AI alerts from medical records to prevent care mistakes
Learning alerting models for clinical care from EMR data and human knowledge
This project trains AI to read hospital records and expert rules to flag when your care deviates from what similar patients usually receive so clinicians can be alerted.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of Pittsburgh at Pittsburgh NIH-funded |
| Lab location | 1 site (Pittsburgh, United States) |
| Project ID | NIH-11160433 on NIH RePORTER |
What this research studies
If you receive care in a hospital, this project uses past electronic medical records plus human clinical knowledge to teach computer models what typical care looks like for patients like you. The system predicts expected clinician actions and raises an alert when current care differs substantially from those expectations, for example when a needed medication is not reordered after surgery. Researchers build these models using historical patient records and expert input, and plan to test how well the alerts can spot preventable mistakes in real clinical settings. The team aims to integrate the alerts into hospital workflows so clinicians can act on them quickly.
Who could benefit from this research
Good fit: Ideal candidates are hospitalized patients whose care is recorded in participating electronic medical record systems, especially those on complex or high‑risk treatments.
Not a fit: Patients treated outside participating hospitals or whose records are incomplete or not compatible with the system are unlikely to benefit directly.
Why it matters
Potential benefit: If successful, this could help catch preventable clinical errors earlier and reduce harm from missed or incorrect treatments.
How similar studies have performed: Computerized alerts have sometimes reduced errors but also caused alert fatigue, and this outlier‑based, AI-plus-human‑knowledge approach is newer and not yet widely proven.
Where this research is happening
Pittsburgh, United States
- University of Pittsburgh at Pittsburgh — Pittsburgh, United States (Active)
Researchers
- Principal investigator: Hauskrecht, Milos — University of Pittsburgh at Pittsburgh
- Study coordinator: Hauskrecht, Milos
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 below.