AI that finds sepsis early using medical records and images
Develop Multi-modal Foundation Models for Sepsis Early Detection
Researchers will build AI that learns from medical records, clinical notes, and medical images to spot early signs of sepsis in hospitalized patients.
Quick facts
| Grant type | NIH-funded research |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of California, San Diego NIH-funded |
| Lab location | 1 site (La Jolla, United States) |
| Project ID | NIH-11259423 on NIH RePORTER |
What this research studies
They will train large "foundation" AI models on multiple types of hospital data—electronic health records, clinical notes, and imaging—to recognize early sepsis patterns. The work uses advanced machine‑learning techniques like self‑supervised learning, multi‑modal modeling, and interpretable AI so clinicians can understand predictions. The team will address differences across hospitals and data formats to make the tool useful in varied care settings. If it works, the system would aim to flag at‑risk patients earlier so care teams can act sooner.
Who could benefit from this research
Good fit: Ideal participants are hospitalized patients whose EHRs, clinical notes, and medical images can be shared for research by their treating hospitals.
Not a fit: People without linked hospital records or imaging data, or those treated entirely outside participating hospitals, are unlikely to be directly involved or benefit.
Why it matters
Potential benefit: If successful, this could help clinicians detect sepsis sooner, reducing organ damage and deaths.
How similar studies have performed: Previous machine‑learning sepsis early‑warning tools have shown mixed but promising results, while large multi‑modal foundation models for sepsis remain a newer, less‑tested approach.
Where this research is happening
La Jolla, United States
- University of California, San Diego — La Jolla, United States (Active)
Researchers
- Principal investigator: Xie, Pengtao — University of California, San Diego
- Study coordinator: Xie, Pengtao
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.