Post

New research · Emergency Medicine
Journal of medical systems · 1d
AI / informaticsJournal of medical systems · 2026

Interpretable Machine Learning Model for Predicting Sepsis and Septic Shock Among Patients with Documented Fever at Emergency Department Triage Using Patients' Historical Data.

Seung Jin Maeng, Ye Rim Lee, Se Uk Lee … Sejin Heo
Read paper
Emergency MedicineAI / informatics

New machine learning model shows high accuracy predicting septic shock in febrile emergency department patients.

Interpretable Machine Learning Model for Predicting Sepsis and Septic Shock Among Patients with Documented Fever at Emergency Department Triage Using Patients' Historical Data.

Seung Jin Maeng et al. · Journal of medical systems · 2026
Purpose

This study aimed to develop an interpretable machine learning-based scoring system for predicting sepsis and septic shock among febrile patients at emergency department (ED) triage using longitudinal data.

Methods

This retrospective, single-center study included adult patients, presented to ED of tertiary academic hospital with fever from January 2016 to December 2021.

Results

how well the new model predicted septic shock

0.844
Our model
0.605
qSOFA
0.678
MEWS
More results

Using the AutoScore framework, we developed a novel scoring system for predicting sepsis and septic shock at the triage stage, incorporating nine variables and a maximum score of 29.

Our model incorporated nine variables including initial vital signs, age, baseline platelet count, total bilirubin, and creatinine levels.

More results

Compared to qSOFA ≥ 2 (AUROC: 0.605) and MEWS ≥ 5 (AUROC: 0.678), our scoring system demonstrated superior predictive performance for septic shock.

For comparable specificity levels (ranging from 0.50 to 0.95), our scoring system achieved higher sensitivity than MEWS.

“
Conclusion

Our scoring system is an interpretable and practical scoring tool for predicting sepsis and septic shock among patients with documented fever at ED triage using patient's longitudinal data.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
Cross-sectional
62%
most clinicians were interested in a research study for lower leg skin infection
Cohort Study
76%
76% of patients offered virtual care were successfully diverted from the physical emergency department.
AI / Informatics
odds ratio 2.37
Artificial intelligence assistance more than doubled the chances of correctly identifying a collapsed lung from injury
Cohort Study
IRR 0.21
Community Care Team Transitional Care Programme was associated with a lower 90-day death rate
Cohort Study
29.0 min
Patients during unscheduled drills had an average door-to-triage time of 29.0 minutes.
Guideline
5 days
new guidelines recommend less antibiotic treatment than this for nonsevere lung infection
AI / Informatics
79.4%
the model correctly identified the right diagnosis for brain and nerve emergencies