AI prediction of one-year death and repeat stroke after ischemic stroke
Official title: Machine Learning Prediction of 1-Year Mortality and Recurrence after Ischemic Stroke Using Enriched EHR data
Funded at Pennsylvania State Univ Hershey Med Ctr · NIH-11144923 · Data updated 2026-11-12
This project uses artificial intelligence and medical record data to predict whether people who had an ischemic stroke will die or have another stroke within one year.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Pennsylvania State Univ Hershey Med Ctr NIH-funded |
| Lab location | 1 site (Hershey, United States) |
| Project start | August 11, 2023 |
| Project end | July 31, 2027 |
| Funding year | 2025 |
| Award (this year) | $637,103 |
| NIH project number | 5R01NS128986-03 |
| Project ID | NIH-11144923 on NIH RePORTER |
Who could benefit from this research
✓Could help
People who have experienced an ischemic stroke and whose electronic health records are available in the participating hospitals would be the ideal candidates for the underlying data used in this work.
✕Less likely to help
People without accessible electronic health records in the participating systems, those with hemorrhagic (non-ischemic) strokes, or patients with very incomplete medical records may not benefit from these predictions.
What this research studies
If you've had an ischemic stroke, researchers will use electronic health records from multiple hospitals to train AI models that estimate your chance of dying or having another stroke within 12 months. They will enrich those records with extra clinical details and compare different machine-learning methods to find the most accurate predictions. The team will test the models across different health systems to reduce bias and improve how well the predictions work for many people. The idea is to make a reliable tool clinicians could use to spot patients at higher risk and guide follow-up care.
Why it matters
Potential benefit: If successful, this could help clinicians identify stroke survivors at high risk and target prevention and follow-up care to reduce deaths and recurrent strokes.
How similar studies have gone: Earlier pilot models using EHR data and machine learning showed promising accuracy but were limited to a single health system and need broader validation.
Where this research is happening
-
Pennsylvania State Univ Hershey Med Ctr
Hershey, United States Active
Researchers
- Principal investigator: Abedi, Vida — Pennsylvania State Univ Hershey Med Ctr
- Study coordinator: Abedi, Vida
Related research
Other NIH-funded projects on the same topics.
-
Predicting 1-year death and recurrence after ischemic stroke using enriched EHR data
Researchers are using advanced computer models to identify which stroke patients are at higher risk of dying or having another stroke within a year, by analyzing enriched electronic health records.
-
Using machine learning to predict death and recurrence after ischemic stroke
This study is looking to help doctors better predict how likely it is for patients to face serious health issues or pass away within a year after having an ischemic stroke, using advanced technology to analyze health records, so they can provide better care and support for those at higher risk.
-
Improving blood flow after stroke by breaking down NETs
Aims to see if breaking down neutrophil extracellular traps (NETs) after an ischemic stroke can improve blood flow and reduce brain injury in patients.
-
EP2 blockers to protect the brain after a stroke
Aims to test brain-permeable EP2 antagonists to reduce damage after ischemic stroke and improve recovery.
-
Genetic testing to find causes of early-onset ischemic stroke
This project uses whole-exome sequencing to look for gene changes that may cause ischemic stroke in adults who had a stroke at a young age.
-
Targeted nanomedicine to protect the brain after stroke
A targeted drug-delivery approach aims to boost blood-brain barrier repair and reduce brain damage after an acute ischemic stroke in patients at risk.
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.