Data-centric approach

We design eProtocol system in the data-centric manner as an "orchestration" of structured data (data entry part) and narrative components (authoring tool part). The problem to solve is repeated appearance of structured data elements within a narrative text which need to match with actual content in an automatic manner.

The data-centric approach to content management generally distinguishes:

1. A pool of data and metadata, managed by data entry forms to be stored as structured data in a database (e.g. Clinical Trial Planned Observation). This structured data can be reflected in the narrative part of the Protocol and can be forwarded to downstream activities, such as eCRF generation process.

2. A narrative depiction of the electronic Protocol, supported by an topic-oriented authoring tool that allows to define the document structure and to edit assigned narrative components consisting of text, lists, tables, etc. The narrative components can contain symbolic references to data elements of the data pool of the Protocol (replaced by their values when rendered in a viewer and as an output in the PDF format).

Other important feature of a data-centric approach is fine grained retrieving. In the authoring tool, each chapter has an identifier (Object Identifier, OID), which allows to narrow retrieval process to specific chapter (e.g. seeking for specific term in all "Primary Objectives" across Clinical Study Protocols). Such a selective match significantly increases the relevance of the context where the term was found.

An additional feature of our data-centric system is the possibility to automatic render some complex chapters like "Study Synopsis", "Visit Schedule", "Blood-log" which are really a nightmare for study planners by doing it manually.


Enter your comment or question