Identifying different types of sepsis linked to antibiotic-resistant infections using advanced technology.
Official title: Identifying Sepsis Phenotypes Associated with Antibiotic-Resistant Pathogens Using Large Language Models and Machine Learning
Funded at Massachusetts General Hospital · NIH-10948425 · Data updated 2026-11-12
This study is looking to help doctors treat sepsis better by using advanced technology to find out which types of sepsis are caused by germs that don't respond to common antibiotics, so they can give the right medicine to patients and avoid using antibiotics that aren't needed.
Quick facts
| Grant type | Career grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Massachusetts General Hospital NIH-funded |
| Lab location | 1 site (Boston, United States) |
| Project start | August 1, 2024 |
| Project end | July 31, 2029 |
| Funding year | 2024 |
| Award (this year) | $147,122 |
| NIH project number | 1K08HS030118-01 |
| Project ID | NIH-10948425 on NIH RePORTER |
Who could benefit from this research
✓Could help
Ideal candidates for this research are patients diagnosed with sepsis or suspected of having sepsis who are receiving or may require antibiotic treatment.
✕Less likely to help
Patients with sepsis caused by non-resistant pathogens or those who do not require antibiotic therapy may not benefit from this research.
What this research studies
This research aims to improve the treatment of sepsis by using large language models and machine learning to identify specific phenotypes of sepsis associated with antibiotic-resistant pathogens. By analyzing detailed electronic health record data, including clinical notes and patient histories, the study seeks to develop more accurate models that can help clinicians determine when to use narrow-spectrum antibiotics instead of broad-spectrum ones. This could lead to more targeted treatments, reducing unnecessary antibiotic use and the associated risks of resistance and side effects. The research will involve collecting and analyzing data from patients with suspected sepsis to enhance understanding of their conditions.
Why it matters
Potential benefit: If successful, this research could lead to more effective and safer antibiotic treatments for patients with sepsis, minimizing the risks of antibiotic resistance and adverse effects.
How similar studies have gone: Previous studies have shown promise in using clustering methods to understand sepsis subtypes, but this approach of integrating clinical notes with machine learning is relatively novel.
Where this research is happening
-
Massachusetts General Hospital
Boston, United States Active
Researchers
- Principal investigator: Pak, Theodore Robertson — Massachusetts General Hospital
- Study coordinator: Pak, Theodore Robertson
Related research
Other NIH-funded projects on the same topics.
-
Role of blood flow mechanics and transfusion in sepsis-related tiny-cl vessel clots
A patient-focused look at how red blood cell flow and transfusions might influence tiny blood vessel clots and organ problems in sepsis.
-
STeroids and enhanced Antibiotics for sepsis in Africa with HIV/TB (STARS)
This grant tests whether adding steroids and stronger antibiotics helps adults with severe sepsis tied to HIV/TB in areas with high TB rates.
-
Steroids and better antibiotics for sepsis in people with AIDS in Africa
This study tests whether adding hydrocortisone and stronger antibiotics can improve survival for adults with sepsis who also have AIDS or advanced immune suppression in regions with high TB.
-
Coordinating ARDS, Pneumonia, and Sepsis research to improve care
A national coordinating center helps researchers studying ARDS and related infections work together more efficiently to learn what helps patients recover.
-
Role of MMP3 in heart dysfunction during sepsis
This research tests whether blocking MMP3 can protect the heart and improve survival during sepsis in adults.
-
Sepsis patient care guided by Ugandan-wide immune profiling
A large NIH-funded study will map how the immune system behaves during sepsis in people in Uganda to improve personalized care for sepsis patients.
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.