• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE International Symposium on Computer-Based Medical Systems >Improving Data Quality in Medical Research: A Monitoring Architecture for Clinical and Translational Data Warehouses
【24h】

Improving Data Quality in Medical Research: A Monitoring Architecture for Clinical and Translational Data Warehouses

机译：改善医学研究中的数据质量：临床和转化数据仓库的监视体系结构

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Clinical and translational data warehouses are important infrastructure building blocks for modern data-driven approaches in medical research. These analytics-oriented databases have been designed to integrate heterogeneous biomedical datasets from different sources and to support use cases such as cohort selection and ad-hoc data analyses. However, the lack of clear definitions of source data and controlled data collection procedures often raises concerns about the quality of data provided in such environments and, consequently, about the evidence level of related findings. To address these problems, we present an architecture that helps to monitor data quality issues when importing data into warehousing solutions using ETL (Extraction, Transformation, Load) processes. Our approach provides software developers with an API (Application Programming Interface) for logging detailed and structured information about data quality issues encountered. This information can then be displayed in dynamic dashboards, the evolution of data quality can be monitored over time, and quality issues can be traced back to their source. Our architecture supports several well-known data quality dimensions, addressing conformance, completeness, and plausibility. We present an open-source implementation, which is compatible with common clinical and translational data warehousing platforms, such as i2b2 and tranSMART, and which can be used in conjunction with many ETL environments.
机译：临床和转换数据仓库是医学研究中现代数据驱动方法的重要基础架构构建块。这些面向分析的数据库旨在整合来自不同来源的异构生物医学数据集，并支持诸如人群选择和临时数据分析之类的使用案例。但是，缺乏对源数据的明确定义和受控的数据收集程序常常引起人们对在这种环境下提供的数据质量以及因此对相关发现的证据水平的关注。为了解决这些问题，我们提出了一种体系结构，当使用ETL（提取，转换，加载）流程将数据导入到仓库解决方案中时，有助于监视数据质量问题。我们的方法为软件开发人员提供了一个API（应用程序编程接口），用于记录有关遇到的数据质量问题的详细和结构化信息。然后，可以在动态仪表板中显示此信息，可以随时间监视数据质量的演变，并且可以将质量问题追溯到其来源。我们的体系结构支持几个众所周知的数据质量维度，以解决一致性，完整性和合理性问题。我们提供了一种开源实现，该实现与常见的临床和转换数据仓库平台（例如i2b2和tranSMART）兼容，并且可以与许多ETL环境结合使用。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号