Mortality prediction in ICU

a methodological advance

John Norrie

Research output: Contribution to journalComment/debate

3 Citations (Scopus)

Abstract

In The Lancet Respiratory Medicine, Pirracchio and colleagues present a new approach to predicting mortality in intensive care units (ICUs). The investigators propose that instead of picking one of the many mortality prediction models available, an ensemble machine learning approach can be used (the non-parametric Super Learner) to leverage the individual candidate models from a pre-specified library, to produce an optimum prediction algorithm. This is an elegant idea that frees the user from making an arbitrary choice of model, and that also guarantees at least as good performance as any individual model within that library.
Original languageEnglish
Pages (from-to)5-6
Number of pages2
JournalThe Lancet. Respiratory medicine
Volume3
Issue number1
Early online date24 Nov 2014
DOIs
Publication statusPublished - Jan 2015

Fingerprint

Libraries
Intensive Care Units
Pulmonary Medicine
Mortality
Research Personnel
Machine Learning

Keywords

  • Mortality prediction
  • ICU

Cite this

Mortality prediction in ICU : a methodological advance. / Norrie, John.

In: The Lancet. Respiratory medicine, Vol. 3, No. 1, 01.2015, p. 5-6.

Research output: Contribution to journalComment/debate

@article{1497c70c96404c8abda11a7ccdef42dc,
title = "Mortality prediction in ICU: a methodological advance",
abstract = "In The Lancet Respiratory Medicine, Pirracchio and colleagues present a new approach to predicting mortality in intensive care units (ICUs). The investigators propose that instead of picking one of the many mortality prediction models available, an ensemble machine learning approach can be used (the non-parametric Super Learner) to leverage the individual candidate models from a pre-specified library, to produce an optimum prediction algorithm. This is an elegant idea that frees the user from making an arbitrary choice of model, and that also guarantees at least as good performance as any individual model within that library.",
keywords = "Mortality prediction, ICU",
author = "John Norrie",
year = "2015",
month = "1",
doi = "10.1016/S2213-2600(14)70268-1",
language = "English",
volume = "3",
pages = "5--6",
journal = "The Lancet. Respiratory medicine",
issn = "2213-2600",
publisher = "Elsevier Limited",
number = "1",

}

TY - JOUR

T1 - Mortality prediction in ICU

T2 - a methodological advance

AU - Norrie, John

PY - 2015/1

Y1 - 2015/1

N2 - In The Lancet Respiratory Medicine, Pirracchio and colleagues present a new approach to predicting mortality in intensive care units (ICUs). The investigators propose that instead of picking one of the many mortality prediction models available, an ensemble machine learning approach can be used (the non-parametric Super Learner) to leverage the individual candidate models from a pre-specified library, to produce an optimum prediction algorithm. This is an elegant idea that frees the user from making an arbitrary choice of model, and that also guarantees at least as good performance as any individual model within that library.

AB - In The Lancet Respiratory Medicine, Pirracchio and colleagues present a new approach to predicting mortality in intensive care units (ICUs). The investigators propose that instead of picking one of the many mortality prediction models available, an ensemble machine learning approach can be used (the non-parametric Super Learner) to leverage the individual candidate models from a pre-specified library, to produce an optimum prediction algorithm. This is an elegant idea that frees the user from making an arbitrary choice of model, and that also guarantees at least as good performance as any individual model within that library.

KW - Mortality prediction

KW - ICU

U2 - 10.1016/S2213-2600(14)70268-1

DO - 10.1016/S2213-2600(14)70268-1

M3 - Comment/debate

VL - 3

SP - 5

EP - 6

JO - The Lancet. Respiratory medicine

JF - The Lancet. Respiratory medicine

SN - 2213-2600

IS - 1

ER -