Clinical predictive modelling of post-surgical recovery in individuals with cervical radiculopathy: a machine learning approach

Research output: Contribution to journalArticlepeer-review

Standard

Clinical predictive modelling of post-surgical recovery in individuals with cervical radiculopathy : a machine learning approach. / Liew, Bernard X W; Peolsson, Anneli; Rugamer, David; Wibault, Johanna; Löfgren, Hakan; Dedering, Asa; Zsigmond, Peter; Falla, Deborah.

In: Scientific Reports, Vol. 10, No. 1, 16782, 08.10.2020, p. 16782.

Research output: Contribution to journalArticlepeer-review

Harvard

Liew, BXW, Peolsson, A, Rugamer, D, Wibault, J, Löfgren, H, Dedering, A, Zsigmond, P & Falla, D 2020, 'Clinical predictive modelling of post-surgical recovery in individuals with cervical radiculopathy: a machine learning approach', Scientific Reports, vol. 10, no. 1, 16782, pp. 16782. https://doi.org/10.1038/s41598-020-73740-7

APA

Vancouver

Author

Liew, Bernard X W ; Peolsson, Anneli ; Rugamer, David ; Wibault, Johanna ; Löfgren, Hakan ; Dedering, Asa ; Zsigmond, Peter ; Falla, Deborah. / Clinical predictive modelling of post-surgical recovery in individuals with cervical radiculopathy : a machine learning approach. In: Scientific Reports. 2020 ; Vol. 10, No. 1. pp. 16782.

Bibtex

@article{00bebbe215d041c8bd34f641fbb75e66,
title = "Clinical predictive modelling of post-surgical recovery in individuals with cervical radiculopathy: a machine learning approach",
abstract = "Prognostic models play an important role in the clinical management of cervical radiculopathy (CR). No study has compared the performance of modern machine learning techniques, against more traditional stepwise regression techniques, when developing prognostic models in individuals with CR. We analysed a prospective cohort dataset of 201 individuals with CR. Four modelling techniques (stepwise regression, least absolute shrinkage and selection operator [LASSO], boosting, and multivariate adaptive regression splines [MuARS]) were each used to form a prognostic model for each of four outcomes obtained at a 12 month follow-up (disability-neck disability index [NDI]), quality of life (EQ5D), present neck pain intensity, and present arm pain intensity). For all four outcomes, the differences in mean performance between all four models were small (difference of NDI < 1 point; EQ5D < 0.1 point; neck and arm pain < 2 points). Given that the predictive accuracy of all four modelling methods were clinically similar, the optimal modelling method may be selected based on the parsimony of predictors. Some of the most parsimonious models were achieved using MuARS, a non-linear technique. Modern machine learning methods may be used to probe relationships along different regions of the predictor space.",
author = "Liew, {Bernard X W} and Anneli Peolsson and David Rugamer and Johanna Wibault and Hakan L{\"o}fgren and Asa Dedering and Peter Zsigmond and Deborah Falla",
year = "2020",
month = oct,
day = "8",
doi = "10.1038/s41598-020-73740-7",
language = "English",
volume = "10",
pages = "16782",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

RIS

TY - JOUR

T1 - Clinical predictive modelling of post-surgical recovery in individuals with cervical radiculopathy

T2 - a machine learning approach

AU - Liew, Bernard X W

AU - Peolsson, Anneli

AU - Rugamer, David

AU - Wibault, Johanna

AU - Löfgren, Hakan

AU - Dedering, Asa

AU - Zsigmond, Peter

AU - Falla, Deborah

PY - 2020/10/8

Y1 - 2020/10/8

N2 - Prognostic models play an important role in the clinical management of cervical radiculopathy (CR). No study has compared the performance of modern machine learning techniques, against more traditional stepwise regression techniques, when developing prognostic models in individuals with CR. We analysed a prospective cohort dataset of 201 individuals with CR. Four modelling techniques (stepwise regression, least absolute shrinkage and selection operator [LASSO], boosting, and multivariate adaptive regression splines [MuARS]) were each used to form a prognostic model for each of four outcomes obtained at a 12 month follow-up (disability-neck disability index [NDI]), quality of life (EQ5D), present neck pain intensity, and present arm pain intensity). For all four outcomes, the differences in mean performance between all four models were small (difference of NDI < 1 point; EQ5D < 0.1 point; neck and arm pain < 2 points). Given that the predictive accuracy of all four modelling methods were clinically similar, the optimal modelling method may be selected based on the parsimony of predictors. Some of the most parsimonious models were achieved using MuARS, a non-linear technique. Modern machine learning methods may be used to probe relationships along different regions of the predictor space.

AB - Prognostic models play an important role in the clinical management of cervical radiculopathy (CR). No study has compared the performance of modern machine learning techniques, against more traditional stepwise regression techniques, when developing prognostic models in individuals with CR. We analysed a prospective cohort dataset of 201 individuals with CR. Four modelling techniques (stepwise regression, least absolute shrinkage and selection operator [LASSO], boosting, and multivariate adaptive regression splines [MuARS]) were each used to form a prognostic model for each of four outcomes obtained at a 12 month follow-up (disability-neck disability index [NDI]), quality of life (EQ5D), present neck pain intensity, and present arm pain intensity). For all four outcomes, the differences in mean performance between all four models were small (difference of NDI < 1 point; EQ5D < 0.1 point; neck and arm pain < 2 points). Given that the predictive accuracy of all four modelling methods were clinically similar, the optimal modelling method may be selected based on the parsimony of predictors. Some of the most parsimonious models were achieved using MuARS, a non-linear technique. Modern machine learning methods may be used to probe relationships along different regions of the predictor space.

UR - http://www.scopus.com/inward/record.url?scp=85092286131&partnerID=8YFLogxK

U2 - 10.1038/s41598-020-73740-7

DO - 10.1038/s41598-020-73740-7

M3 - Article

C2 - 33033308

VL - 10

SP - 16782

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 16782

ER -