Concatenated spatially-localized random forests for hippocampus labeling in adult and infant MR brain images

Lichi Zhang, Qian Wang, Yaozong Gao, Hongxin Li, Guorong Wu, Dinggang Shen

Research output: Contribution to journalArticle

  • 6 Citations

Abstract

Automatic labeling of the hippocampus in brain MR images is highly demanded, as it has played an important role in imaging-based brain studies. However, accurate labeling of the hippocampus is still challenging, partially due to the ambiguous intensity boundary between the hippocampus and surrounding anatomies. In this paper, we propose a concatenated set of spatially-localized random forests for multi-atlas-based hippocampus labeling of adult/infant brain MR images. The contribution in our work is two-fold. First, each forest classifier is trained to label just a specific sub-region of the hippocampus, thus enhancing the labeling accuracy. Second, a novel forest selection strategy is proposed, such that each voxel in the test image can automatically select a set of optimal forests, and then dynamically fuses their respective outputs for determining the final label. Furthermore, we enhance the spatially-localized random forests with the aid of the auto-context strategy. In this way, our proposed learning framework can gradually refine the tentative labeling result for better performance. Experiments show that, regarding the large datasets of both adult and infant brain MR images, our method owns satisfactory scalability by segmenting the hippocampus accurately and efficiently.

LanguageEnglish (US)
Pages3-12
Number of pages10
JournalNeurocomputing
Volume229
DOIs
StatePublished - Mar 15 2017

Fingerprint

Labeling
Hippocampus
Brain
Labels
Electric fuses
Atlases
Scalability
Neuroimaging
Classifiers
Forests
Anatomy
Imaging techniques
Learning
Experiments

Keywords

  • Atlas selection
  • Brain MR images
  • Clustering
  • Image segmentation
  • Random forest

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Concatenated spatially-localized random forests for hippocampus labeling in adult and infant MR brain images. / Zhang, Lichi; Wang, Qian; Gao, Yaozong; Li, Hongxin; Wu, Guorong; Shen, Dinggang.

In: Neurocomputing, Vol. 229, 15.03.2017, p. 3-12.

Research output: Contribution to journalArticle

@article{c321779bc1b04cc793d7f3061ccea8d9,
title = "Concatenated spatially-localized random forests for hippocampus labeling in adult and infant MR brain images",
abstract = "Automatic labeling of the hippocampus in brain MR images is highly demanded, as it has played an important role in imaging-based brain studies. However, accurate labeling of the hippocampus is still challenging, partially due to the ambiguous intensity boundary between the hippocampus and surrounding anatomies. In this paper, we propose a concatenated set of spatially-localized random forests for multi-atlas-based hippocampus labeling of adult/infant brain MR images. The contribution in our work is two-fold. First, each forest classifier is trained to label just a specific sub-region of the hippocampus, thus enhancing the labeling accuracy. Second, a novel forest selection strategy is proposed, such that each voxel in the test image can automatically select a set of optimal forests, and then dynamically fuses their respective outputs for determining the final label. Furthermore, we enhance the spatially-localized random forests with the aid of the auto-context strategy. In this way, our proposed learning framework can gradually refine the tentative labeling result for better performance. Experiments show that, regarding the large datasets of both adult and infant brain MR images, our method owns satisfactory scalability by segmenting the hippocampus accurately and efficiently.",
keywords = "Atlas selection, Brain MR images, Clustering, Image segmentation, Random forest",
author = "Lichi Zhang and Qian Wang and Yaozong Gao and Hongxin Li and Guorong Wu and Dinggang Shen",
year = "2017",
month = "3",
day = "15",
doi = "10.1016/j.neucom.2016.05.082",
language = "English (US)",
volume = "229",
pages = "3--12",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Concatenated spatially-localized random forests for hippocampus labeling in adult and infant MR brain images

AU - Zhang,Lichi

AU - Wang,Qian

AU - Gao,Yaozong

AU - Li,Hongxin

AU - Wu,Guorong

AU - Shen,Dinggang

PY - 2017/3/15

Y1 - 2017/3/15

N2 - Automatic labeling of the hippocampus in brain MR images is highly demanded, as it has played an important role in imaging-based brain studies. However, accurate labeling of the hippocampus is still challenging, partially due to the ambiguous intensity boundary between the hippocampus and surrounding anatomies. In this paper, we propose a concatenated set of spatially-localized random forests for multi-atlas-based hippocampus labeling of adult/infant brain MR images. The contribution in our work is two-fold. First, each forest classifier is trained to label just a specific sub-region of the hippocampus, thus enhancing the labeling accuracy. Second, a novel forest selection strategy is proposed, such that each voxel in the test image can automatically select a set of optimal forests, and then dynamically fuses their respective outputs for determining the final label. Furthermore, we enhance the spatially-localized random forests with the aid of the auto-context strategy. In this way, our proposed learning framework can gradually refine the tentative labeling result for better performance. Experiments show that, regarding the large datasets of both adult and infant brain MR images, our method owns satisfactory scalability by segmenting the hippocampus accurately and efficiently.

AB - Automatic labeling of the hippocampus in brain MR images is highly demanded, as it has played an important role in imaging-based brain studies. However, accurate labeling of the hippocampus is still challenging, partially due to the ambiguous intensity boundary between the hippocampus and surrounding anatomies. In this paper, we propose a concatenated set of spatially-localized random forests for multi-atlas-based hippocampus labeling of adult/infant brain MR images. The contribution in our work is two-fold. First, each forest classifier is trained to label just a specific sub-region of the hippocampus, thus enhancing the labeling accuracy. Second, a novel forest selection strategy is proposed, such that each voxel in the test image can automatically select a set of optimal forests, and then dynamically fuses their respective outputs for determining the final label. Furthermore, we enhance the spatially-localized random forests with the aid of the auto-context strategy. In this way, our proposed learning framework can gradually refine the tentative labeling result for better performance. Experiments show that, regarding the large datasets of both adult and infant brain MR images, our method owns satisfactory scalability by segmenting the hippocampus accurately and efficiently.

KW - Atlas selection

KW - Brain MR images

KW - Clustering

KW - Image segmentation

KW - Random forest

UR - http://www.scopus.com/inward/record.url?scp=85003520708&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85003520708&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2016.05.082

DO - 10.1016/j.neucom.2016.05.082

M3 - Article

VL - 229

SP - 3

EP - 12

JO - Neurocomputing

T2 - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -