Characterization of clot composition in acute cerebral infarct using machine learning techniques

Jong Won Chung, Yoon Chul Kim, Jihoon Cha, Eun Hyeok Choi, Byung Moon Kim, Woo Keun Seo, Gyeong Moon Kim, Oh Young Bang

Research output: Contribution to journalArticle

Abstract

Objective: Clot characteristics can provide information on the cause of cerebral artery occlusion and may guide acute revascularization and secondary prevention strategies. We developed a rapid automated clot analysis system using machine learning (ML) and validated its accuracy in patients undergoing endovascular treatment. Methods: Pre-endovascular treatment gradient echo (GRE) images from consecutive patients with middle cerebral artery occlusion were utilized to develop and validate an ML system to predict whether atrial fibrillation (AF) was the underlying cause of ischemic stroke. The accuracy of the ML algorithm was compared with that of visual inspection by neuroimaging specialists for the presence of blooming artifact. Endovascular procedures and outcomes were compared in patients with and without AF. Results: Of 67 patients, 29 (43.3%) had AF. Of these, 13 had known AF and 16 were newly diagnosed with cardiac monitoring. By visual inspection, interrater correlation for blooming artifact was 0.73 and sensitivity and specificity for AF were 0.79 and 0.63, respectively. For AF classification, the ML algorithms yielded an average accuracy of > 75.4% in fivefold cross-validation with clot signal profiles obtained from 52 patients and an area under the curve >0.87 for the average AF probability from five signal profiles in external validation (n = 15). Analysis with an in-house interface took approximately 3 min per patient. Absence of AF was associated with increased number of passes by stentriever, high reocclusion frequency, and additional use of rescue stenting and/or glycogen IIb/IIIa blocker for recanalization. Interpretation: ML-based rapid clot analysis is feasible and can identify AF with high accuracy, enabling selection of endovascular treatment strategy.

Original languageEnglish
Pages (from-to)739-747
Number of pages9
JournalAnnals of Clinical and Translational Neurology
Volume6
Issue number4
DOIs
Publication statusPublished - 2019 Apr

Fingerprint

Atrial Fibrillation
Artifacts
Machine Learning
Endovascular Procedures
Cerebral Arteries
Middle Cerebral Artery Infarction
Secondary Prevention
Glycogen
Neuroimaging
Area Under Curve
Therapeutics
Stroke
Sensitivity and Specificity

All Science Journal Classification (ASJC) codes

  • Neuroscience(all)
  • Clinical Neurology

Cite this

Chung, Jong Won ; Kim, Yoon Chul ; Cha, Jihoon ; Choi, Eun Hyeok ; Kim, Byung Moon ; Seo, Woo Keun ; Kim, Gyeong Moon ; Bang, Oh Young. / Characterization of clot composition in acute cerebral infarct using machine learning techniques. In: Annals of Clinical and Translational Neurology. 2019 ; Vol. 6, No. 4. pp. 739-747.
@article{67b8329dd015408cac48e6d54433ead1,
title = "Characterization of clot composition in acute cerebral infarct using machine learning techniques",
abstract = "Objective: Clot characteristics can provide information on the cause of cerebral artery occlusion and may guide acute revascularization and secondary prevention strategies. We developed a rapid automated clot analysis system using machine learning (ML) and validated its accuracy in patients undergoing endovascular treatment. Methods: Pre-endovascular treatment gradient echo (GRE) images from consecutive patients with middle cerebral artery occlusion were utilized to develop and validate an ML system to predict whether atrial fibrillation (AF) was the underlying cause of ischemic stroke. The accuracy of the ML algorithm was compared with that of visual inspection by neuroimaging specialists for the presence of blooming artifact. Endovascular procedures and outcomes were compared in patients with and without AF. Results: Of 67 patients, 29 (43.3{\%}) had AF. Of these, 13 had known AF and 16 were newly diagnosed with cardiac monitoring. By visual inspection, interrater correlation for blooming artifact was 0.73 and sensitivity and specificity for AF were 0.79 and 0.63, respectively. For AF classification, the ML algorithms yielded an average accuracy of > 75.4{\%} in fivefold cross-validation with clot signal profiles obtained from 52 patients and an area under the curve >0.87 for the average AF probability from five signal profiles in external validation (n = 15). Analysis with an in-house interface took approximately 3 min per patient. Absence of AF was associated with increased number of passes by stentriever, high reocclusion frequency, and additional use of rescue stenting and/or glycogen IIb/IIIa blocker for recanalization. Interpretation: ML-based rapid clot analysis is feasible and can identify AF with high accuracy, enabling selection of endovascular treatment strategy.",
author = "Chung, {Jong Won} and Kim, {Yoon Chul} and Jihoon Cha and Choi, {Eun Hyeok} and Kim, {Byung Moon} and Seo, {Woo Keun} and Kim, {Gyeong Moon} and Bang, {Oh Young}",
year = "2019",
month = "4",
doi = "10.1002/acn3.751",
language = "English",
volume = "6",
pages = "739--747",
journal = "Annals of Clinical and Translational Neurology",
issn = "2328-9503",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

Characterization of clot composition in acute cerebral infarct using machine learning techniques. / Chung, Jong Won; Kim, Yoon Chul; Cha, Jihoon; Choi, Eun Hyeok; Kim, Byung Moon; Seo, Woo Keun; Kim, Gyeong Moon; Bang, Oh Young.

