Building safer, clearer medical summaries from big language models
Official title: Building Safety Guards into LLMs for Trustworthy Automatic Simplification of Medical Documents
Funded at University of Texas at Austin · NIH-11403974 · Data updated 2026-11-12
This project aims to make trustworthy, easy-to-understand versions of medical papers for the general public while preventing misinformation.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of Texas at Austin NIH-funded |
| Lab location | 1 site (Austin, United States) |
| Project start | September 18, 2024 |
| Project end | August 31, 2028 |
| Funding year | 2026 |
| Award (this year) | $336,527 |
| NIH project number | 5R01LM014600-03 |
| Project ID | NIH-11403974 on NIH RePORTER |
Who could benefit from this research
✓Could help
Not applicable; this grant focuses on developing safer AI for medical text simplification rather than recruiting patients.
✕Less likely to help
Direct patient participation is not part of the grant activities.
What this research studies
The team will create controllable, transparent, and moderately sized language models to produce simplified medical texts. They will develop error-detection tools and safeguards so summaries stay accurate and helpful for readers without medical background. The work includes designing tools that let communities balance usefulness with critical evaluation of outputs, aiming to improve health literacy. This is meant to empower people to better understand new medical findings through safer AI-generated explanations.
Why it matters
Potential benefit: If successful, patients and the public could access clearer explanations of medical research, helping them make informed health decisions.
How similar studies have gone: Previous work shows AI can summarize text, but reliability and safety remain major challenges, making this approach novel and necessary.
Where this research is happening
-
University of Texas at Austin
Austin, United States Active
Researchers
- Principal investigator: Li, Junyi Jessy — University of Texas at Austin
- Study coordinator: Li, Junyi Jessy
Related research
Other NIH-funded projects on the same topics.
-
Translating new nerve grafts to clinical use
Aims to bring advanced nerve graft technology that uses PEG-fusion and localized immunosuppression to patients with nerve gaps.
-
GABA-A receptor gating mechanisms and how drugs affect them
Researchers are studying how GABA-A receptor gates work and how medicines can modulate them to help people with epilepsy, Alzheimer's-related symptoms, and other brain disorders.
-
Nanoparticles that calm inflammation after artery procedures
A new nanoparticle inspired by dying cells is being tested to reduce inflammation after artery procedures and help PAD patients heal better.
-
Silent speech interface using a wearable tongue tracker to help people after laryngectomy communicate
A wearable system with tongue sensors and intraoral components aims to let people without a vocal cord use natural-sounding speech in daily life.
-
Understanding and reversing abnormal brain activity in Alzheimer's disease
Aims to understand how unusual brain activity contributes to Alzheimer’s and test ways to reverse it to help people with the condition.
-
CFDE Cloud Workspace to help researchers share and analyze data
A cloud-based workspace will let a wide range of researchers import, integrate, and analyze data from CFDE and other sources using familiar tools and workflows.
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page.