Topological correction of infant white matter surfaces using anatomically constrained convolutional neural network

UNC/UMN Baby Connectome Project Consortium

Research output: Contribution to journalArticle

Abstract

Reconstruction of accurate cortical surfaces without topological errors (i.e., handles and holes) from infant brain MR images is very important in early brain development studies. However, infant brain MR images typically suffer extremely low tissue contrast and dynamic imaging appearance patterns. Thus, it is inevitable to have large amounts of topological errors in the segmented infant brain tissue images, which lead to inaccurately reconstructed cortical surfaces with topological errors. To address this issue, inspired by recent advances in deep learning, we propose an anatomically constrained network for topological correction on infant cortical surfaces. Specifically, in our method, we first locate regions of potential topological defects by leveraging a topology-preserving level set method. Then, we propose an anatomically constrained network to correct those candidate voxels in the located regions. Since infant cortical surfaces often contain large and complex handles or holes, it is difficult to completely correct all errors using one-shot correction. Therefore, we further enroll these two steps into an iterative framework to gradually correct large topological errors. To the best of our knowledge, this is the first work to introduce deep learning approach for topological correction of infant cortical surfaces. We compare our method with the state-of-the-art methods on both simulated topological errors and real topological errors in human infant brain MR images. Moreover, we also validate our method on the infant brain MR images of macaques. All experimental results show the superior performance of the proposed method.

Original languageEnglish
Pages (from-to)114-124
Number of pages11
JournalNeuroImage
Volume198
DOIs
Publication statusPublished - 2019 Sep 1

Fingerprint

Brain
Learning
White Matter
Macaca

Keywords

  • Anatomically constrained network
  • Infant cortical surfaces
  • Topological correction

ASJC Scopus subject areas

  • Neurology
  • Cognitive Neuroscience

Cite this

Topological correction of infant white matter surfaces using anatomically constrained convolutional neural network. / UNC/UMN Baby Connectome Project Consortium.

In: NeuroImage, Vol. 198, 01.09.2019, p. 114-124.

Research output: Contribution to journalArticle

UNC/UMN Baby Connectome Project Consortium. / Topological correction of infant white matter surfaces using anatomically constrained convolutional neural network. In: NeuroImage. 2019 ; Vol. 198. pp. 114-124.
@article{4bea891ed0e74c53b192bbcbd2e36b82,
title = "Topological correction of infant white matter surfaces using anatomically constrained convolutional neural network",
abstract = "Reconstruction of accurate cortical surfaces without topological errors (i.e., handles and holes) from infant brain MR images is very important in early brain development studies. However, infant brain MR images typically suffer extremely low tissue contrast and dynamic imaging appearance patterns. Thus, it is inevitable to have large amounts of topological errors in the segmented infant brain tissue images, which lead to inaccurately reconstructed cortical surfaces with topological errors. To address this issue, inspired by recent advances in deep learning, we propose an anatomically constrained network for topological correction on infant cortical surfaces. Specifically, in our method, we first locate regions of potential topological defects by leveraging a topology-preserving level set method. Then, we propose an anatomically constrained network to correct those candidate voxels in the located regions. Since infant cortical surfaces often contain large and complex handles or holes, it is difficult to completely correct all errors using one-shot correction. Therefore, we further enroll these two steps into an iterative framework to gradually correct large topological errors. To the best of our knowledge, this is the first work to introduce deep learning approach for topological correction of infant cortical surfaces. We compare our method with the state-of-the-art methods on both simulated topological errors and real topological errors in human infant brain MR images. Moreover, we also validate our method on the infant brain MR images of macaques. All experimental results show the superior performance of the proposed method.",
keywords = "Anatomically constrained network, Infant cortical surfaces, Topological correction",
author = "{UNC/UMN Baby Connectome Project Consortium} and Liang Sun and Daoqiang Zhang and Chunfeng Lian and Li Wang and Zhengwang Wu and Wei Shao and Weili Lin and Dinggang Shen and Gang Li",
year = "2019",
month = "9",
day = "1",
doi = "10.1016/j.neuroimage.2019.05.037",
language = "English",
volume = "198",
pages = "114--124",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Topological correction of infant white matter surfaces using anatomically constrained convolutional neural network

AU - UNC/UMN Baby Connectome Project Consortium

AU - Sun, Liang

AU - Zhang, Daoqiang

AU - Lian, Chunfeng

AU - Wang, Li

AU - Wu, Zhengwang

AU - Shao, Wei

AU - Lin, Weili

AU - Shen, Dinggang

AU - Li, Gang

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Reconstruction of accurate cortical surfaces without topological errors (i.e., handles and holes) from infant brain MR images is very important in early brain development studies. However, infant brain MR images typically suffer extremely low tissue contrast and dynamic imaging appearance patterns. Thus, it is inevitable to have large amounts of topological errors in the segmented infant brain tissue images, which lead to inaccurately reconstructed cortical surfaces with topological errors. To address this issue, inspired by recent advances in deep learning, we propose an anatomically constrained network for topological correction on infant cortical surfaces. Specifically, in our method, we first locate regions of potential topological defects by leveraging a topology-preserving level set method. Then, we propose an anatomically constrained network to correct those candidate voxels in the located regions. Since infant cortical surfaces often contain large and complex handles or holes, it is difficult to completely correct all errors using one-shot correction. Therefore, we further enroll these two steps into an iterative framework to gradually correct large topological errors. To the best of our knowledge, this is the first work to introduce deep learning approach for topological correction of infant cortical surfaces. We compare our method with the state-of-the-art methods on both simulated topological errors and real topological errors in human infant brain MR images. Moreover, we also validate our method on the infant brain MR images of macaques. All experimental results show the superior performance of the proposed method.

AB - Reconstruction of accurate cortical surfaces without topological errors (i.e., handles and holes) from infant brain MR images is very important in early brain development studies. However, infant brain MR images typically suffer extremely low tissue contrast and dynamic imaging appearance patterns. Thus, it is inevitable to have large amounts of topological errors in the segmented infant brain tissue images, which lead to inaccurately reconstructed cortical surfaces with topological errors. To address this issue, inspired by recent advances in deep learning, we propose an anatomically constrained network for topological correction on infant cortical surfaces. Specifically, in our method, we first locate regions of potential topological defects by leveraging a topology-preserving level set method. Then, we propose an anatomically constrained network to correct those candidate voxels in the located regions. Since infant cortical surfaces often contain large and complex handles or holes, it is difficult to completely correct all errors using one-shot correction. Therefore, we further enroll these two steps into an iterative framework to gradually correct large topological errors. To the best of our knowledge, this is the first work to introduce deep learning approach for topological correction of infant cortical surfaces. We compare our method with the state-of-the-art methods on both simulated topological errors and real topological errors in human infant brain MR images. Moreover, we also validate our method on the infant brain MR images of macaques. All experimental results show the superior performance of the proposed method.

KW - Anatomically constrained network

KW - Infant cortical surfaces

KW - Topological correction

UR - http://www.scopus.com/inward/record.url?scp=85065924616&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065924616&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2019.05.037

DO - 10.1016/j.neuroimage.2019.05.037

M3 - Article

C2 - 31112785

AN - SCOPUS:85065924616

VL - 198

SP - 114

EP - 124

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

ER -