Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patients

Introduction In stroke patients, accurate infarct volume assessment at 24 hours typically requires manual segmentation of lesions that often are not well defined. We aimed to validate an automated machine learning algorithm (MethinksFIV) specifically trained to automatically segment and measure suba...

Full description

Saved in:
Bibliographic Details
Main Authors: Magda Jablonska, João André Sousa, Marc Molina, Pere Canals, Alvaro Garcia‐Tornel, Marc Rodrigo‐Gisbert, Federica Rizzo, Marta Olivé‐Gadea, Manuel Requena, David Rodriguez‐Luna, Noelia Rodriguez‐Villatoro, Jesús M. Juega, Marián Muchada, Jorge Pagola, Marta Rubiera, Alejandro Tomasello, Carlos A. Molina, Leonardo Tanzi, Victor Salvia, Cristian Marti, Marc Ribo
Format: Article
Language:English
Published: Wiley 2023-11-01
Series:Stroke: Vascular and Interventional Neurology
Online Access:https://www.ahajournals.org/doi/10.1161/SVIN.03.suppl_2.268
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849714771678461952
author Magda Jablonska
João André Sousa
Marc Molina
Pere Canals
Alvaro Garcia‐Tornel
Marc Rodrigo‐Gisbert
Federica Rizzo
Marta Olivé‐Gadea
Manuel Requena
David Rodriguez‐Luna
Noelia Rodriguez‐Villatoro
Jesús M. Juega
Marián Muchada
Jorge Pagola
Marta Rubiera
Alejandro Tomasello
Carlos A. Molina
Leonardo Tanzi
Victor Salvia
Cristian Marti
Marc Ribo
author_facet Magda Jablonska
João André Sousa
Marc Molina
Pere Canals
Alvaro Garcia‐Tornel
Marc Rodrigo‐Gisbert
Federica Rizzo
Marta Olivé‐Gadea
Manuel Requena
David Rodriguez‐Luna
Noelia Rodriguez‐Villatoro
Jesús M. Juega
Marián Muchada
Jorge Pagola
Marta Rubiera
Alejandro Tomasello
Carlos A. Molina
Leonardo Tanzi
Victor Salvia
Cristian Marti
Marc Ribo
author_sort Magda Jablonska
collection DOAJ
description Introduction In stroke patients, accurate infarct volume assessment at 24 hours typically requires manual segmentation of lesions that often are not well defined. We aimed to validate an automated machine learning algorithm (MethinksFIV) specifically trained to automatically segment and measure subacute infarcts on non‐contrast CT (NCCT) scans. Methods We retrospectively studied stroke patients with a large vessel occlusion prospectively admitted to a comprehensive stroke center. The final infarct volume was segmented on the 24h NCCT manually (man‐FIV) and with the automated method (auto‐FIV). Both measurements were correlated with clinical outcomes. Results We included 346 patients. On 24h‐NCCT median auto‐FIV was lower than man‐FIV (8.6 ml [2.3‐40.7] vs. 16.9 ml [1.1‐70.8], p<0.001). Auto‐FIV and man‐FIV were highly correlated (r=0.8; p<0.001; mean dice coefficient of 0.61). The Auto‐FIV correlated better than man‐FIV with 24h (r=0.54 vs. r=0.51, both p<0.001) and discharge NIHSS (r=0.59 vs r=0.54, both p<0.001). An increase of 1 ml in auto‐FIV was associated with an increase in the odds of higher ordinal mRS at three months (OR=1.01 95%CI 1.01‐1.02, p<0.001). Auto‐FIV was a better predictor of mRS>2 at three months (OR: 1.02 95%CI 1.01‐1.13, p=0.002) than manual segmentation (OR=1.01, 95%CI 1.00‐1.01, p=0.01). In a multivariable analysis, only auto‐FIV remained a significant predictor of mRS>2 at three months (aOR: 1.03 95% CI 1.01‐1.05, p=0.004). Conclusion An automated method to measure subacute infarct volumes in stroke patients is accurate and correlates well with further clinical outcomes. The method appears to outperform manual segmentation and may be used to facilitate infarct characterization in large clinical trials.
format Article
id doaj-art-e29d6ea02a7e490fafc6d62f95e122cb
institution DOAJ
issn 2694-5746
language English
publishDate 2023-11-01
publisher Wiley
record_format Article
series Stroke: Vascular and Interventional Neurology
spelling doaj-art-e29d6ea02a7e490fafc6d62f95e122cb2025-08-20T03:13:36ZengWileyStroke: Vascular and Interventional Neurology2694-57462023-11-013S210.1161/SVIN.03.suppl_2.268Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patientsMagda Jablonska0João André Sousa1Marc Molina2Pere Canals3Alvaro Garcia‐Tornel4Marc Rodrigo‐Gisbert5Federica Rizzo6Marta Olivé‐Gadea7Manuel Requena8David Rodriguez‐Luna9Noelia Rodriguez‐Villatoro10Jesús M. Juega11Marián Muchada12Jorge Pagola13Marta Rubiera14Alejandro Tomasello15Carlos A. Molina16Leonardo Tanzi17Victor Salvia18Cristian Marti19Marc Ribo20Stroke Unit Vall d’Hebron University Hospital Barcelona SpainNeurology Department, Centro Hospitalar e Universitario de Coimbra Coimbra PortugalStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainNeurointerventional Radiology Department Vall d’Hebron University