Administrative Supplement: Enhance the Utility of Data Available through the Childhood Cancer Data Initiative (CCDI) Ecosystem

项目总结/摘要 美国国家癌症研究所（NCI）儿童癌症数据倡议（CCDI）正在建立一个生态系统， 从所有儿童癌症患者身上学习。其中一个重要组成部分是增加 纵向临床数据的可用性，这些数据可以与提交给CCDI的其他关键数据相结合。的 当前用于获取这些类型的数据的标准实践是通过临床研究 协调员，这是耗时的，并受到人为错误。电子健康记录（EHR）具有 有可能扩大流入CCDI生态系统的数据量。然而，他们也提出了几个 由于缺乏结构和标准化而面临的挑战。本研究的目的是将联合收割机 先前开发的获取数据的方法，以便向医疗机构提供干净的临床相关数据。 CCDI。第一个包ExtractEHR从EHR中提取数据，并利用一系列后处理 编码将原始EHR数据转换为临床相关的、人类可理解的关于儿童的数据 癌症经历第二种方法使用基于快速医疗保健互操作性资源（FHIR）的 方法，为更可扩展的工作流程和临床数据交换提供技术框架。 本研究旨在将联合收割机ExtractEHR与FHIR框架开发相结合，以扩展 CCDI提供的临床数据。该提案将通过以下方式确定持续贡献的可扩展方式： 实现以下两个具体目标。第一个目标是提取和处理电子病历数据，以提供临床 关于CCDI生态系统中具有分子数据的患者的治疗和结局的背景。实现 为此，ExtractEHR将提取先前同意加入儿童脑肿瘤网络的患者的数据 （CBTN）协议在两个大型儿童医院，费城儿童医院（CHOP）和 亚特兰大儿童保健（CHOA）。将对提取的数据进行处理，以提供临床相关性 治疗和结果数据将被转移到CCDI。这将证明可行性和 通过向CCDI提供一组EHR数据来促进协作， 与CCDI团队讨论工作流程或数据的哪些方面需要优化和/或扩展。 第二个目标是开发一个基于FHIR的ExtractEHR版本，并确定其可扩展性和成功性， 与获得有临床意义的数据所需的后处理工作流程进行交互。将努力 将ExtractEHR功能导出到CHOP的FHIR框架中，并在CHOA测试此流程。 这项研究将创建管道，从清理和处理的EHR数据中提供粒度数据， CCDI中包含的患者的临床背景，并将确定是否可以使用基于FHR的方法作为 一种可扩展的方法，用于使用ExtractEHR来确定要包含在CCDI生态系统中的数据。

PROJECT SUMMARY/ABSTRACT The National Cancer Institute (NCI) Childhood Cancer Data Initiative (CCDI) is building an ecosystem to gather data and learn from all patients with childhood cancer. An important component of this is increasing the availability of longitudinal clinical data that can be integrated with other crucial data submitted to the CCDI. The current standard practice for acquiring these types of data is manual abstraction by a clinical research coordinator, which is time consuming and subject to human error. Electronic health records (EHRs) have the potential to expand the amount of data flowing into the CCDI ecosystem. However, they also present several challenges due to lack of structure and standardization. This objective of this study is to combine two previously developed approaches to obtaining data in order to provision clean, clinically-relevant data to the CCDI. The first package, ExtractEHR, extracts data from the EHR, and utilizes a series of post-processing coding to transform the raw EHR data into clinically relevant, human understandable data about the childhood cancer experience. The second approach uses Fast Healthcare Interoperability Resources (FHIR)-based methods to provide a technical framework for more scalable workflows and interchange of clinical data. This study aims to combine ExtractEHR with the FHIR framework development to enable the expansion of clinical data available to CCDI. The proposal will determine scalable ways for ongoing contributions by achieving the following two specific aims. The first aims is to extract and process EHR data to provide clinical context regarding treatment and outcomes for patients with molecular data in the CCDI ecosystem. To achieve this goal, ExtractEHR will be extract data on patients previously consented to Children’s Brain Tumor Network (CBTN) protocols at two large children’s hospitals, the Children’s Hospital of Philadelphia (CHOP) and Children’s Healthcare of Atlanta (CHOA). Extracted data will be processed to provide clinically-relevant treatment and outcomes data that will be transferred to the CCDI. This will demonstrate the feasibility and operability of a seamless workflow by delivering a set of EHR data to CCDI to facilitate collaboration and discussion with CCDI teams on which aspects of the workflow or data to optimize and/or expand on. The second aim is to develop a FHIR-based version of ExtractEHR and determine its scalability and success in interacting with the post-processing workflow needed to obtain clinically meaningful data. Efforts will be made to export ExtractEHR functionality into a FHIR framework at CHOP and to test this process at CHOA. This study will create pipelines that provide granular data from cleaned and processed EHR data that offers clinical context for patients included in the CCDI and will determine if FHIR-based approaches can be used as a scalable method for using ExtractEHR to ascertain data to be included in the CCDI ecosystem.