Robust brain ROI segmentation by deformation regression and deformable shape model

Zhengwang Wu, Yanrong Guo, Sang Hyun Park, Yaozong Gao, Pei Dong, Seong Whan Lee, Dinggang Shen

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

We propose a robust and efficient learning-based deformable model for segmenting regions of interest (ROIs) from structural MR brain images. Different from the conventional deformable-model-based methods that deform a shape model locally around the initialization location, we learn an image-based regressor to guide the deformable model to fit for the target ROI. Specifically, given any voxel in a new image, the image-based regressor can predict the displacement vector from this voxel towards the boundary of target ROI, which can be used to guide the deformable segmentation. By predicting the displacement vector maps for the whole image, our deformable model is able to use multiple non-boundary predictions to jointly determine and iteratively converge the initial shape model to the target ROI boundary, which is more robust to the local prediction error and initialization. In addition, by introducing the prior shape model, our segmentation avoids the isolated segmentations as often occurred in the previous multi-atlas-based methods. In order to learn an image-based regressor for displacement vector prediction, we adopt the following novel strategies in the learning procedure: (1) a joint classification and regression random forest is proposed to learn an image-based regressor together with an ROI classifier in a multi-task manner; (2) high-level context features are extracted from intermediate (estimated) displacement vector and classification maps to enforce the relationship between predicted displacement vectors at neighboring voxels. To validate our method, we compare it with the state-of-the-art multi-atlas-based methods and other learning-based methods on three public brain MR datasets. The results consistently show that our method is better in terms of both segmentation accuracy and computational efficiency.

Original languageEnglish
Pages (from-to)198-213
Number of pages16
JournalMedical Image Analysis
Volume43
DOIs
Publication statusPublished - 2018 Jan 1

Fingerprint

Brain
Atlases
Learning
Computational efficiency
Classifiers
Joints
Efficiency

Keywords

  • Auto-context model
  • Brain ROI segmentation
  • Deformable model
  • Joint classification and regression random forest

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Graphics and Computer-Aided Design

Cite this

Robust brain ROI segmentation by deformation regression and deformable shape model. / Wu, Zhengwang; Guo, Yanrong; Park, Sang Hyun; Gao, Yaozong; Dong, Pei; Lee, Seong Whan; Shen, Dinggang.

In: Medical Image Analysis, Vol. 43, 01.01.2018, p. 198-213.

Research output: Contribution to journalArticle

Wu, Zhengwang ; Guo, Yanrong ; Park, Sang Hyun ; Gao, Yaozong ; Dong, Pei ; Lee, Seong Whan ; Shen, Dinggang. / Robust brain ROI segmentation by deformation regression and deformable shape model. In: Medical Image Analysis. 2018 ; Vol. 43. pp. 198-213.
@article{409b3326b79f4daaa6014d9c50eb75a3,
title = "Robust brain ROI segmentation by deformation regression and deformable shape model",
abstract = "We propose a robust and efficient learning-based deformable model for segmenting regions of interest (ROIs) from structural MR brain images. Different from the conventional deformable-model-based methods that deform a shape model locally around the initialization location, we learn an image-based regressor to guide the deformable model to fit for the target ROI. Specifically, given any voxel in a new image, the image-based regressor can predict the displacement vector from this voxel towards the boundary of target ROI, which can be used to guide the deformable segmentation. By predicting the displacement vector maps for the whole image, our deformable model is able to use multiple non-boundary predictions to jointly determine and iteratively converge the initial shape model to the target ROI boundary, which is more robust to the local prediction error and initialization. In addition, by introducing the prior shape model, our segmentation avoids the isolated segmentations as often occurred in the previous multi-atlas-based methods. In order to learn an image-based regressor for displacement vector prediction, we adopt the following novel strategies in the learning procedure: (1) a joint classification and regression random forest is proposed to learn an image-based regressor together with an ROI classifier in a multi-task manner; (2) high-level context features are extracted from intermediate (estimated) displacement vector and classification maps to enforce the relationship between predicted displacement vectors at neighboring voxels. To validate our method, we compare it with the state-of-the-art multi-atlas-based methods and other learning-based methods on three public brain MR datasets. The results consistently show that our method is better in terms of both segmentation accuracy and computational efficiency.",
keywords = "Auto-context model, Brain ROI segmentation, Deformable model, Joint classification and regression random forest",
author = "Zhengwang Wu and Yanrong Guo and Park, {Sang Hyun} and Yaozong Gao and Pei Dong and Lee, {Seong Whan} and Dinggang Shen",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.media.2017.11.001",
language = "English",
volume = "43",
pages = "198--213",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",