In: Annals of Clinical and Translational Neurology, Vol. 6, No. 4, 04.2019, p. 739-747.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Characterization of clot composition in acute cerebral infarct using machine learning techniques

AU - Chung, Jong Won

AU - Kim, Yoon Chul

AU - Cha, Jihoon

AU - Choi, Eun Hyeok

AU - Kim, Byung Moon

AU - Seo, Woo Keun

AU - Kim, Gyeong Moon

AU - Bang, Oh Young

PY - 2019/4

Y1 - 2019/4

N2 - Objective: Clot characteristics can provide information on the cause of cerebral artery occlusion and may guide acute revascularization and secondary prevention strategies. We developed a rapid automated clot analysis system using machine learning (ML) and validated its accuracy in patients undergoing endovascular treatment. Methods: Pre-endovascular treatment gradient echo (GRE) images from consecutive patients with middle cerebral artery occlusion were utilized to develop and validate an ML system to predict whether atrial fibrillation (AF) was the underlying cause of ischemic stroke. The accuracy of the ML algorithm was compared with that of visual inspection by neuroimaging specialists for the presence of blooming artifact. Endovascular procedures and outcomes were compared in patients with and without AF. Results: Of 67 patients, 29 (43.3%) had AF. Of these, 13 had known AF and 16 were newly diagnosed with cardiac monitoring. By visual inspection, interrater correlation for blooming artifact was 0.73 and sensitivity and specificity for AF were 0.79 and 0.63, respectively. For AF classification, the ML algorithms yielded an average accuracy of > 75.4% in fivefold cross-validation with clot signal profiles obtained from 52 patients and an area under the curve >0.87 for the average AF probability from five signal profiles in external validation (n = 15). Analysis with an in-house interface took approximately 3 min per patient. Absence of AF was associated with increased number of passes by stentriever, high reocclusion frequency, and additional use of rescue stenting and/or glycogen IIb/IIIa blocker for recanalization. Interpretation: ML-based rapid clot analysis is feasible and can identify AF with high accuracy, enabling selection of endovascular treatment strategy.

AB - Objective: Clot characteristics can provide information on the cause of cerebral artery occlusion and may guide acute revascularization and secondary prevention strategies. We developed a rapid automated clot analysis system using machine learning (ML) and validated its accuracy in patients undergoing endovascular treatment. Methods: Pre-endovascular treatment gradient echo (GRE) images from consecutive patients with middle cerebral artery occlusion were utilized to develop and validate an ML system to predict whether atrial fibrillation (AF) was the underlying cause of ischemic stroke. The accuracy of the ML algorithm was compared with that of visual inspection by neuroimaging specialists for the presence of blooming artifact. Endovascular procedures and outcomes were compared in patients with and without AF. Results: Of 67 patients, 29 (43.3%) had AF. Of these, 13 had known AF and 16 were newly diagnosed with cardiac monitoring. By visual inspection, interrater correlation for blooming artifact was 0.73 and sensitivity and specificity for AF were 0.79 and 0.63, respectively. For AF classification, the ML algorithms yielded an average accuracy of > 75.4% in fivefold cross-validation with clot signal profiles obtained from 52 patients and an area under the curve >0.87 for the average AF probability from five signal profiles in external validation (n = 15). Analysis with an in-house interface took approximately 3 min per patient. Absence of AF was associated with increased number of passes by stentriever, high reocclusion frequency, and additional use of rescue stenting and/or glycogen IIb/IIIa blocker for recanalization. Interpretation: ML-based rapid clot analysis is feasible and can identify AF with high accuracy, enabling selection of endovascular treatment strategy.

UR - http://www.scopus.com/inward/record.url?scp=85064500204&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064500204&partnerID=8YFLogxK

U2 - 10.1002/acn3.751

DO - 10.1002/acn3.751

M3 - Article

AN - SCOPUS:85064500204

VL - 6

SP - 739

EP - 747

JO - Annals of Clinical and Translational Neurology

JF - Annals of Clinical and Translational Neurology

SN - 2328-9503

IS - 4

ER -