Quantitative prediction of acute ischemic tissue fate using support vector machine

Shiliang Huang, Qiang Shen, Timothy Q. Duong

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Accurate and quantitative prediction of ischemic tissue fate could improve decision-making in the clinical treatment of acute stroke. The goal of the present study is to explore the novel use of support vector machine (SVM) to predict infarct on a pixel-by-pixel basis using only acute cerebral blood flow (CBF), apparent diffusion coefficient (ADC) MRI data. The efficacy of SVM prediction model was tested on three stroke groups: 30-min, 60-min, and permanent middle cerebral-artery occlusion (n = 12 rats for each group). CBF, ADC and relaxation time constant (T2) were acquired during the acute phase up to 3 h and again at 24 h. Infarct was predicted using only acute (30-min) stroke data. Receiver-operating characteristic (ROC) analysis was used to quantify prediction accuracy. The areas under the receiver-operating curves were 86 ± 2.7%, 89 ± 1.4%, and 93 ± 0.8% using ADC + CBF data for the 30-min, 60-min and permanent middle cerebral artery occlusion (MCAO) group, respectively. Adding neighboring pixel information and spatial infarction incidence improved performance to 88 ± 2.8%, 94 ± 0.8%, and 97 ± 0.9%, respectively. SVM prediction compares favorably to a previously published artificial neural network (ANN) prediction algorithm operated on the same data sets. SVM prediction model has the potential to provide quantitative frameworks to aid clinical decision-making in the treatment of acute stroke.

Original languageEnglish (US)
Pages (from-to)77-84
Number of pages8
JournalBrain Research
Volume1405
DOIs
StatePublished - Aug 8 2011

Fingerprint

Cerebrovascular Circulation
Stroke
Middle Cerebral Artery Infarction
Diffusion Magnetic Resonance Imaging
ROC Curve
Infarction
Support Vector Machine
Incidence
Clinical Decision-Making

Keywords

  • ADC
  • ANN
  • CBF
  • DWI
  • fMRI
  • Perfusion-diffusion mismatch
  • Predictive model
  • PWI
  • Spatial infarction incidence
  • SVM

ASJC Scopus subject areas

  • Neuroscience(all)
  • Clinical Neurology
  • Developmental Biology
  • Molecular Biology

Cite this

Quantitative prediction of acute ischemic tissue fate using support vector machine. / Huang, Shiliang; Shen, Qiang; Duong, Timothy Q.

In: Brain Research, Vol. 1405, 08.08.2011, p. 77-84.

Research output: Contribution to journalArticle

Huang, Shiliang ; Shen, Qiang ; Duong, Timothy Q. / Quantitative prediction of acute ischemic tissue fate using support vector machine. In: Brain Research. 2011 ; Vol. 1405. pp. 77-84.
@article{ebf2494ce8a04aff8556e9ba36de2e94,
title = "Quantitative prediction of acute ischemic tissue fate using support vector machine",
abstract = "Accurate and quantitative prediction of ischemic tissue fate could improve decision-making in the clinical treatment of acute stroke. The goal of the present study is to explore the novel use of support vector machine (SVM) to predict infarct on a pixel-by-pixel basis using only acute cerebral blood flow (CBF), apparent diffusion coefficient (ADC) MRI data. The efficacy of SVM prediction model was tested on three stroke groups: 30-min, 60-min, and permanent middle cerebral-artery occlusion (n = 12 rats for each group). CBF, ADC and relaxation time constant (T2) were acquired during the acute phase up to 3 h and again at 24 h. Infarct was predicted using only acute (30-min) stroke data. Receiver-operating characteristic (ROC) analysis was used to quantify prediction accuracy. The areas under the receiver-operating curves were 86 ± 2.7{\%}, 89 ± 1.4{\%}, and 93 ± 0.8{\%} using ADC + CBF data for the 30-min, 60-min and permanent middle cerebral artery occlusion (MCAO) group, respectively. Adding neighboring pixel information and spatial infarction incidence improved performance to 88 ± 2.8{\%}, 94 ± 0.8{\%}, and 97 ± 0.9{\%}, respectively. SVM prediction compares favorably to a previously published artificial neural network (ANN) prediction algorithm operated on the same data sets. SVM prediction model has the potential to provide quantitative frameworks to aid clinical decision-making in the treatment of acute stroke.",
keywords = "ADC, ANN, CBF, DWI, fMRI, Perfusion-diffusion mismatch, Predictive model, PWI, Spatial infarction incidence, SVM",
author = "Shiliang Huang and Qiang Shen and Duong, {Timothy Q.}",
year = "2011",
month = "8",
day = "8",
doi = "10.1016/j.brainres.2011.05.066",
language = "English (US)",
volume = "1405",
pages = "77--84",
journal = "Brain Research",
issn = "0006-8993",
publisher = "Elsevier",

