Anderson Wes, Bhatnagar Roopal, Scollick Keith, Schito Marco, Walls Ramona, Podichetty Jagdeep T
Critical Path Institute, Tucson, Arizona, USA.
Clin Transl Sci. 2024 Dec;17(12):e70078. doi: 10.1111/cts.70078.
In the rapidly evolving landscape of healthcare and drug development, the ability to efficiently collect, process, and analyze large volumes of real-world data (RWD) is critical for advancing drug development. This article provides a blueprint for establishing an end-to-end data and analytics pipeline in a cloud-based environment. The pipeline presented here includes four major components, including data ingestion, transformation, visualization, and analytics, each supported by a suite of Amazon Web Services (AWS) tools. The pipeline is exemplified through the CURE ID platform, a collaborative tool designed to capture and analyze real-world, off-label treatment administrations. By using services such as AWS Lambda, Amazon Relational Database Service (RDS), Amazon QuickSight, and Amazon SageMaker, the pipeline facilitates the ingestion of diverse data sources, the transformation of raw data into structured formats, the creation of interactive dashboards for data visualization, and the application of advanced machine learning models for data analytics. The described architecture not only supports the needs of the CURE ID platform, but also offers a scalable and adaptable framework that can be applied across various domains to enhance data-driven decision making beyond drug repurposing.
在快速发展的医疗保健和药物研发领域，有效收集、处理和分析大量真实世界数据（RWD）的能力对于推进药物研发至关重要。本文提供了一个在基于云的环境中建立端到端数据和分析管道的蓝图。这里介绍的管道包括四个主要组件，即数据摄取、转换、可视化和分析，每个组件都由一套亚马逊网络服务（AWS）工具提供支持。该管道通过CURE ID平台进行示例说明，CURE ID平台是一个协作工具，旨在捕获和分析真实世界中的非标签治疗用药情况。通过使用AWS Lambda、亚马逊关系数据库服务（RDS）、亚马逊QuickSight和亚马逊SageMaker等服务，该管道便于摄取各种数据源，将原始数据转换为结构化格式，创建用于数据可视化的交互式仪表板，并应用先进的机器学习模型进行数据分析。所描述的架构不仅支持CURE ID平台的需求，还提供了一个可扩展且适应性强的框架，该框架可应用于各个领域，以加强超越药物重新利用的数据驱动决策。
