Intervening Nidal Brain Parenchyma and Risk of Radiation-Induced Changes After Radiosurgery for Brain Arteriovenous Malformation

A Study Using an Unsupervised Machine Learning Algorithm

Cheng Chia Lee, Huai Che Yang, Chung Jung Lin, Ching Jen Chen, Hsiu Mei Wu, Cheng Ying Shiau, Wan Yuo Guo, David Hung-Chi Pan, Kang Du Liu, Wen Yuh Chung, Syu Jyun Peng

Research output: Contribution to journalArticle

Abstract

Objective: To assess the sensitivity and specificity of arteriovenous malformation (AVM) nidal component identification and quantification using an unsupervised machine learning algorithm and to evaluate the association between intervening nidal brain parenchyma and radiation-induced changes (RICs) after stereotactic radiosurgery. Methods: Fully automated segmentation via unsupervised classification with fuzzy c-means clustering was used to analyze the AVM nidus on T2-weighted magnetic resonance imaging studies. The proportions of vasculature, brain parenchyma, and cerebrospinal fluid were quantified. These were compared with the results from manual segmentation. The association between the brain parenchyma component and RIC development was assessed. Results: The proposed algorithm was applied to 39 unruptured AVMs in 39 patients (17 female and 22 male patients), with a median age of 27 years. The median proportion of the constituents was as follows: vasculature, 31.3%; brain parenchyma, 48.4%; and cerebrospinal fluid, 16.8%. RICs were identified in 17 of the 39 patients (43.6%). Compared with manual segmentation, the automated algorithm was able to achieve a Dice similarity index of 79.5% (sensitivity, 73.5%; specificity, 85.5%). RICs were associated with a greater proportion of intervening nidal brain parenchyma (52.0% vs. 45.3%; P = 0.015). Obliteration was not associated with greater proportions of nidal vasculature (36.0% vs. 31.2%; P = 0.152). Conclusions: The automated segmentation algorithm was able to achieve classification of the AVM nidus components with relative accuracy. Greater proportions of intervening nidal brain parenchyma were associated with RICs.

Original languageEnglish
Pages (from-to)e132-e138
JournalWorld Neurosurgery
Volume125
DOIs
Publication statusPublished - May 1 2019
Externally publishedYes

Fingerprint

Radiosurgery
Arteriovenous Malformations
Radiation
Brain
Cerebrospinal Fluid
Cluster Analysis
Unsupervised Machine Learning
Magnetic Resonance Imaging
Sensitivity and Specificity

Keywords

  • Adverse radiation effects
  • Arteriovenous malformation
  • Fuzzy c-means
  • Gamma knife radiosurgery
  • Image analysis
  • Radiation-induced changes
  • Stereotactic radiosurgery

ASJC Scopus subject areas

  • Surgery
  • Clinical Neurology

Cite this

Intervening Nidal Brain Parenchyma and Risk of Radiation-Induced Changes After Radiosurgery for Brain Arteriovenous Malformation : A Study Using an Unsupervised Machine Learning Algorithm. / Lee, Cheng Chia; Yang, Huai Che; Lin, Chung Jung; Chen, Ching Jen; Wu, Hsiu Mei; Shiau, Cheng Ying; Guo, Wan Yuo; Hung-Chi Pan, David; Liu, Kang Du; Chung, Wen Yuh; Peng, Syu Jyun.

In: World Neurosurgery, Vol. 125, 01.05.2019, p. e132-e138.

Research output: Contribution to journalArticle

Lee, Cheng Chia ; Yang, Huai Che ; Lin, Chung Jung ; Chen, Ching Jen ; Wu, Hsiu Mei ; Shiau, Cheng Ying ; Guo, Wan Yuo ; Hung-Chi Pan, David ; Liu, Kang Du ; Chung, Wen Yuh ; Peng, Syu Jyun. / Intervening Nidal Brain Parenchyma and Risk of Radiation-Induced Changes After Radiosurgery for Brain Arteriovenous Malformation : A Study Using an Unsupervised Machine Learning Algorithm. In: World Neurosurgery. 2019 ; Vol. 125. pp. e132-e138.
@article{13cc075c99b34ec89dc440f1e6dfc7ca,
title = "Intervening Nidal Brain Parenchyma and Risk of Radiation-Induced Changes After Radiosurgery for Brain Arteriovenous Malformation: A Study Using an Unsupervised Machine Learning Algorithm",
abstract = "Objective: To assess the sensitivity and specificity of arteriovenous malformation (AVM) nidal component identification and quantification using an unsupervised machine learning algorithm and to evaluate the association between intervening nidal brain parenchyma and radiation-induced changes (RICs) after stereotactic radiosurgery. Methods: Fully automated segmentation via unsupervised classification with fuzzy c-means clustering was used to analyze the AVM nidus on T2-weighted magnetic resonance imaging studies. The proportions of vasculature, brain parenchyma, and cerebrospinal fluid were quantified. These were compared with the results from manual segmentation. The association between the brain parenchyma component and RIC development was assessed. Results: The proposed algorithm was applied to 39 unruptured AVMs in 39 patients (17 female and 22 male patients), with a median age of 27 years. The median proportion of the constituents was as follows: vasculature, 31.3{\%}; brain parenchyma, 48.4{\%}; and cerebrospinal fluid, 16.8{\%}. RICs were identified in 17 of the 39 patients (43.6{\%}). Compared with manual segmentation, the automated algorithm was able to achieve a Dice similarity index of 79.5{\%} (sensitivity, 73.5{\%}; specificity, 85.5{\%}). RICs were associated with a greater proportion of intervening nidal brain parenchyma (52.0{\%} vs. 45.3{\%}; P = 0.015). Obliteration was not associated with greater proportions of nidal vasculature (36.0{\%} vs. 31.2{\%}; P = 0.152). Conclusions: The automated segmentation algorithm was able to achieve classification of the AVM nidus components with relative accuracy. Greater proportions of intervening nidal brain parenchyma were associated with RICs.",
keywords = "Adverse radiation effects, Arteriovenous malformation, Fuzzy c-means, Gamma knife radiosurgery, Image analysis, Radiation-induced changes, Stereotactic radiosurgery",
author = "Lee, {Cheng Chia} and Yang, {Huai Che} and Lin, {Chung Jung} and Chen, {Ching Jen} and Wu, {Hsiu Mei} and Shiau, {Cheng Ying} and Guo, {Wan Yuo} and {Hung-Chi Pan}, David and Liu, {Kang Du} and Chung, {Wen Yuh} and Peng, {Syu Jyun}",
year = "2019",
month = "5",
day = "1",
doi = "10.1016/j.wneu.2018.12.220",
language = "English",
volume = "125",
pages = "e132--e138",
journal = "World Neurosurgery",
issn = "1878-8750",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - Intervening Nidal Brain Parenchyma and Risk of Radiation-Induced Changes After Radiosurgery for Brain Arteriovenous Malformation

