Early prediction of sepsis in intensive care patients using the machine learning algorithm NAVOY® Sepsis, a prospective randomized clinical validation study
Problem
Sepsis in ICU patients is often detected too late, so clinicians need tools that predict its onset hours in advance.
Why it matters
Early sepsis detection in ICUs can mean the difference between life and death for critically ill patients.
What the AI did
The NAVOY® Sepsis algorithm uses up to 20 routinely collected vital, blood gas, and lab values to predict sepsis hours ahead; it was prospectively tested in a blinded randomized study of 304 ICU patients.
Results
not stated in source
Impact
A prospective, blinded, randomized validation in a real ICU was conducted, but the abstract is cut off before any accuracy results.
Record
| Venue | Journal of Critical Care |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Inger Persson, Andreas Macura, David Becedas, Fredrik Sjövall |
| Citations | 43 (OpenAlex, at import time) |
| References | 24 |
| Model or method | NAVOY® Sepsis |
| Organization | Skåne University Hospital Malmö |
| Primary source | https://doi.org/10.1016/j.jcrc.2023.154400 |
| Open access | hybrid |
| Keywords | sepsis, ICU, machine learning, prospective validation, early prediction, NAVOY |
Importance rating
significant — thousands to millions affected
Early sepsis prediction in ICUs affects life-and-death care; abstract states a prospective blinded randomized study of 304 ICU patients with sepsis predicted 3 hours ahead (accuracy 0.79, sensitivity 0.80).