...

The process of getting data into a data warehouse is often called ETL: *E*xtraction, *T*ransformation, *L*oading. Extraction means that data is gathered from a data source by parsing, therefore, an understanding of the underlying data model is required. Transformation is the process of mapping the source data model to the target data model, in this case i2b2's database schema for every data instance. Loading is a feature to persist the transformation by adding the data to the target (the i2b2 database).

...

The IMT operates on two levels The . As shown in the big picture is shown belowbelow, it provides basic ETL for various data sources and formats by using dedicated "extractors". Additionally, it allows for modifying the i2b2 ontology.