Automatic prostate MR image segmentation with sparse label propagation and domain-specific manifold regularization.

Shu Liao, Yaozong Gao, Yinghuan Shi, Ambereen Yousuf, Ibrahim Karademir, Aytekin Oto, Dinggang Shen

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Automatic prostate segmentation in MR images plays an important role in prostate cancer diagnosis. However, there are two main challenges: (1) Large inter-subject prostate shape variations; (2) Inhomogeneous prostate appearance. To address these challenges, we propose a new hierarchical prostate MR segmentation method, with the main contributions lying in the following aspects: First, the most salient features are learnt from atlases based on a subclass discriminant analysis (SDA) method, which aims to find a discriminant feature subspace by simultaneously maximizing the inter-class distance and minimizing the intra-class variations. The projected features, instead of only voxel-wise intensity, will be served as anatomical signature of each voxel. Second, based on the projected features, a new multi-atlases sparse label fusion framework is proposed to estimate the prostate likelihood of each voxel in the target image from the coarse level. Third, a domain-specific semi-supervised manifold regularization method is proposed to incorporate the most reliable patient-specific information identified by the prostate likelihood map to refine the segmentation result from the fine level. Our method is evaluated on a T2 weighted prostate MR image dataset consisting of 66 patients and compared with two state-of-the-art segmentation methods. Experimental results show that our method consistently achieves the highest segmentation accuracies than other methods under comparison.

Original languageEnglish
Pages (from-to)511-523
Number of pages13
JournalInformation processing in medical imaging : proceedings of the ... conference
Volume23
Publication statusPublished - 2013 Jan 1

Fingerprint

Prostate
Atlases
Discriminant Analysis
Prostatic Neoplasms

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Automatic prostate MR image segmentation with sparse label propagation and domain-specific manifold regularization. / Liao, Shu; Gao, Yaozong; Shi, Yinghuan; Yousuf, Ambereen; Karademir, Ibrahim; Oto, Aytekin; Shen, Dinggang.

In: Information processing in medical imaging : proceedings of the ... conference, Vol. 23, 01.01.2013, p. 511-523.

Research output: Contribution to journalArticle

@article{6fe429ba150e40a790f31eaea46af449,
title = "Automatic prostate MR image segmentation with sparse label propagation and domain-specific manifold regularization.",
abstract = "Automatic prostate segmentation in MR images plays an important role in prostate cancer diagnosis. However, there are two main challenges: (1) Large inter-subject prostate shape variations; (2) Inhomogeneous prostate appearance. To address these challenges, we propose a new hierarchical prostate MR segmentation method, with the main contributions lying in the following aspects: First, the most salient features are learnt from atlases based on a subclass discriminant analysis (SDA) method, which aims to find a discriminant feature subspace by simultaneously maximizing the inter-class distance and minimizing the intra-class variations. The projected features, instead of only voxel-wise intensity, will be served as anatomical signature of each voxel. Second, based on the projected features, a new multi-atlases sparse label fusion framework is proposed to estimate the prostate likelihood of each voxel in the target image from the coarse level. Third, a domain-specific semi-supervised manifold regularization method is proposed to incorporate the most reliable patient-specific information identified by the prostate likelihood map to refine the segmentation result from the fine level. Our method is evaluated on a T2 weighted prostate MR image dataset consisting of 66 patients and compared with two state-of-the-art segmentation methods. Experimental results show that our method consistently achieves the highest segmentation accuracies than other methods under comparison.",
author = "Shu Liao and Yaozong Gao and Yinghuan Shi and Ambereen Yousuf and Ibrahim Karademir and Aytekin Oto and Dinggang Shen",
year = "2013",
month = "1",
day = "1",
language = "English",
volume = "23",
pages = "511--523",
journal = "Information processing in medical imaging : proceedings of the ... conference",
issn = "1011-2499",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Automatic prostate MR image segmentation with sparse label propagation and domain-specific manifold regularization.

AU - Liao, Shu

AU - Gao, Yaozong

AU - Shi, Yinghuan

AU - Yousuf, Ambereen

AU - Karademir, Ibrahim

AU - Oto, Aytekin

AU - Shen, Dinggang

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Automatic prostate segmentation in MR images plays an important role in prostate cancer diagnosis. However, there are two main challenges: (1) Large inter-subject prostate shape variations; (2) Inhomogeneous prostate appearance. To address these challenges, we propose a new hierarchical prostate MR segmentation method, with the main contributions lying in the following aspects: First, the most salient features are learnt from atlases based on a subclass discriminant analysis (SDA) method, which aims to find a discriminant feature subspace by simultaneously maximizing the inter-class distance and minimizing the intra-class variations. The projected features, instead of only voxel-wise intensity, will be served as anatomical signature of each voxel. Second, based on the projected features, a new multi-atlases sparse label fusion framework is proposed to estimate the prostate likelihood of each voxel in the target image from the coarse level. Third, a domain-specific semi-supervised manifold regularization method is proposed to incorporate the most reliable patient-specific information identified by the prostate likelihood map to refine the segmentation result from the fine level. Our method is evaluated on a T2 weighted prostate MR image dataset consisting of 66 patients and compared with two state-of-the-art segmentation methods. Experimental results show that our method consistently achieves the highest segmentation accuracies than other methods under comparison.

AB - Automatic prostate segmentation in MR images plays an important role in prostate cancer diagnosis. However, there are two main challenges: (1) Large inter-subject prostate shape variations; (2) Inhomogeneous prostate appearance. To address these challenges, we propose a new hierarchical prostate MR segmentation method, with the main contributions lying in the following aspects: First, the most salient features are learnt from atlases based on a subclass discriminant analysis (SDA) method, which aims to find a discriminant feature subspace by simultaneously maximizing the inter-class distance and minimizing the intra-class variations. The projected features, instead of only voxel-wise intensity, will be served as anatomical signature of each voxel. Second, based on the projected features, a new multi-atlases sparse label fusion framework is proposed to estimate the prostate likelihood of each voxel in the target image from the coarse level. Third, a domain-specific semi-supervised manifold regularization method is proposed to incorporate the most reliable patient-specific information identified by the prostate likelihood map to refine the segmentation result from the fine level. Our method is evaluated on a T2 weighted prostate MR image dataset consisting of 66 patients and compared with two state-of-the-art segmentation methods. Experimental results show that our method consistently achieves the highest segmentation accuracies than other methods under comparison.

UR - http://www.scopus.com/inward/record.url?scp=84901279625&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901279625&partnerID=8YFLogxK

M3 - Article

C2 - 24683995

AN - SCOPUS:84901279625

VL - 23

SP - 511

EP - 523

JO - Information processing in medical imaging : proceedings of the ... conference

JF - Information processing in medical imaging : proceedings of the ... conference

SN - 1011-2499

ER -