Gabriel Wardi
University of California-San Diego, CA·Medicine & Clinical Research·Early-career
Sepsis, the body's runaway response to infection, kills about 250,000 people a year in the United States. Almost every new therapy tested has failed to help patients live longer.
- Fields
Medicine & Clinical Research
Computer Science
Public Health & Epidemiology
- Topics
sepsis
machine learning
deep learning
- Grants
- 1 known grant, through Apr 2027
What they work on
- Identify clinical phenotypes, or distinct patient subgroups, of sepsis that develops after hospital admission, using deep-learning analysis of electronic health record data.
- Test whether sepsis phenotypes reflect disease paths that accepted therapies can change, such as the timing and amount of fluid resuscitation, and whether new inpatient phenotypes exist.
- Determine why sepsis patients are readmitted within 30 days by clustering patients at discharge and using natural language processing to mine clinical notes for actionable reasons.
- Test whether a wearable patch combined with a machine-learning algorithm reduces unexpected 30-day sepsis readmissions among patients discharged to long-term acute care hospitals.
More about the research
One reason may be that sepsis care uses "one-size-fits-all" treatment bundles, which can harm some subgroups of patients. This project asks whether artificial intelligence (AI) applied to electronic health record data can sort sepsis patients into meaningful groups and guide more personalized care. The work spans the hospital stay through the period after discharge.
Why it matters
Sepsis causes roughly 250,000 U.S. deaths each year, and few new treatments have improved survival. Algorithms that recognize distinct sepsis subgroups could help clinicians tailor treatments such as fluid resuscitation and avoid harming patients in some subgroups. Remote monitoring after discharge may also catch patients at risk of returning to the hospital. Patients who develop sepsis during a hospital stay, and those discharged to long-term care, could benefit most.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Matthew Churpek
University of Wisconsin-Madison, WI·Medicine & Clinical Research
Improve early identification and risk stratification of infected hospital patients by applying natural language processing and deep learning to structured and free-text electronic health record data from multiple centers.
Shamim Nemati
University of California-San Diego, CA·Biomedical Engineering
Compare a large language model–supported differential diagnosis system against the binary sepsis model COMPOSER for accuracy, with expert clinician review of the diagnosis lists, using retrospective hospital data.
Romain Pirracchio
University of California-San Francisco, CA·Medicine & Clinical Research
Train machine learning models on electronic health record data from three health systems to estimate each patient's individual treatment effect from antibiotics, fluid resuscitation and vasopressors in suspected community-onset lung sepsis.
The school
University of California-San Diego: 84 labs on PI Pages, $1.9B a year on research, 637 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.