stationarity, full symmetry, separability and supported compactness. ¿ Develop and test a prototype module that will guide non-expert users through the selection and optimal fitting of space-time covariance models, followed by the interpolation of space-time data based on BioMedware's space-time visualization and analysis technology. ¿ Conduct a usability study and identify additional methods and tools to consider in Phase II. These technologic, scientific and commercial innovations will revolutionize our ability to model geostatistically space-time phenomena and compute estimates and the associated uncertainty at the scale (e.g. point location, census-tract level) the most relevant for environmental epidemiological studies.