}

TY - JOUR

T1 - Robust brain ROI segmentation by deformation regression and deformable shape model

AU - Wu, Zhengwang

AU - Guo, Yanrong

AU - Park, Sang Hyun

AU - Gao, Yaozong

AU - Dong, Pei

AU - Lee, Seong Whan

AU - Shen, Dinggang

PY - 2018/1/1

Y1 - 2018/1/1

N2 - We propose a robust and efficient learning-based deformable model for segmenting regions of interest (ROIs) from structural MR brain images. Different from the conventional deformable-model-based methods that deform a shape model locally around the initialization location, we learn an image-based regressor to guide the deformable model to fit for the target ROI. Specifically, given any voxel in a new image, the image-based regressor can predict the displacement vector from this voxel towards the boundary of target ROI, which can be used to guide the deformable segmentation. By predicting the displacement vector maps for the whole image, our deformable model is able to use multiple non-boundary predictions to jointly determine and iteratively converge the initial shape model to the target ROI boundary, which is more robust to the local prediction error and initialization. In addition, by introducing the prior shape model, our segmentation avoids the isolated segmentations as often occurred in the previous multi-atlas-based methods. In order to learn an image-based regressor for displacement vector prediction, we adopt the following novel strategies in the learning procedure: (1) a joint classification and regression random forest is proposed to learn an image-based regressor together with an ROI classifier in a multi-task manner; (2) high-level context features are extracted from intermediate (estimated) displacement vector and classification maps to enforce the relationship between predicted displacement vectors at neighboring voxels. To validate our method, we compare it with the state-of-the-art multi-atlas-based methods and other learning-based methods on three public brain MR datasets. The results consistently show that our method is better in terms of both segmentation accuracy and computational efficiency.

AB - We propose a robust and efficient learning-based deformable model for segmenting regions of interest (ROIs) from structural MR brain images. Different from the conventional deformable-model-based methods that deform a shape model locally around the initialization location, we learn an image-based regressor to guide the deformable model to fit for the target ROI. Specifically, given any voxel in a new image, the image-based regressor can predict the displacement vector from this voxel towards the boundary of target ROI, which can be used to guide the deformable segmentation. By predicting the displacement vector maps for the whole image, our deformable model is able to use multiple non-boundary predictions to jointly determine and iteratively converge the initial shape model to the target ROI boundary, which is more robust to the local prediction error and initialization. In addition, by introducing the prior shape model, our segmentation avoids the isolated segmentations as often occurred in the previous multi-atlas-based methods. In order to learn an image-based regressor for displacement vector prediction, we adopt the following novel strategies in the learning procedure: (1) a joint classification and regression random forest is proposed to learn an image-based regressor together with an ROI classifier in a multi-task manner; (2) high-level context features are extracted from intermediate (estimated) displacement vector and classification maps to enforce the relationship between predicted displacement vectors at neighboring voxels. To validate our method, we compare it with the state-of-the-art multi-atlas-based methods and other learning-based methods on three public brain MR datasets. The results consistently show that our method is better in terms of both segmentation accuracy and computational efficiency.

KW - Auto-context model

KW - Brain ROI segmentation

KW - Deformable model

KW - Joint classification and regression random forest

UR - http://www.scopus.com/inward/record.url?scp=85034095050&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85034095050&partnerID=8YFLogxK

U2 - 10.1016/j.media.2017.11.001

DO - 10.1016/j.media.2017.11.001

M3 - Article

C2 - 29149715

AN - SCOPUS:85034095050

VL - 43

SP - 198

EP - 213

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

ER -