LABEL: Pediatric brain extraction using learning-based meta-algorithm

Feng Shi, Li Wang, Yakang Dai, John H. Gilmore, Weili Lin, Dinggang Shen

Research output: Contribution to journalArticle

102 Citations (Scopus)

Abstract

Magnetic resonance imaging of pediatric brain provides valuable information for early brain development studies. Automated brain extraction is challenging due to the small brain size and dynamic change of tissue contrast in the developing brains. In this paper, we propose a novel Learning Algorithm for Brain Extraction and Labeling (LABEL) specially for the pediatric MR brain images. The idea is to perform multiple complementary brain extractions on a given testing image by using a meta-algorithm, including BET and BSE, where the parameters of each run of the meta-algorithm are effectively learned from the training data. Also, the representative subjects are selected as exemplars and used to guide brain extraction of new subjects in different age groups. We further develop a level-set based fusion method to combine multiple brain extractions together with a closed smooth surface for obtaining the final extraction. The proposed method has been extensively evaluated in subjects of three representative age groups, such as neonate (less than 2. months), infant (1-2. years), and child (5-18. years). Experimental results show that, with 45 subjects for training (15 neonates, 15 infant, and 15 children), the proposed method can produce more accurate brain extraction results on 246 testing subjects (75 neonates, 126 infants, and 45 children), i.e., at average Jaccard Index of 0.953, compared to those by BET (0.918), BSE (0.902), ROBEX (0.901), GCUT (0.856), and other fusion methods such as Majority Voting (0.919) and STAPLE (0.941). Along with the largely-improved computational efficiency, the proposed method demonstrates its ability of automated brain extraction for pediatric MR images in a large age range.

Original languageEnglish
Pages (from-to)1975-1986
Number of pages12
JournalNeuroImage
Volume62
Issue number3
DOIs
Publication statusPublished - 2012 Sep 1
Externally publishedYes

Fingerprint

Learning
Pediatrics
Brain
Newborn Infant
Age Groups
Aptitude
Politics
Magnetic Resonance Imaging
Efficiency

Keywords

  • Affinity propagation
  • Brain extraction
  • Infant brain analysis
  • Label fusion
  • Level-set
  • Meta-algorithm
  • Skull stripping

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

LABEL : Pediatric brain extraction using learning-based meta-algorithm. / Shi, Feng; Wang, Li; Dai, Yakang; Gilmore, John H.; Lin, Weili; Shen, Dinggang.

In: NeuroImage, Vol. 62, No. 3, 01.09.2012, p. 1975-1986.

Research output: Contribution to journalArticle

Shi, Feng ; Wang, Li ; Dai, Yakang ; Gilmore, John H. ; Lin, Weili ; Shen, Dinggang. / LABEL : Pediatric brain extraction using learning-based meta-algorithm. In: NeuroImage. 2012 ; Vol. 62, No. 3. pp. 1975-1986.
@article{8861a491ca3e45cca45b2b4e60c71daf,
title = "LABEL: Pediatric brain extraction using learning-based meta-algorithm",
abstract = "Magnetic resonance imaging of pediatric brain provides valuable information for early brain development studies. Automated brain extraction is challenging due to the small brain size and dynamic change of tissue contrast in the developing brains. In this paper, we propose a novel Learning Algorithm for Brain Extraction and Labeling (LABEL) specially for the pediatric MR brain images. The idea is to perform multiple complementary brain extractions on a given testing image by using a meta-algorithm, including BET and BSE, where the parameters of each run of the meta-algorithm are effectively learned from the training data. Also, the representative subjects are selected as exemplars and used to guide brain extraction of new subjects in different age groups. We further develop a level-set based fusion method to combine multiple brain extractions together with a closed smooth surface for obtaining the final extraction. The proposed method has been extensively evaluated in subjects of three representative age groups, such as neonate (less than 2. months), infant (1-2. years), and child (5-18. years). Experimental results show that, with 45 subjects for training (15 neonates, 15 infant, and 15 children), the proposed method can produce more accurate brain extraction results on 246 testing subjects (75 neonates, 126 infants, and 45 children), i.e., at average Jaccard Index of 0.953, compared to those by BET (0.918), BSE (0.902), ROBEX (0.901), GCUT (0.856), and other fusion methods such as Majority Voting (0.919) and STAPLE (0.941). Along with the largely-improved computational efficiency, the proposed method demonstrates its ability of automated brain extraction for pediatric MR images in a large age range.",
keywords = "Affinity propagation, Brain extraction, Infant brain analysis, Label fusion, Level-set, Meta-algorithm, Skull stripping",
author = "Feng Shi and Li Wang and Yakang Dai and Gilmore, {John H.} and Weili Lin and Dinggang Shen",
year = "2012",
month = "9",
day = "1",
doi = "10.1016/j.neuroimage.2012.05.042",
language = "English",
volume = "62",
pages = "1975--1986",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "3",

