Steps to avoid overuse and misuse of machine learning in clinical research

Victor Volovici; Nicholas L. Syn; Ari Ercole; Joseph J. Zhao; Nan Liu

doi:10.1038/s41591-022-01961-6

Steps to avoid overuse and misuse of machine learning in clinical research

Victor Volovici^*
, Nicholas L. Syn
, Ari Ercole
, Joseph J. Zhao
, Nan Liu

^*Corresponding author for this work

Neurosurgery

Yong Loo Lin School of Medicine
National University Hospital (Singapore)
University of Cambridge
Duke-NUS Graduate Medical School

Research output: Contribution to journal › Comment/Letter to the editor › Academic › peer-review

98 Citations (Scopus)

223 Downloads (Pure)

Abstract

Machine learning algorithms are a powerful tool in healthcare, but sometimes perform no better than traditional statistical techniques. Steps should be taken to ensure that algorithms are not overused or misused, in order to provide genuine benefit for patients.

Original language	English
Pages (from-to)	1996-1999
Number of pages	4
Journal	Nature Medicine
Volume	28
Issue number	10
Early online date	12 Sept 2022
DOIs	https://doi.org/10.1038/s41591-022-01961-6
Publication status	Published - Oct 2022

Bibliographical note

Funding Information:
We would like to thank M. van Bilsen for the figure and F. Liu for her valuable advice. V.V. wishes to thank D. Volovici for opening up the world of probability, statistics and machine learning.

Access to Document

10.1038/s41591-022-01961-6

s41591-022-01961-6Final published version, 914 KB

https://www.nature.com/articles/s41591-022-01961-6.pdf

Cite this

@article{6ed363252a7a4c6cab6bb3d7c71c1c1c,

title = "Steps to avoid overuse and misuse of machine learning in clinical research",

abstract = "Machine learning algorithms are a powerful tool in healthcare, but sometimes perform no better than traditional statistical techniques. Steps should be taken to ensure that algorithms are not overused or misused, in order to provide genuine benefit for patients.",

author = "Victor Volovici and Syn, \{Nicholas L.\} and Ari Ercole and Zhao, \{Joseph J.\} and Nan Liu",

note = "Funding Information: We would like to thank M. van Bilsen for the figure and F. Liu for her valuable advice. V.V. wishes to thank D. Volovici for opening up the world of probability, statistics and machine learning.",

year = "2022",

month = oct,

doi = "10.1038/s41591-022-01961-6",

language = "English",

volume = "28",

pages = "1996--1999",

journal = "Nature Medicine",

issn = "1078-8956",

publisher = "Nature Research",

number = "10",

}

TY - JOUR

T1 - Steps to avoid overuse and misuse of machine learning in clinical research

AU - Volovici, Victor

AU - Syn, Nicholas L.

AU - Ercole, Ari

AU - Zhao, Joseph J.

AU - Liu, Nan

N1 - Funding Information: We would like to thank M. van Bilsen for the figure and F. Liu for her valuable advice. V.V. wishes to thank D. Volovici for opening up the world of probability, statistics and machine learning.

PY - 2022/10

Y1 - 2022/10

N2 - Machine learning algorithms are a powerful tool in healthcare, but sometimes perform no better than traditional statistical techniques. Steps should be taken to ensure that algorithms are not overused or misused, in order to provide genuine benefit for patients.

AB - Machine learning algorithms are a powerful tool in healthcare, but sometimes perform no better than traditional statistical techniques. Steps should be taken to ensure that algorithms are not overused or misused, in order to provide genuine benefit for patients.

UR - https://www.scopus.com/pages/publications/85137938353

U2 - 10.1038/s41591-022-01961-6

DO - 10.1038/s41591-022-01961-6

M3 - Comment/Letter to the editor

C2 - 36097217

SN - 1078-8956

VL - 28

SP - 1996

EP - 1999

JO - Nature Medicine

JF - Nature Medicine

IS - 10

ER -

Steps to avoid overuse and misuse of machine learning in clinical research

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this