Authors: Martha Powell, Future Science Group
A computer-aided model that uses routinely collected data could identify the early symptoms of sepsis therefore allowing rapid diagnosis, according to new research.
Early detection of sepsis is key to preventing mortality, each hour diagnosis is delayed has been linked to a 7% reduction in survival for patients.
In the UK the National Early Warning Score (NEWS) is used in NHS hospitals to help identify patients with sepsis. This uses routinely collected vital signs to identify possible sepsis cases, prompting researchers to build on this system by developing a computer-aided NEWS (cNEWS) to determine if this could enhance to accuracy of predicting sepsis.
The team developed three models for cNEWS, evaluating each and identifying the model with the highest accuracy.