}

TY - JOUR

T1 - Quantitative prediction of acute ischemic tissue fate using support vector machine

AU - Huang, Shiliang

AU - Shen, Qiang

AU - Duong, Timothy Q.

PY - 2011/8/8

Y1 - 2011/8/8

N2 - Accurate and quantitative prediction of ischemic tissue fate could improve decision-making in the clinical treatment of acute stroke. The goal of the present study is to explore the novel use of support vector machine (SVM) to predict infarct on a pixel-by-pixel basis using only acute cerebral blood flow (CBF), apparent diffusion coefficient (ADC) MRI data. The efficacy of SVM prediction model was tested on three stroke groups: 30-min, 60-min, and permanent middle cerebral-artery occlusion (n = 12 rats for each group). CBF, ADC and relaxation time constant (T2) were acquired during the acute phase up to 3 h and again at 24 h. Infarct was predicted using only acute (30-min) stroke data. Receiver-operating characteristic (ROC) analysis was used to quantify prediction accuracy. The areas under the receiver-operating curves were 86 ± 2.7%, 89 ± 1.4%, and 93 ± 0.8% using ADC + CBF data for the 30-min, 60-min and permanent middle cerebral artery occlusion (MCAO) group, respectively. Adding neighboring pixel information and spatial infarction incidence improved performance to 88 ± 2.8%, 94 ± 0.8%, and 97 ± 0.9%, respectively. SVM prediction compares favorably to a previously published artificial neural network (ANN) prediction algorithm operated on the same data sets. SVM prediction model has the potential to provide quantitative frameworks to aid clinical decision-making in the treatment of acute stroke.

AB - Accurate and quantitative prediction of ischemic tissue fate could improve decision-making in the clinical treatment of acute stroke. The goal of the present study is to explore the novel use of support vector machine (SVM) to predict infarct on a pixel-by-pixel basis using only acute cerebral blood flow (CBF), apparent diffusion coefficient (ADC) MRI data. The efficacy of SVM prediction model was tested on three stroke groups: 30-min, 60-min, and permanent middle cerebral-artery occlusion (n = 12 rats for each group). CBF, ADC and relaxation time constant (T2) were acquired during the acute phase up to 3 h and again at 24 h. Infarct was predicted using only acute (30-min) stroke data. Receiver-operating characteristic (ROC) analysis was used to quantify prediction accuracy. The areas under the receiver-operating curves were 86 ± 2.7%, 89 ± 1.4%, and 93 ± 0.8% using ADC + CBF data for the 30-min, 60-min and permanent middle cerebral artery occlusion (MCAO) group, respectively. Adding neighboring pixel information and spatial infarction incidence improved performance to 88 ± 2.8%, 94 ± 0.8%, and 97 ± 0.9%, respectively. SVM prediction compares favorably to a previously published artificial neural network (ANN) prediction algorithm operated on the same data sets. SVM prediction model has the potential to provide quantitative frameworks to aid clinical decision-making in the treatment of acute stroke.

KW - ADC

KW - ANN

KW - CBF

KW - DWI

KW - fMRI

KW - Perfusion-diffusion mismatch

KW - Predictive model

KW - PWI

KW - Spatial infarction incidence

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=79960740340&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960740340&partnerID=8YFLogxK

U2 - 10.1016/j.brainres.2011.05.066

DO - 10.1016/j.brainres.2011.05.066

M3 - Article

C2 - 21741624

AN - SCOPUS:79960740340

VL - 1405

SP - 77

EP - 84

JO - Brain Research

JF - Brain Research

SN - 0006-8993

ER -