}

TY - JOUR

T1 - LABEL

T2 - Pediatric brain extraction using learning-based meta-algorithm

AU - Shi, Feng

AU - Wang, Li

AU - Dai, Yakang

AU - Gilmore, John H.

AU - Lin, Weili

AU - Shen, Dinggang

PY - 2012/9/1

Y1 - 2012/9/1

N2 - Magnetic resonance imaging of pediatric brain provides valuable information for early brain development studies. Automated brain extraction is challenging due to the small brain size and dynamic change of tissue contrast in the developing brains. In this paper, we propose a novel Learning Algorithm for Brain Extraction and Labeling (LABEL) specially for the pediatric MR brain images. The idea is to perform multiple complementary brain extractions on a given testing image by using a meta-algorithm, including BET and BSE, where the parameters of each run of the meta-algorithm are effectively learned from the training data. Also, the representative subjects are selected as exemplars and used to guide brain extraction of new subjects in different age groups. We further develop a level-set based fusion method to combine multiple brain extractions together with a closed smooth surface for obtaining the final extraction. The proposed method has been extensively evaluated in subjects of three representative age groups, such as neonate (less than 2. months), infant (1-2. years), and child (5-18. years). Experimental results show that, with 45 subjects for training (15 neonates, 15 infant, and 15 children), the proposed method can produce more accurate brain extraction results on 246 testing subjects (75 neonates, 126 infants, and 45 children), i.e., at average Jaccard Index of 0.953, compared to those by BET (0.918), BSE (0.902), ROBEX (0.901), GCUT (0.856), and other fusion methods such as Majority Voting (0.919) and STAPLE (0.941). Along with the largely-improved computational efficiency, the proposed method demonstrates its ability of automated brain extraction for pediatric MR images in a large age range.

AB - Magnetic resonance imaging of pediatric brain provides valuable information for early brain development studies. Automated brain extraction is challenging due to the small brain size and dynamic change of tissue contrast in the developing brains. In this paper, we propose a novel Learning Algorithm for Brain Extraction and Labeling (LABEL) specially for the pediatric MR brain images. The idea is to perform multiple complementary brain extractions on a given testing image by using a meta-algorithm, including BET and BSE, where the parameters of each run of the meta-algorithm are effectively learned from the training data. Also, the representative subjects are selected as exemplars and used to guide brain extraction of new subjects in different age groups. We further develop a level-set based fusion method to combine multiple brain extractions together with a closed smooth surface for obtaining the final extraction. The proposed method has been extensively evaluated in subjects of three representative age groups, such as neonate (less than 2. months), infant (1-2. years), and child (5-18. years). Experimental results show that, with 45 subjects for training (15 neonates, 15 infant, and 15 children), the proposed method can produce more accurate brain extraction results on 246 testing subjects (75 neonates, 126 infants, and 45 children), i.e., at average Jaccard Index of 0.953, compared to those by BET (0.918), BSE (0.902), ROBEX (0.901), GCUT (0.856), and other fusion methods such as Majority Voting (0.919) and STAPLE (0.941). Along with the largely-improved computational efficiency, the proposed method demonstrates its ability of automated brain extraction for pediatric MR images in a large age range.

KW - Affinity propagation

KW - Brain extraction

KW - Infant brain analysis

KW - Label fusion

KW - Level-set

KW - Meta-algorithm

KW - Skull stripping

UR - http://www.scopus.com/inward/record.url?scp=84863506007&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863506007&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2012.05.042

DO - 10.1016/j.neuroimage.2012.05.042

M3 - Article

C2 - 22634859

AN - SCOPUS:84863506007

VL - 62

SP - 1975

EP - 1986

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 3

ER -