i2b2-etl

i2b2-etl imports ontology and patient data from simplified input file schemas into the i2b2 clinical research data warehouse to enable EHR-based research, population health management, and epidemiological analyses.

Key Features:

• Simplified data loading: Maps simplified input file schemas for ontology and patient data to the i2b2 schema for ingestion.
• Import and deletion operations: Supports creation and removal of i2b2 concepts, facts, and encounter records.
• Record-number de-identification: Performs record-number de-identification to remove direct patient identifiers prior to loading.
• Data validation: Executes validation checks on input data to ensure integrity of imported records.
• Implementation: Implemented in Python.

Scientific Applications:

• EHR-based research: Enables loading of electronic health record data into i2b2 for cohort discovery and clinical studies.
• Population health management: Supports population-management initiatives such as the deployment reported at MassGeneral Brigham.
• Clinical decision support and epidemiology: Facilitates construction of datasets for clinical decision support systems and epidemiological analyses.

Methodology:

Implemented in Python, i2b2-etl parses simplified input file schemas to generate i2b2 ontology and fact records and applies record-number de-identification and data validation.