Our population is aging rapidly. More people are living longer with multiple conditions, and more people are at risk for falls, dementia, and medication harm. Yet healthcare systems don't consistently meet the complex needs and priorities of older adults, older patients often receive care that is fragmented, disease-focused, and burdensome.
Real-world data refers to health information that is routinely collected during everyday patient care. For complex patients, there is often no clear clinical guideline, no expert consensus, and no randomized controlled trial to guide the decision, yet every patient encounter generates data that could improve the next patient's care, if only we build the science to make that possible.
LEAF-AI Lab is guided by the integration of two complementary frameworks.
Led by The John A. Hartford Foundation and the Institute for Healthcare Improvement (IHI), and adopted by more than 3,000 healthcare organizations worldwide. Its 4Ms Framework, What Matters, Medication, Mentation, and Mobility, provides an evidence-based approach to delivering person-centered, age-friendly care.
A cyclical process that continuously transforms real-world data into actionable knowledge, integrates that knowledge into practice, and generates new evidence through an ongoing cycle of learning.
The two frameworks complete each other: the 4Ms define what to deliver, while Learning Health Systems define how to continuously improve care. Today, however, healthcare systems are unable to fully leverage real-world data to support continuous learning and personalized, age-friendly care.
“Our research develops human-centered, trustworthy AI solutions through team-science approaches to accelerate the use of real-world data for improving age-friendly care.”
Our research follows three interconnected pillars: Modeling, Translating, and Scaling. We develop AI methods that transform clinical data into real-world evidence, translate those innovations into scalable healthcare solutions, and build the collaborative infrastructure that enables learning health ecosystems to continuously improve care. Learn more about our applications →
Most of what we know about how older adults are actually doing, whether they can still dress themselves, whether they fell, whether they're confused, is buried in unstructured clinical text, not billing codes. We build NLP systems, from rule-based to fine-tuned to LLM-based, that turn that language into standardized, usable knowledge about geriatric syndromes like delirium, falls, and medication risk.
A model that works in one hospital often breaks in the next, different EHRs, different documentation habits, different infrastructure. We study and solve the real barriers, legal, technical, data quality, institutional variation, that stand between a validated model and one that actually helps patients at scale.
None of this scales alone. We help build and grow the shared infrastructure, data networks, NLP consortia, consensus standards, and working groups, that let institutions across the country learn from each other's real-world data, together.
We work in close partnership with a growing set of national research networks and professional communities, helping shape the infrastructure and standards our research depends on.
A federated research network linking EHRs across 140M+ patients and 50+ CTSAs nationwide.
Open Health Natural Language Processing Consortium, open-source clinical NLP development since 2009, 100+ repositories.
A multi-country working group advancing NLP methods for Age-Friendly Health Systems, 44 members across 15 countries.
Its mission is to facilitate communication, collaboration, training, and networking for researchers who develop, apply, and promote NLP in biomedical science, patient care, public health, and biomedical education. Sunyang serves as Vice Chair.
Network for Investigation of Delirium: Unifying Scientists, a national delirium research network.
The 4Ms framework (What Matters, Medication, Mentation, Mobility) anchoring our clinical vision.
At LEAF-AI, we are building the scientific foundation for age-friendly, AI-enabled healthcare. Our work transforms real-world clinical data into actionable knowledge that helps healthcare systems continuously learn, improve, and deliver better care for older adults.
As part of Dell Medical School at The University of Texas at Austin, we are contributing to a bold vision for the future of healthcare. Dell Medical School's mission is to revolutionize how people get and stay healthy, and the university's recent investment in an AI-native hospital and research campus creates an extraordinary opportunity to shape how AI is developed, evaluated, and translated into clinical practice.
We are looking for motivated students, trainees, and collaborators who are curious, collaborative, and passionate about solving meaningful healthcare challenges. If you enjoy working as part of a team, tackling complex problems, and advancing AI that makes a real difference in people's lives, we invite you to join LEAF-AI and help build that future together.