Hospital Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainMethinks Software Barcelona SpainMethinks Software Barcelona SpainMethinks Software Barcelona SpainStroke Unit Vall d’Hebron University Hospital Barcelona SpainIntroduction In stroke patients, accurate infarct volume assessment at 24 hours typically requires manual segmentation of lesions that often are not well defined. We aimed to validate an automated machine learning algorithm (MethinksFIV) specifically trained to automatically segment and measure subacute infarcts on non‐contrast CT (NCCT) scans. Methods We retrospectively studied stroke patients with a large vessel occlusion prospectively admitted to a comprehensive stroke center. The final infarct volume was segmented on the 24h NCCT manually (man‐FIV) and with the automated method (auto‐FIV). Both measurements were correlated with clinical outcomes. Results We included 346 patients. On 24h‐NCCT median auto‐FIV was lower than man‐FIV (8.6 ml [2.3‐40.7] vs. 16.9 ml [1.1‐70.8], p<0.001). Auto‐FIV and man‐FIV were highly correlated (r=0.8; p<0.001; mean dice coefficient of 0.61). The Auto‐FIV correlated better than man‐FIV with 24h (r=0.54 vs. r=0.51, both p<0.001) and discharge NIHSS (r=0.59 vs r=0.54, both p<0.001). An increase of 1 ml in auto‐FIV was associated with an increase in the odds of higher ordinal mRS at three months (OR=1.01 95%CI 1.01‐1.02, p<0.001). Auto‐FIV was a better predictor of mRS>2 at three months (OR: 1.02 95%CI 1.01‐1.13, p=0.002) than manual segmentation (OR=1.01, 95%CI 1.00‐1.01, p=0.01). In a multivariable analysis, only auto‐FIV remained a significant predictor of mRS>2 at three months (aOR: 1.03 95% CI 1.01‐1.05, p=0.004). Conclusion An automated method to measure subacute infarct volumes in stroke patients is accurate and correlates well with further clinical outcomes. The method appears to outperform manual segmentation and may be used to facilitate infarct characterization in large clinical trials.https://www.ahajournals.org/doi/10.1161/SVIN.03.suppl_2.268
spellingShingle Magda Jablonska
João André Sousa
Marc Molina
Pere Canals
Alvaro Garcia‐Tornel
Marc Rodrigo‐Gisbert
Federica Rizzo
Marta Olivé‐Gadea
Manuel Requena
David Rodriguez‐Luna
Noelia Rodriguez‐Villatoro
Jesús M. Juega
Marián Muchada
Jorge Pagola
Marta Rubiera
Alejandro Tomasello
Carlos A. Molina
Leonardo Tanzi
Victor Salvia
Cristian Marti
Marc Ribo
Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patients
Stroke: Vascular and Interventional Neurology
title Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patients
title_full Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patients
title_fullStr Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patients
title_full_unstemmed Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patients
title_short Abstract 268: Validation of an automated method to assess 24‐hour infarct volume in stroke patients
title_sort abstract 268 validation of an automated method to assess 24 hour infarct volume in stroke patients
url https://www.ahajournals.org/doi/10.1161/SVIN.03.suppl_2.268
work_keys_str_mv AT magdajablonska abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT joaoandresousa abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT marcmolina abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT perecanals abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT alvarogarciatornel abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT marcrodrigogisbert abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT federicarizzo abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT martaolivegadea abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT manuelrequena abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT davidrodriguezluna abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT noeliarodriguezvillatoro abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT jesusmjuega abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT marianmuchada abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT jorgepagola abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT martarubiera abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT alejandrotomasello abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT carlosamolina abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT leonardotanzi abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT victorsalvia abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT cristianmarti abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients
AT marcribo abstract268validationofanautomatedmethodtoassess24hourinfarctvolumeinstrokepatients