T2 - A Study Using an Unsupervised Machine Learning Algorithm

AU - Lee, Cheng Chia

AU - Yang, Huai Che

AU - Lin, Chung Jung

AU - Chen, Ching Jen

AU - Wu, Hsiu Mei

AU - Shiau, Cheng Ying

AU - Guo, Wan Yuo

AU - Hung-Chi Pan, David

AU - Liu, Kang Du

AU - Chung, Wen Yuh

AU - Peng, Syu Jyun

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Objective: To assess the sensitivity and specificity of arteriovenous malformation (AVM) nidal component identification and quantification using an unsupervised machine learning algorithm and to evaluate the association between intervening nidal brain parenchyma and radiation-induced changes (RICs) after stereotactic radiosurgery. Methods: Fully automated segmentation via unsupervised classification with fuzzy c-means clustering was used to analyze the AVM nidus on T2-weighted magnetic resonance imaging studies. The proportions of vasculature, brain parenchyma, and cerebrospinal fluid were quantified. These were compared with the results from manual segmentation. The association between the brain parenchyma component and RIC development was assessed. Results: The proposed algorithm was applied to 39 unruptured AVMs in 39 patients (17 female and 22 male patients), with a median age of 27 years. The median proportion of the constituents was as follows: vasculature, 31.3%; brain parenchyma, 48.4%; and cerebrospinal fluid, 16.8%. RICs were identified in 17 of the 39 patients (43.6%). Compared with manual segmentation, the automated algorithm was able to achieve a Dice similarity index of 79.5% (sensitivity, 73.5%; specificity, 85.5%). RICs were associated with a greater proportion of intervening nidal brain parenchyma (52.0% vs. 45.3%; P = 0.015). Obliteration was not associated with greater proportions of nidal vasculature (36.0% vs. 31.2%; P = 0.152). Conclusions: The automated segmentation algorithm was able to achieve classification of the AVM nidus components with relative accuracy. Greater proportions of intervening nidal brain parenchyma were associated with RICs.

AB - Objective: To assess the sensitivity and specificity of arteriovenous malformation (AVM) nidal component identification and quantification using an unsupervised machine learning algorithm and to evaluate the association between intervening nidal brain parenchyma and radiation-induced changes (RICs) after stereotactic radiosurgery. Methods: Fully automated segmentation via unsupervised classification with fuzzy c-means clustering was used to analyze the AVM nidus on T2-weighted magnetic resonance imaging studies. The proportions of vasculature, brain parenchyma, and cerebrospinal fluid were quantified. These were compared with the results from manual segmentation. The association between the brain parenchyma component and RIC development was assessed. Results: The proposed algorithm was applied to 39 unruptured AVMs in 39 patients (17 female and 22 male patients), with a median age of 27 years. The median proportion of the constituents was as follows: vasculature, 31.3%; brain parenchyma, 48.4%; and cerebrospinal fluid, 16.8%. RICs were identified in 17 of the 39 patients (43.6%). Compared with manual segmentation, the automated algorithm was able to achieve a Dice similarity index of 79.5% (sensitivity, 73.5%; specificity, 85.5%). RICs were associated with a greater proportion of intervening nidal brain parenchyma (52.0% vs. 45.3%; P = 0.015). Obliteration was not associated with greater proportions of nidal vasculature (36.0% vs. 31.2%; P = 0.152). Conclusions: The automated segmentation algorithm was able to achieve classification of the AVM nidus components with relative accuracy. Greater proportions of intervening nidal brain parenchyma were associated with RICs.

KW - Adverse radiation effects

KW - Arteriovenous malformation

KW - Fuzzy c-means

KW - Gamma knife radiosurgery

KW - Image analysis

KW - Radiation-induced changes

KW - Stereotactic radiosurgery

UR - http://www.scopus.com/inward/record.url?scp=85061637355&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061637355&partnerID=8YFLogxK

U2 - 10.1016/j.wneu.2018.12.220

DO - 10.1016/j.wneu.2018.12.220

M3 - Article

VL - 125

SP - e132-e138

JO - World Neurosurgery

JF - World Neurosurgery

SN - 1878-8750

ER -