Data Science Methods to Enable Real-world Evidence to Support Stroke Care (DRESS)
Problem and Need for the Study
Acute ischemic stroke (AIS) is a dangerous yet prevalent brain condition that can cause instant death or inflict chronic disability. Because of the complexity of the human brain, there is an unmet need to supplement data from randomized controlled trials and accelerate the investigation of patient responses to different treatments under various disease conditions.
Innovation and Impact
To inform treatment decisions for AIS, we propose developing an evidence-based precision medicine framework by mining the electronic health records (EHRs) at M Health Fairview. The overall goal is to enable real-world evidence to be used to support stroke care.
Our aims are to develop data science methods that:
- Curate functional outcomes data from EHRs
- Derive imaging markers that link damaged and salvageable locations in the brain to functional outcomes
- Assess treatment effect while adjusting for background temporal shifts in patient outcomes
Key Personnel and Performance Sites
University of Minnesota
- Principal Investigators: Jue Hou, Margy McCullough-Hicks, Rui Zhang, Ju Sun, Christopher Streib
The Data Science Initiative Seed Grant is a 1 year, $111,615 award.
Project dates: 01-January-2024 to 31-December-2024