Prediction of stroke severity: systematic evaluation of lesion representations.

Bonkhoff, AK; Cohen, AL; Drew, W; Ferguson, MA; Hussain, A; Lin, C; Schaper, FLWVJ; Bourached, A; Giese, A-K; Oliveira, LC; Regenhardt, RW; Schirmer, MD; Jern, C; Lindgren, AG; Maguire, J; Wu, O; Zafar, S; Rhee, JY; Kimchi, EY; Corbetta, M; Rost, NS; Fox, MD; MRI‐GENIE and GISCOME Investigators and the International Stroke Genetics Consortium,

Prediction of stroke severity: systematic evaluation of lesion representations.

Bonkhoff, AK Cohen, AL Drew, W Ferguson, MA Hussain, A Lin, C Schaper, FLWVJ Bourached, A Giese, A-K Oliveira, LC Regenhardt, RW Schirmer, MD Jern, C Lindgren, AG Maguire, J

Wu, O Zafar, S Rhee, JY Kimchi, EY Corbetta, M Rost, NS Fox, MD MRI‐GENIE and GISCOME Investigators and the International Stroke Genetics Consortium,

Permalink

Publisher:: Wiley
Publication Type:: Journal Article
Citation:: Ann Clin Transl Neurol, 2024
Issue Date:: 2024-10-11

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Published versionAdobe PDF (2.84 MB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Bonkhoff, AK
dc.contributor.author	Cohen, AL
dc.contributor.author	Drew, W
dc.contributor.author	Ferguson, MA
dc.contributor.author	Hussain, A
dc.contributor.author	Lin, C
dc.contributor.author	Schaper, FLWVJ
dc.contributor.author	Bourached, A
dc.contributor.author	Giese, A-K
dc.contributor.author	Oliveira, LC
dc.contributor.author	Regenhardt, RW
dc.contributor.author	Schirmer, MD
dc.contributor.author	Jern, C
dc.contributor.author	Lindgren, AG
dc.contributor.author	Maguire, J https://orcid.org/0000-0001-5722-8311
dc.contributor.author	Wu, O
dc.contributor.author	Zafar, S
dc.contributor.author	Rhee, JY
dc.contributor.author	Kimchi, EY
dc.contributor.author	Corbetta, M
dc.contributor.author	Rost, NS
dc.contributor.author	Fox, MD
dc.contributor.author	MRI‐GENIE and GISCOME Investigators and the International Stroke Genetics Consortium,
dc.date.accessioned	2024-10-30T01:00:59Z
dc.date.available	2024-09-08
dc.date.available	2024-10-30T01:00:59Z
dc.date.issued	2024-10-11
dc.identifier.citation	Ann Clin Transl Neurol, 2024
dc.identifier.issn	2328-9503
dc.identifier.issn	2328-9503
dc.identifier.uri	http://hdl.handle.net/10453/181617
dc.description.abstract	OBJECTIVE: To systematically evaluate which lesion-based imaging features and methods allow for the best statistical prediction of poststroke deficits across independent datasets. METHODS: We utilized imaging and clinical data from three independent datasets of patients experiencing acute stroke (N1 = 109, N2 = 638, N3 = 794) to statistically predict acute stroke severity (NIHSS) based on lesion volume, lesion location, and structural and functional disconnection with the lesion location using normative connectomes. RESULTS: We found that prediction models trained on small single-center datasets could perform well using within-dataset cross-validation, but results did not generalize to independent datasets (median R2 N1 = 0.2%). Performance across independent datasets improved using large single-center training data (R2 N2 = 15.8%) and improved further using multicenter training data (R2 N3 = 24.4%). These results were consistent across lesion attributes and prediction models. Including either structural or functional disconnection in the models outperformed prediction based on volume or location alone (P < 0.001, FDR-corrected). INTERPRETATION: We conclude that (1) prediction performance in independent datasets of patients with acute stroke cannot be inferred from cross-validated results within a dataset, as performance results obtained via these two methods differed consistently, (2) prediction performance can be improved by training on large and, importantly, multicenter datasets, and (3) structural and functional disconnection allow for improved prediction of acute stroke severity.
dc.format	Print-Electronic
dc.language	eng
dc.publisher	Wiley
dc.relation.ispartof	Ann Clin Transl Neurol
dc.relation.isbasedon	10.1002/acn3.52215
dc.rights	info:eu-repo/semantics/openAccess
dc.subject	1103 Clinical Sciences, 1109 Neurosciences
dc.subject.classification	3209 Neurosciences
dc.subject.classification	5203 Clinical and health psychology
dc.title	Prediction of stroke severity: systematic evaluation of lesion representations.
dc.type	Journal Article
utslib.location.activity	United States
utslib.for	1103 Clinical Sciences
utslib.for	1109 Neurosciences
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Health
pubs.organisational-group	University of Technology Sydney/Faculty of Health/School of Nursing and Midwifery
utslib.copyright.status	open_access	*
dc.rights.license	This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-nd/4.0/
dc.date.updated	2024-10-30T01:00:58Z
pubs.publication-status	Published online

Abstract:

OBJECTIVE: To systematically evaluate which lesion-based imaging features and methods allow for the best statistical prediction of poststroke deficits across independent datasets. METHODS: We utilized imaging and clinical data from three independent datasets of patients experiencing acute stroke (N1 = 109, N2 = 638, N3 = 794) to statistically predict acute stroke severity (NIHSS) based on lesion volume, lesion location, and structural and functional disconnection with the lesion location using normative connectomes. RESULTS: We found that prediction models trained on small single-center datasets could perform well using within-dataset cross-validation, but results did not generalize to independent datasets (median R2 N1 = 0.2%). Performance across independent datasets improved using large single-center training data (R2 N2 = 15.8%) and improved further using multicenter training data (R2 N3 = 24.4%). These results were consistent across lesion attributes and prediction models. Including either structural or functional disconnection in the models outperformed prediction based on volume or location alone (P < 0.001, FDR-corrected). INTERPRETATION: We conclude that (1) prediction performance in independent datasets of patients with acute stroke cannot be inferred from cross-validated results within a dataset, as performance results obtained via these two methods differed consistently, (2) prediction performance can be improved by training on large and, importantly, multicenter datasets, and (3) structural and functional disconnection allow for improved prediction of acute stroke severity.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/181617