Diagnostic and prognostic capabilities of a biomarker and EMR‐based machine learning algorithm for sepsis
Problem
Early recognition of sepsis in emergency department patients, which is difficult but critical for timely antibiotic treatment.
Why it matters
Sepsis is a major cause of hospital mortality and faster treatment improves outcomes.
What the AI did
A machine-learning algorithm using clinical EMR data and uncommonly measured biomarkers outputs a score with diagnostic and prognostic capability for hospital length of stay, 30-day mortality, and 3-day readmission.
Results
AUROC of 0.83 for sepsis diagnosis in a two-center cohort of 1400 adult ED patients; prognostic power for LOS, 30-day mortality, and 3-day readmission.
Impact
Peer-reviewed two-center cohort study demonstrating diagnostic and prognostic performance, though the abstract is truncated before full results.
Record
| Venue | Clinical and Translational Science |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Ishan Taneja, Gregory L. Damhorst, Carlos G. Lopez-Espina, Sihai Dave Zhao, Ruoqing Zhu, Shah Alam Khan, Karen White, James S Kumar et al. |
| Citations | 44 (OpenAlex, at import time) |
| References | 35 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1111/cts.13030 |
| Open access | gold |
| Keywords | sepsis, machine learning, biomarkers, AUROC, prognosis, emergency department |
Importance rating
major — millions of lives, an industry or a field
Abstract states sepsis is a major cause of mortality among hospitalized patients worldwide; early recognition is critical for timely antibiotics.