ABOUT THE EVENT
Date: Wednesday, August 27, 2025 1–2pm EDT
Event Location: Virtual
Real-world data offers powerful opportunities to study disease progression, treatment safety and effectiveness, and patient outcomes—but some of the most clinically meaningful details are buried in unstructured sources like clinical notes and diagnostic reports.
Truveta makes these data accessible and research ready. Truveta Data includes complete, de-identified EHR data from more than 120 million patients across the US—including insights from more than 7 billion clinical notes. Clinical concepts are extracted from notes using advanced AI and normalized to a common data model, enabling seamless integration with structured EHR data and facilitating powerful, precise analyses across therapeutic areas. Truveta Data is also linked with closed claims and mortality data for a complete view of the patient journey.
In this webinar, we’ll explore how the Truveta Language Model (TLM)—a multi-modal AI model trained on EHR data—unlocks insights from clinical notes at scale. We’ll spotlight cardiovascular research that relied on concepts like left ventricle ejection fraction (LVEF) and NYHA class to improve classification of heart failure and aortic stenosis, and to support deeper treatment analysis. We’ll also discuss how AI-extracted clinical concepts are being used across therapeutic areas to accelerate research across the product lifecycle.
What you’ll learn:
- How TLM extracts clinical concepts at scale from notes, such as LVEF, NYHA class, seizure frequency, migraine severity, ECOG status, and more
- How granular, normalized data improves cohort definitions and outcomes analyses
- How leading researchers are integrating AI-extracted concepts with EHR data to uncover new insights
Featured Speakers


Host
![Lori Ellis[square]](https://static.biospace.com/dims4/default/30ef6ab/2147483647/strip/true/crop/300x300+0+0/resize/100x100!/quality/90/?url=https%3A%2F%2F4413123.fs1.hubspotusercontent-na1.net%2Fhubfs%2F4413123%2FLori%20Ellis%20headshot%20300x300.png)