Secure and PrivAte Collaborative EnvironmentS (SPACES) for biomedical analytics

用于生物医学分析的安全且私密的协作环境 隐私和机密性对医疗保健至关重要。然而，保护隐私 是一项不平凡的任务，因为任何保护方案本质上都涉及 与数据实用程序的权衡。此外，缺乏对生物医学数据的访问可能导致 到分散的医疗保健，导致更高的经济和社会成本，更高 可避免的药物相互作用副作用的安全风险，是一个重要的 生物医学研究的障碍。该项目的主要目的是促进 通过开发技术在协作环境中进行生物医学研究 安全和隐私保护的探索性分析。这对大数据至关重要 到知识(BD2K)倡议，因为它可以促进 协作环境。 拟议工作涉及两个相辅相成的目标。首先，我们将发展 能够对数据进行探索性分析以确定其可用性的技术 并与特定的生物医学研究任务相关。我们还将发展 能够测量和缓解其他 由于访问此数据而带来的隐私/安全风险，从而实现对 数据。建议的解决方案将利用并提升最先进的技术 保护隐私的采样和分析等技术解决方案 算法、基于风险的访问控制和查询审计。通过这样做， 拟议的工作将加强生物医学研究设计、探索性数据分析、 和假设发现，而不会损害隐私。 该项目将产生开放源码、免费可用的软件工具来执行 效用分析和风险分析。这些将被整合到RedCap，一个数据 一种广泛用于提供翻译服务的采集管理系统 研究信息学支持。罗格斯大学和加州大学圣地亚哥分校正在进行的几个项目将 作为拟议工作的初始试点客户，与其他研究小组一起 在后来的阶段也参与其中。

Secure and Private Collaborative Environments for Biomedical Analytics Privacy and confidentiality are critical to healthcare. However, preserving privacy is a non-trivial task because any protection scheme essentially involves a tradeoff with data utility. Furthermore, lack of access to biomedical data can lead to fragmentation of care, resulting in higher economic and social costs, higher safety risk from avoidable drug interactions side-effects, and is a significant impediment to biomedical research. The primary aim of this project is to facilitate biomedical research in collaborative environments by developing technologies for secure and privacy-preserving exploratory analysis. This is critical to the big data to knowledge (BD2K) initiative since it can facilitate biomedical research in collaborative environments. The proposed work addresses two complementary aims. First, we will develop technologies that enable exploratory analysis of data to determine its usability and relevance to the specific biomedical research task. We will also develop technologies to enable the measurement and mitigation of additional privacy/security risk due to accessing this data, thus enabling proper control over the data. The proposed solutions will utilize and enhance the state of the art technological solutions such as privacy-preserving sampling and analysis algorithms, risk-based access control, and query auditing. By doing so, the proposed work will enhance biomedical study design, exploratory data analysis, and hypothesis discovery without compromising on privacy. The project will result in open-source, freely available software tools to perform utility analysis and risk analysis. These will be integrated into REDCap, a data collection and management system used widely for providing translational research informatics support. Several ongoing projects at Rutgers and UCSD will serve as initial pilot customers for the proposed work, with other research groups also being involved at a later stage.