At the last Digital Data Flow Solution Showcase, Regina Preciado from Content Rules and I had a chance to share how CDISC’s Unified Study Definition Model (USDM data standard) supports content reuse from clinical trial protocols.
While USDM was initially conceived to facilitate auto-configuration of clinical systems based on more structured protocol elements (like the Schedule of Assessments), the last few versions of the standard account for all protocol content, including narrative (i.e., longer text) sections.
In the webinar, we discussed how semantic markup helps machines identify content by type and purpose, allowing better search, reuse, and long-term content management across the development lifecycle. We concluded that CDISC’s USDM provides a semantic standard for machine-friendly, interoperable content.
We also noted that metadata-tagged content improves LLM training and helps ensure accurate, reliable results from AI content generation (like RAG). This is a critical aspect, given our industry’s current focus on AI’s role in making clinical trials more efficient.
And of course, our colleagues Kirsten Walther Langendorf and Johannes Ulander from data4knowledge gave a fantastic overview of how USDM provides structure to all other aspects of study protocols.
If you didn’t see it live, you can watch the recording here.
futurpositif 