Center for Environmental and Health Effects of PFAS

项目摘要 意义：在过去的几十年里，我们对环境的不利影响的认识， 人类健康的质量大幅提高，需要监测和预测的工具 环境完整性。因此，已作出努力，有效地监测和管理潜在的 环境压力，以改善环境健康，如EPA的环境质量指数。 当前的环境建模工具试图对地理空间数据进行建模以预测环境质量（EQ）， 包括汇总不同环境领域的监测和社会人口数据。 创新/独特性：然而，目前的建模工具并不提供全面和动态的 为环境和人类健康进行建模预测。环境完整性框架（EIF） 来填补这些空白。拟议的研究旨在建立一个综合框架，提供可更新的 通过整合多个环境领域并包括其他环境领域， 环境污染物，如PFAS。监督机器学习（ML）或“白盒”方法将 生成可解释的EQ预测，并可通过交互式仪表板访问广泛的用户群。一个 将成立由社区利益相关者和SRP研究人员组成的EIF发展咨询小组， 进一步努力优先考虑高危社区最关心的问题，并保持一定程度的 在EIF的发展过程中保持信任和透明度。这个广泛的用户群将包括研究人员和风险 在研究中心监测EQ的评估员。此外，我们还寻求让风险社区的青年参与增强综合框架 讲习班仪表板，以促进环境卫生方面的科学知识和交流 问题. 我们的目标是： 1.采用先进和创新的集成机器学习（ML）技术， 综合和可互操作的建模环境完整性预测。使用多个ML 模型的集成方法，以产生特定领域的EQ分数的组合环境 风险场所的完整性评分。PFAS浓度与污染物的关系 不同领域的暴露水平将被纳入综合评分， 建模框架（EIF）。 2.人类健康风险成果和方法的翻译传播 在NIEHS转化研究框架内。专注于沟通和社区 通过与SRP研究人员和社区利益相关者的合作，加强EIF的参与部分。 通过向风险社区通报PFAS， 当地社区的环境质量问题。

PROJECT SUMMARY Significance: Over the past several decades, our understanding of the adverse effects of environmental quality on human health has risen substantially, and there is a need for tools that monitor and predict environmental integrity. As a result, efforts have been made to effectively monitor and regulate potential environmental stressors to improve environmental health, such as the EPA’s Environmental Quality Index. Current environmental modeling tools try to model geospatial data to predict environmental quality (EQ), including aggregating different environmental domain monitoring and sociodemographic data. Innovation/Uniqueness: However, current modeling tools do not provide comprehensive and dynamic modeling predictions for environmental and human health. The Environmental Integrity Framework (EIF) seeks to fill these gaps. The proposed research aims to build a comprehensive framework that provides updateable EQ and human health predictions by integrating several environmental domains and including other environmental contaminants such as PFAS. Supervised machine learning (ML) or “white-box” methods will generate interpretable EQ predictions and be accessible via interactive dashboards to a broad user base. An EIF development Advisory group composed of community stakeholders and SRP researchers will be formed to further our efforts to prioritize what at-risk communities are concerned about the most and uphold a level of trust and transparency during the development of the EIF. This wide userbase will include researchers and risk assessors monitoring EQ at a site. Also, we seek to engage youth in at-risk communities with the EIF dashboard in workshops to promote science literacy and communication regarding environmental health issues. Our aims are to: 1. Incorporate advanced and innovative ensemble machine learning (ML) techniques for integrative and interoperable modeling Environmental Integrity Predictions. Employ several ML models in an ensemble approach to generate domain-specific EQ scores for a combined Environmental Integrity Score for at-risk sites. The relationships of the PFAS concentrations and contaminant exposure levels across the different domains will be integrated into a comprehensive scoring and modeling framework (EIF). 2. Prioritize Translational Dissemination of human health risks outcomes and methodologies within the NIEHS Translational Research Framework. Focus on the communication and community engagement portion of the EIF via collaborations with SRP researchers and community stakeholders. Focus on establishing trust and transparency by informing at-risk communities on PFAS and environmental quality-related issues in their local communities.