Computer-Aided Detection of Hyperacute Stroke Based on Relative Radiomic Patterns in Computed Tomography

Chung-Ming Lo, Peng-Hsiang Hung, Kevin Li-Chun Hsieh

Research output: Contribution to journalArticle

Abstract

Ischemic stroke is one of the leading causes of disability and death. To achieve timely assessments, a computer-aided diagnosis (CAD) system was proposed to perform early recognition of hyperacute ischemic stroke based on non-contrast computed tomography (NCCT). In total, 26 patients with hyperacute ischemic stroke (with onset <6 h previous) and 56 normal controls composed the image database. For each NCCT slice, textural features were extracted from Ranklet-transformed images which had enhanced local contrast. Textural differences between the two sides of an image were calculated and combined in a machine learning classifier to detect stroke areas. The proposed CAD system using Ranklet features achieved significantly higher accuracy (81% vs. 71%), specificity (90% vs. 79%), and area under the curve (Az) (0.81 vs. 0.73) than conventional textural features. Diagnostic suggestions provided by the CAD system are fast and promising and could be useful in the pipeline of hyperacute ischemic stroke assessments.
Original languageTraditional Chinese
Pages (from-to)1668
Number of pages11
JournalApplied Sciences
Volume9
Issue number8
DOIs
Publication statusPublished - 2019

Cite this

Computer-Aided Detection of Hyperacute Stroke Based on Relative Radiomic Patterns in Computed Tomography. / Lo, Chung-Ming; Hung, Peng-Hsiang; Hsieh, Kevin Li-Chun.

In: Applied Sciences, Vol. 9, No. 8, 2019, p. 1668.

Research output: Contribution to journalArticle

@article{e2aa0521f9af4259b9651972ba638a7f,
title = "Computer-Aided Detection of Hyperacute Stroke Based on Relative Radiomic Patterns in Computed Tomography",
abstract = "Ischemic stroke is one of the leading causes of disability and death. To achieve timely assessments, a computer-aided diagnosis (CAD) system was proposed to perform early recognition of hyperacute ischemic stroke based on non-contrast computed tomography (NCCT). In total, 26 patients with hyperacute ischemic stroke (with onset <6 h previous) and 56 normal controls composed the image database. For each NCCT slice, textural features were extracted from Ranklet-transformed images which had enhanced local contrast. Textural differences between the two sides of an image were calculated and combined in a machine learning classifier to detect stroke areas. The proposed CAD system using Ranklet features achieved significantly higher accuracy (81{\%} vs. 71{\%}), specificity (90{\%} vs. 79{\%}), and area under the curve (Az) (0.81 vs. 0.73) than conventional textural features. Diagnostic suggestions provided by the CAD system are fast and promising and could be useful in the pipeline of hyperacute ischemic stroke assessments.",
author = "Chung-Ming Lo and Peng-Hsiang Hung and Hsieh, {Kevin Li-Chun}",
year = "2019",
doi = "https://doi.org/10.3390/app9081668",
language = "繁體中文",
volume = "9",
pages = "1668",
journal = "Applied Sciences",
issn = "1454-5101",
publisher = "Politechnica University of Bucharest",
number = "8",

}

TY - JOUR

T1 - Computer-Aided Detection of Hyperacute Stroke Based on Relative Radiomic Patterns in Computed Tomography

AU - Lo, Chung-Ming

AU - Hung, Peng-Hsiang

AU - Hsieh, Kevin Li-Chun

PY - 2019

Y1 - 2019

N2 - Ischemic stroke is one of the leading causes of disability and death. To achieve timely assessments, a computer-aided diagnosis (CAD) system was proposed to perform early recognition of hyperacute ischemic stroke based on non-contrast computed tomography (NCCT). In total, 26 patients with hyperacute ischemic stroke (with onset <6 h previous) and 56 normal controls composed the image database. For each NCCT slice, textural features were extracted from Ranklet-transformed images which had enhanced local contrast. Textural differences between the two sides of an image were calculated and combined in a machine learning classifier to detect stroke areas. The proposed CAD system using Ranklet features achieved significantly higher accuracy (81% vs. 71%), specificity (90% vs. 79%), and area under the curve (Az) (0.81 vs. 0.73) than conventional textural features. Diagnostic suggestions provided by the CAD system are fast and promising and could be useful in the pipeline of hyperacute ischemic stroke assessments.

AB - Ischemic stroke is one of the leading causes of disability and death. To achieve timely assessments, a computer-aided diagnosis (CAD) system was proposed to perform early recognition of hyperacute ischemic stroke based on non-contrast computed tomography (NCCT). In total, 26 patients with hyperacute ischemic stroke (with onset <6 h previous) and 56 normal controls composed the image database. For each NCCT slice, textural features were extracted from Ranklet-transformed images which had enhanced local contrast. Textural differences between the two sides of an image were calculated and combined in a machine learning classifier to detect stroke areas. The proposed CAD system using Ranklet features achieved significantly higher accuracy (81% vs. 71%), specificity (90% vs. 79%), and area under the curve (Az) (0.81 vs. 0.73) than conventional textural features. Diagnostic suggestions provided by the CAD system are fast and promising and could be useful in the pipeline of hyperacute ischemic stroke assessments.

U2 - https://doi.org/10.3390/app9081668

DO - https://doi.org/10.3390/app9081668

M3 - 文章

VL - 9

SP - 1668

JO - Applied Sciences

JF - Applied Sciences

SN - 1454-5101

IS - 8

ER -