Personalized AI to predict sudden kidney injury
Official title: Personalized Machine Learning for Acute Kidney Injury Prediction and Prognosis
Funded at University of Florida · NIH-11345244 · Data updated 2026-11-12
This project uses personalized machine learning on hospital records to predict which hospitalized patients are likely to develop acute kidney injury so clinicians can act sooner.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of Florida NIH-funded |
| Lab location | 1 site (Gainesville, United States) |
| Project start | April 15, 2024 |
| Project end | March 31, 2027 |
| Funding year | 2026 |
| Award (this year) | $255,004 |
| NIH project number | 5R01DK137881-03 |
| Project ID | NIH-11345244 on NIH RePORTER |
Who could benefit from this research
✓Could help
Hospitalized adults with risk factors for acute kidney injury—especially critically ill patients—with available electronic health records are the most likely candidates.
✕Less likely to help
People who are not hospitalized, who lack comprehensive hospital EHR data, or whose kidney problems are chronic and unrelated to in-hospital acute events may not benefit from this work.
What this research studies
If I'm a hospitalized patient, researchers will use my electronic health record—including labs, medications, and vital signs—to build a computer model tailored to patients like me that estimates short-term risk of acute kidney injury. They will compare these personalized models to traditional one-size-fits-all models to find hidden patient subgroups and improve prediction accuracy. The team will train and test models using hospital EHR data and prior examples where global models miss high-risk patients. The aim is to give doctors earlier, more specific warnings so they can target prevention and reduce kidney damage.
Why it matters
Potential benefit: If successful, this could provide earlier, tailored warnings that help prevent or reduce the severity of acute kidney injury and its long-term complications.
How similar studies have gone: Other machine-learning models have improved AKI prediction in hospitals, but personalized, subgroup-focused approaches are newer and less tested in routine care.
Where this research is happening
-
University of Florida
Gainesville, United States Active
Researchers
- Principal investigator: Liu, Mei — University of Florida
- Study coordinator: Liu, Mei
Related research
Other NIH-funded projects on the same topics.
-
Using machine learning to predict and understand acute kidney injury
This study is working on using advanced computer techniques to better predict and understand acute kidney injury (AKI) for each patient, so doctors can spot those at higher risk early and provide the right care to prevent serious problems.
-
Colorectal cancer prognosis through a broad, collaborative study
A large, multi-center study follows people newly diagnosed with colorectal cancer to find biomarkers and patterns that predict outcomes and guide future care.
-
AI-powered digital pathology to tailor cancer treatment decisions
A team at Emory is building an AI-powered platform that analyzes cancer slides to help doctors choose the right therapies and avoid unnecessary chemotherapy or radiation for patients.
-
Harnessing the body's response to leptospirosis to aid diagnosis and prognosis
This grant aims to develop new ways to detect leptospirosis earlier and predict how severe it might become by studying the body's own response.
-
Harnessing the body’s response to Leptospirosis to improve diagnosis and prognosis
A research program at UCLA aims to use how the body responds to leptospirosis to make better diagnoses and predict outcomes for patients.
-
Harnessing the body's response to a dangerous infection to help diagnose and predict outcomes
A large, multi-project effort aims to use how the body reacts to leptospirosis to improve diagnosis and prognosis for patients affected by this infection.
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.