Slide #1 image

i2b2-Ontology-Editor

Create new i2b2-Ontologies by coping and renaming i2b2-Ontologies, mapping items, adding start- and enddates and merging patient data with terminologies.

Slide #2 image

IDRT-Import-Tool

Import CDISC ODM, CSV/SQL and terminologies to i2b2 with a wizard-style GUI.

Slide #3 image

i2b2 Wizard

Install and configure i2b2.

NextPrev

Integrated Data Repository Toolkit (IDRT)

Introduction

i2b2 has turned out to be a very valuable component for secondary use of routine clinical data. Its pragmatic database schema allows merging of data from heterogeneous data sources, and the intuitive user interface enables easy querying and powerful processing. However, it's a component rather than a complete solution: The user is facing several barriers when integrating i2b2 into the operational workflow. Even though the i2b2 team did an exceptionally good job in documenting the whole system and providing an easy-to-use virtual machine for demonstration purposes, it's still hard to combine i2b2 with custom solutions. Although the database schema is easy to understand, a significant effort has to be spent to preprocess and import data as i2b2 provides only limited support for importing data through the workbench, but no way to automate a suitable ETL process (Extraction, Transformation and Loading). This also applies for terminologies that are required to access the data.

The Integrated Data Repository Toolkit (IDRT) was conceived to address both the issue of setup and administration as well as the ability to import clinical data in several standard formats, including terminologies.

Tools

  • i2b2 Wizard

    automates many core aspects of setting up and administrating an i2b2 instance, drastically reducing the barrier-to-entry for running an individual i2b2 installation.

  • IDRT-Import-Tool

    a Java GUI with an i2b2 server browser and convenient configuration wizards for starting the ETL. Therefor an extensive set of configurable Talend Open Studio ETL-jobs has been developed to import fact data from common standard formats such as CSV, CDISC ODM and SQL databases into the i2b2 system. The ETL process supports the automatic derivation of i2b2 ontologies from metadata (such as column names in CSV files) and the automatic replacement of flat code lists with standard terminological hierarchies (e.g. ICD-10) in the i2b2 ontology. We've also implemented import jobs for German standard terminologies and the German "§21" benchmarking data set.

  • i2b2-Ontology-Editor (IOE)

    a Java GUI to edit i2b2 Ontologies.