Hippocampal Segmentation from Longitudinal Infant Brain MR Images via Classification-Guided Boundary Regression

Yeqin Shao, Jaeil Kim, Yaozong Gao, Qian Wang, Weili Lin, Dinggang Shen

Research output: Contribution to journalArticle

Abstract

Hippocampal segmentation from infant brain MR images is indispensable for studying early brain development. However, most of the hippocampal segmentation methods were developed for population-based adult brain images, which are not suitable for longitudinal infant brain images acquired in the first year of life due to the low image contrast and variable development patterns of the hippocampal structure. To address these challenges, we propose a classification-guided boundary regression method to first detect hippocampal boundaries in the longitudinal infant brain images and then use those detected boundaries to guide the deformable model for final segmentation. Specifically, we first employ a classification-guided regression forest to predict the 3D displacements from individual image voxels to the potential hippocampal boundaries. These predicted displacements then determine the boundary maps by a voting strategy. Second, we iteratively enhance the voted hippocampal boundary map by incorporating the spatial context information given the tentative boundary estimation of the current time point. Besides, the longitudinal context information from all time points of the temporal sequence of the same subject (i.e., given their tentative segmentation results) is also utilized to facilitate accurate segmentation. Finally, a deformable model is applied to the enhanced voted boundary maps for achieving the longitudinal hippocampal segmentation. The experiments on infant brain MR images acquired from 2-week-old to 1-year-old show promising hippocampal segmentation results, indicating the applicability of our method in early brain development study.

Original languageEnglish
Article number8664424
Pages (from-to)33728-33740
Number of pages13
JournalIEEE Access
Volume7
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Brain
Experiments

Keywords

  • classification-guided boundary regression
  • deformable segmentation
  • hippocampal segmentation
  • Infant brain
  • longitudinal MR image

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Cite this

Hippocampal Segmentation from Longitudinal Infant Brain MR Images via Classification-Guided Boundary Regression. / Shao, Yeqin; Kim, Jaeil; Gao, Yaozong; Wang, Qian; Lin, Weili; Shen, Dinggang.

In: IEEE Access, Vol. 7, 8664424, 01.01.2019, p. 33728-33740.

Research output: Contribution to journalArticle

Shao, Yeqin ; Kim, Jaeil ; Gao, Yaozong ; Wang, Qian ; Lin, Weili ; Shen, Dinggang. / Hippocampal Segmentation from Longitudinal Infant Brain MR Images via Classification-Guided Boundary Regression. In: IEEE Access. 2019 ; Vol. 7. pp. 33728-33740.
@article{bb94c2988ddd4f638749521786744907,
title = "Hippocampal Segmentation from Longitudinal Infant Brain MR Images via Classification-Guided Boundary Regression",
abstract = "Hippocampal segmentation from infant brain MR images is indispensable for studying early brain development. However, most of the hippocampal segmentation methods were developed for population-based adult brain images, which are not suitable for longitudinal infant brain images acquired in the first year of life due to the low image contrast and variable development patterns of the hippocampal structure. To address these challenges, we propose a classification-guided boundary regression method to first detect hippocampal boundaries in the longitudinal infant brain images and then use those detected boundaries to guide the deformable model for final segmentation. Specifically, we first employ a classification-guided regression forest to predict the 3D displacements from individual image voxels to the potential hippocampal boundaries. These predicted displacements then determine the boundary maps by a voting strategy. Second, we iteratively enhance the voted hippocampal boundary map by incorporating the spatial context information given the tentative boundary estimation of the current time point. Besides, the longitudinal context information from all time points of the temporal sequence of the same subject (i.e., given their tentative segmentation results) is also utilized to facilitate accurate segmentation. Finally, a deformable model is applied to the enhanced voted boundary maps for achieving the longitudinal hippocampal segmentation. The experiments on infant brain MR images acquired from 2-week-old to 1-year-old show promising hippocampal segmentation results, indicating the applicability of our method in early brain development study.",
keywords = "classification-guided boundary regression, deformable segmentation, hippocampal segmentation, Infant brain, longitudinal MR image",
author = "Yeqin Shao and Jaeil Kim and Yaozong Gao and Qian Wang and Weili Lin and Dinggang Shen",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/ACCESS.2019.2904143",
language = "English",
volume = "7",
pages = "33728--33740",
journal = "IEEE Access",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Hippocampal Segmentation from Longitudinal Infant Brain MR Images via Classification-Guided Boundary Regression

AU - Shao, Yeqin

AU - Kim, Jaeil

AU - Gao, Yaozong

AU - Wang, Qian

AU - Lin, Weili

AU - Shen, Dinggang

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Hippocampal segmentation from infant brain MR images is indispensable for studying early brain development. However, most of the hippocampal segmentation methods were developed for population-based adult brain images, which are not suitable for longitudinal infant brain images acquired in the first year of life due to the low image contrast and variable development patterns of the hippocampal structure. To address these challenges, we propose a classification-guided boundary regression method to first detect hippocampal boundaries in the longitudinal infant brain images and then use those detected boundaries to guide the deformable model for final segmentation. Specifically, we first employ a classification-guided regression forest to predict the 3D displacements from individual image voxels to the potential hippocampal boundaries. These predicted displacements then determine the boundary maps by a voting strategy. Second, we iteratively enhance the voted hippocampal boundary map by incorporating the spatial context information given the tentative boundary estimation of the current time point. Besides, the longitudinal context information from all time points of the temporal sequence of the same subject (i.e., given their tentative segmentation results) is also utilized to facilitate accurate segmentation. Finally, a deformable model is applied to the enhanced voted boundary maps for achieving the longitudinal hippocampal segmentation. The experiments on infant brain MR images acquired from 2-week-old to 1-year-old show promising hippocampal segmentation results, indicating the applicability of our method in early brain development study.

AB - Hippocampal segmentation from infant brain MR images is indispensable for studying early brain development. However, most of the hippocampal segmentation methods were developed for population-based adult brain images, which are not suitable for longitudinal infant brain images acquired in the first year of life due to the low image contrast and variable development patterns of the hippocampal structure. To address these challenges, we propose a classification-guided boundary regression method to first detect hippocampal boundaries in the longitudinal infant brain images and then use those detected boundaries to guide the deformable model for final segmentation. Specifically, we first employ a classification-guided regression forest to predict the 3D displacements from individual image voxels to the potential hippocampal boundaries. These predicted displacements then determine the boundary maps by a voting strategy. Second, we iteratively enhance the voted hippocampal boundary map by incorporating the spatial context information given the tentative boundary estimation of the current time point. Besides, the longitudinal context information from all time points of the temporal sequence of the same subject (i.e., given their tentative segmentation results) is also utilized to facilitate accurate segmentation. Finally, a deformable model is applied to the enhanced voted boundary maps for achieving the longitudinal hippocampal segmentation. The experiments on infant brain MR images acquired from 2-week-old to 1-year-old show promising hippocampal segmentation results, indicating the applicability of our method in early brain development study.

KW - classification-guided boundary regression

KW - deformable segmentation

KW - hippocampal segmentation

KW - Infant brain

KW - longitudinal MR image

UR - http://www.scopus.com/inward/record.url?scp=85063870741&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063870741&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2019.2904143

DO - 10.1109/ACCESS.2019.2904143

M3 - Article

VL - 7

SP - 33728

EP - 33740

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

M1 - 8664424

ER -