King's College London

Research portal

A systematic review of machine learning models for predicting outcomes of stroke with structured data

Research output: Contribution to journalReview article

Standard

A systematic review of machine learning models for predicting outcomes of stroke with structured data. / Wang, Wenjuan; Kiik, Martin ; Peek, Niels; Curcin, Vasa; Marshall, Iain; Rudd, Anthony; Wang, Yanzhong; Douiri, Abdel; Wolfe, Charles; Bray, Benjamin.

In: PLOS One, Vol. 15, No. 6, e0234722, 12.06.2020, p. e0234722.

Research output: Contribution to journalReview article

Harvard

Wang, W, Kiik, M, Peek, N, Curcin, V, Marshall, I, Rudd, A, Wang, Y, Douiri, A, Wolfe, C & Bray, B 2020, 'A systematic review of machine learning models for predicting outcomes of stroke with structured data', PLOS One, vol. 15, no. 6, e0234722, pp. e0234722. https://doi.org/10.1371/journal.pone.0234722

APA

Wang, W., Kiik, M., Peek, N., Curcin, V., Marshall, I., Rudd, A., ... Bray, B. (2020). A systematic review of machine learning models for predicting outcomes of stroke with structured data. PLOS One, 15(6), e0234722. [e0234722]. https://doi.org/10.1371/journal.pone.0234722

Vancouver

Wang W, Kiik M, Peek N, Curcin V, Marshall I, Rudd A et al. A systematic review of machine learning models for predicting outcomes of stroke with structured data. PLOS One. 2020 Jun 12;15(6):e0234722. e0234722. https://doi.org/10.1371/journal.pone.0234722

Author

Wang, Wenjuan ; Kiik, Martin ; Peek, Niels ; Curcin, Vasa ; Marshall, Iain ; Rudd, Anthony ; Wang, Yanzhong ; Douiri, Abdel ; Wolfe, Charles ; Bray, Benjamin. / A systematic review of machine learning models for predicting outcomes of stroke with structured data. In: PLOS One. 2020 ; Vol. 15, No. 6. pp. e0234722.

Bibtex Download

@article{039cdcf8fb494ce8afead862077efa97,
title = "A systematic review of machine learning models for predicting outcomes of stroke with structured data",
abstract = "Background and Purpose Machine learning (ML) has attracted much attention with the hope that it could make use of large, routinely collected datasets and deliver accurate personalised prognosis. The aim of this systematic review is to identify and critically appraise the reporting and developing of ML models for predicting outcomes after stroke.Methods We searched PubMed and Web of Science from 1990 to March 2019, using previously published search filters for stroke, ML, and prediction models. We focused on structured clinical data, excluding image and text analysis. This review was registered with PROSPERO (CRD42019127154).Results Eighteen studies were eligible for inclusion. Most studies reported less than half of the terms in the reporting quality checklist. The most frequently predicted stroke outcomes were mortality (7 studies) and functional outcome (5 studies). The most commonly used ML methods were random forests (9 studies), support vector machines (8 studies), decision trees (6 studies), and neural networks (6 studies). The median sample size was 475 (range 70–3184), with a median of 22 predictors (range 4–152) considered. All studies evaluated discrimination with thirteen using area under the ROC curve whilst calibration was assessed in three. Two studies performed external validation. None described the final model sufficiently well to reproduce it.Conclusions The use of ML for predicting stroke outcomes is increasing. However, few met basic reporting standards for clinical prediction tools and none made their models available in a way which could be used or evaluated. Major improvements in ML study conduct and reporting are needed before it can meaningfully be considered for practice.",
author = "Wenjuan Wang and Martin Kiik and Niels Peek and Vasa Curcin and Iain Marshall and Anthony Rudd and Yanzhong Wang and Abdel Douiri and Charles Wolfe and Benjamin Bray",
year = "2020",
month = "6",
day = "12",
doi = "https://doi.org/10.1371/journal.pone.0234722",
language = "English",
volume = "15",
pages = "e0234722",
journal = "PLOS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "6",

}

RIS (suitable for import to EndNote) Download

TY - JOUR

T1 - A systematic review of machine learning models for predicting outcomes of stroke with structured data

AU - Wang, Wenjuan

AU - Kiik, Martin

AU - Peek, Niels

AU - Curcin, Vasa

AU - Marshall, Iain

AU - Rudd, Anthony

AU - Wang, Yanzhong

AU - Douiri, Abdel

AU - Wolfe, Charles

AU - Bray, Benjamin

PY - 2020/6/12

Y1 - 2020/6/12

N2 - Background and Purpose Machine learning (ML) has attracted much attention with the hope that it could make use of large, routinely collected datasets and deliver accurate personalised prognosis. The aim of this systematic review is to identify and critically appraise the reporting and developing of ML models for predicting outcomes after stroke.Methods We searched PubMed and Web of Science from 1990 to March 2019, using previously published search filters for stroke, ML, and prediction models. We focused on structured clinical data, excluding image and text analysis. This review was registered with PROSPERO (CRD42019127154).Results Eighteen studies were eligible for inclusion. Most studies reported less than half of the terms in the reporting quality checklist. The most frequently predicted stroke outcomes were mortality (7 studies) and functional outcome (5 studies). The most commonly used ML methods were random forests (9 studies), support vector machines (8 studies), decision trees (6 studies), and neural networks (6 studies). The median sample size was 475 (range 70–3184), with a median of 22 predictors (range 4–152) considered. All studies evaluated discrimination with thirteen using area under the ROC curve whilst calibration was assessed in three. Two studies performed external validation. None described the final model sufficiently well to reproduce it.Conclusions The use of ML for predicting stroke outcomes is increasing. However, few met basic reporting standards for clinical prediction tools and none made their models available in a way which could be used or evaluated. Major improvements in ML study conduct and reporting are needed before it can meaningfully be considered for practice.

AB - Background and Purpose Machine learning (ML) has attracted much attention with the hope that it could make use of large, routinely collected datasets and deliver accurate personalised prognosis. The aim of this systematic review is to identify and critically appraise the reporting and developing of ML models for predicting outcomes after stroke.Methods We searched PubMed and Web of Science from 1990 to March 2019, using previously published search filters for stroke, ML, and prediction models. We focused on structured clinical data, excluding image and text analysis. This review was registered with PROSPERO (CRD42019127154).Results Eighteen studies were eligible for inclusion. Most studies reported less than half of the terms in the reporting quality checklist. The most frequently predicted stroke outcomes were mortality (7 studies) and functional outcome (5 studies). The most commonly used ML methods were random forests (9 studies), support vector machines (8 studies), decision trees (6 studies), and neural networks (6 studies). The median sample size was 475 (range 70–3184), with a median of 22 predictors (range 4–152) considered. All studies evaluated discrimination with thirteen using area under the ROC curve whilst calibration was assessed in three. Two studies performed external validation. None described the final model sufficiently well to reproduce it.Conclusions The use of ML for predicting stroke outcomes is increasing. However, few met basic reporting standards for clinical prediction tools and none made their models available in a way which could be used or evaluated. Major improvements in ML study conduct and reporting are needed before it can meaningfully be considered for practice.

UR - http://www.scopus.com/inward/record.url?scp=85086524908&partnerID=8YFLogxK

U2 - https://doi.org/10.1371/journal.pone.0234722

DO - https://doi.org/10.1371/journal.pone.0234722

M3 - Review article

VL - 15

SP - e0234722

JO - PLOS One

JF - PLOS One

SN - 1932-6203

IS - 6

M1 - e0234722

ER -

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454