logo
ResearchBunny Logo
Artificial intelligence sepsis prediction algorithm learns to say “I don’t know”

Medicine and Health

Artificial intelligence sepsis prediction algorithm learns to say “I don’t know”

S. P. Shashikumar, G. Wardi, et al.

Discover COMPOSER, a groundbreaking deep learning model developed by Supreeth P. Shashikumar, Gabriel Wardi, Atul Malhotra, and Shamim Nemati for early sepsis prediction. This innovative approach minimizes false alarms and offers timely warnings for critical patient situations, potentially saving lives.

00:00
00:00
Playback language: English
Abstract
Sepsis is a leading cause of morbidity and mortality. This paper introduces COMPOSER, a deep learning model for early sepsis prediction designed to minimize false alarms by identifying unfamiliar patient situations. Using six patient cohorts (515,720 patients) from two US healthcare systems, COMPOSER achieved high AUC (ICU: 0.925–0.953; ED: 0.938–0.945) and provided early warnings within a clinically actionable timeframe. Approximately 20% and 8% of prediction windows were flagged as indeterminate for non-septic and septic patients, respectively.
Publisher
npj Digital Medicine
Published On
Sep 09, 2021
Authors
Supreeth P. Shashikumar, Gabriel Wardi, Atul Malhotra, Shamim Nemati
Tags
sepsis
deep learning
early prediction
clinical healthcare
false alarms
patient monitoring

Related Publications

Explore these studies to deepen your understanding of the subject.

Freud and the algorithm: neuropsychoanalysis as a framework to understand artificial general intelligence
Computer Science

Freud and the algorithm: neuropsychoanalysis as a framework to understand artificial general intelligence

L. M. Possati

Artificial intelligence in sepsis early prediction and diagnosis using unstructured data in healthcare
Medicine and Health

Artificial intelligence in sepsis early prediction and diagnosis using unstructured data in healthcare

K. H. Goh, L. Wang, et al.

A multi-institutional study using artificial intelligence to provide reliable and fair feedback to surgeons
Medicine and Health

A multi-institutional study using artificial intelligence to provide reliable and fair feedback to surgeons

D. Kiyasseh, J. Laca, et al.

Artificial intelligence achieves easy-to-adapt nonlinear global temperature reconstructions using minimal local data
Earth Sciences

Artificial intelligence achieves easy-to-adapt nonlinear global temperature reconstructions using minimal local data

M. Wegmann and F. Jaume-santero

Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny