MR prostate segmentation via distributed discriminative dictionary (DDD) learning

Yanrong Guo, Yiqiang Zhan, Yaozong Gao, Jianguo Jiang, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

Segmenting prostate from MR images is important yet challenging. Due to non-Gaussian distribution of prostate appearances in MR images, the popular active appearance model (AAM) has its limited performance. Although the newly developed sparse dictionary learning method[1, 2] can model the image appearance in a non-parametric fashion, the learned dictionaries still lack the discriminative power between prostate and non-prostate tissues, which is critical for accurate prostate segmentation. In this paper, we propose to integrate deformable model with a novel learning scheme, namely the Distributed Discriminative Dictionary (DDD) learning, which can capture image appearance in a non-parametric and discriminative fashion. In particular, three strategies are designed to boost the tissue discriminative power of DDD. First, minimum Redundancy Maximum Relevance (mRMR) feature selection is performed to constrain the dictionary learning in a discriminative feature space. Second, linear discriminant analysis (LDA) is employed to assemble residuals from different dictionaries for optimal separation between prostate and non-prostate tissues. Third, instead of learning the global dictionaries, we learn a set of local dictionaries for the local regions (each with small appearance variations) along prostate boundary, thus achieving better tissue differentiation locally. In the application stage, DDDs will provide the appearance cues to robustly drive the deformable model onto the prostate boundary. Experiments on 50 MR prostate images show that our method can yield a Dice Ratio of 88% compared to the manual segmentations, and have 7% improvement over the conventional AAM.

Original languageEnglish
Title of host publicationProceedings - International Symposium on Biomedical Imaging
Pages868-871
Number of pages4
DOIs
Publication statusPublished - 2013 Aug 22
Externally publishedYes
Event2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013 - San Francisco, CA, United States
Duration: 2013 Apr 72013 Apr 11

Other

Other2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013
CountryUnited States
CitySan Francisco, CA
Period13/4/713/4/11

Fingerprint

Glossaries
Prostate
Learning
Tissue
Dichlorodiphenyldichloroethane
Discriminant Analysis
Discriminant analysis
Cues
Redundancy
Feature extraction

Keywords

  • deformable segmentation
  • magnetic resonance image
  • Prostate segmentation
  • sparse dictionary learning

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Guo, Y., Zhan, Y., Gao, Y., Jiang, J., & Shen, D. (2013). MR prostate segmentation via distributed discriminative dictionary (DDD) learning. In Proceedings - International Symposium on Biomedical Imaging (pp. 868-871). [6556613] https://doi.org/10.1109/ISBI.2013.6556613

MR prostate segmentation via distributed discriminative dictionary (DDD) learning. / Guo, Yanrong; Zhan, Yiqiang; Gao, Yaozong; Jiang, Jianguo; Shen, Dinggang.

Proceedings - International Symposium on Biomedical Imaging. 2013. p. 868-871 6556613.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Guo, Y, Zhan, Y, Gao, Y, Jiang, J & Shen, D 2013, MR prostate segmentation via distributed discriminative dictionary (DDD) learning. in Proceedings - International Symposium on Biomedical Imaging., 6556613, pp. 868-871, 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013, San Francisco, CA, United States, 13/4/7. https://doi.org/10.1109/ISBI.2013.6556613
Guo Y, Zhan Y, Gao Y, Jiang J, Shen D. MR prostate segmentation via distributed discriminative dictionary (DDD) learning. In Proceedings - International Symposium on Biomedical Imaging. 2013. p. 868-871. 6556613 https://doi.org/10.1109/ISBI.2013.6556613
Guo, Yanrong ; Zhan, Yiqiang ; Gao, Yaozong ; Jiang, Jianguo ; Shen, Dinggang. / MR prostate segmentation via distributed discriminative dictionary (DDD) learning. Proceedings - International Symposium on Biomedical Imaging. 2013. pp. 868-871
@inproceedings{bbfe06c449ec45f7b28667c88634afbd,
title = "MR prostate segmentation via distributed discriminative dictionary (DDD) learning",
abstract = "Segmenting prostate from MR images is important yet challenging. Due to non-Gaussian distribution of prostate appearances in MR images, the popular active appearance model (AAM) has its limited performance. Although the newly developed sparse dictionary learning method[1, 2] can model the image appearance in a non-parametric fashion, the learned dictionaries still lack the discriminative power between prostate and non-prostate tissues, which is critical for accurate prostate segmentation. In this paper, we propose to integrate deformable model with a novel learning scheme, namely the Distributed Discriminative Dictionary (DDD) learning, which can capture image appearance in a non-parametric and discriminative fashion. In particular, three strategies are designed to boost the tissue discriminative power of DDD. First, minimum Redundancy Maximum Relevance (mRMR) feature selection is performed to constrain the dictionary learning in a discriminative feature space. Second, linear discriminant analysis (LDA) is employed to assemble residuals from different dictionaries for optimal separation between prostate and non-prostate tissues. Third, instead of learning the global dictionaries, we learn a set of local dictionaries for the local regions (each with small appearance variations) along prostate boundary, thus achieving better tissue differentiation locally. In the application stage, DDDs will provide the appearance cues to robustly drive the deformable model onto the prostate boundary. Experiments on 50 MR prostate images show that our method can yield a Dice Ratio of 88{\%} compared to the manual segmentations, and have 7{\%} improvement over the conventional AAM.",
keywords = "deformable segmentation, magnetic resonance image, Prostate segmentation, sparse dictionary learning",
author = "Yanrong Guo and Yiqiang Zhan and Yaozong Gao and Jianguo Jiang and Dinggang Shen",
year = "2013",
month = "8",
day = "22",
doi = "10.1109/ISBI.2013.6556613",
language = "English",
isbn = "9781467364546",
pages = "868--871",
booktitle = "Proceedings - International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - MR prostate segmentation via distributed discriminative dictionary (DDD) learning

AU - Guo, Yanrong

AU - Zhan, Yiqiang

AU - Gao, Yaozong

AU - Jiang, Jianguo

AU - Shen, Dinggang

PY - 2013/8/22

Y1 - 2013/8/22

N2 - Segmenting prostate from MR images is important yet challenging. Due to non-Gaussian distribution of prostate appearances in MR images, the popular active appearance model (AAM) has its limited performance. Although the newly developed sparse dictionary learning method[1, 2] can model the image appearance in a non-parametric fashion, the learned dictionaries still lack the discriminative power between prostate and non-prostate tissues, which is critical for accurate prostate segmentation. In this paper, we propose to integrate deformable model with a novel learning scheme, namely the Distributed Discriminative Dictionary (DDD) learning, which can capture image appearance in a non-parametric and discriminative fashion. In particular, three strategies are designed to boost the tissue discriminative power of DDD. First, minimum Redundancy Maximum Relevance (mRMR) feature selection is performed to constrain the dictionary learning in a discriminative feature space. Second, linear discriminant analysis (LDA) is employed to assemble residuals from different dictionaries for optimal separation between prostate and non-prostate tissues. Third, instead of learning the global dictionaries, we learn a set of local dictionaries for the local regions (each with small appearance variations) along prostate boundary, thus achieving better tissue differentiation locally. In the application stage, DDDs will provide the appearance cues to robustly drive the deformable model onto the prostate boundary. Experiments on 50 MR prostate images show that our method can yield a Dice Ratio of 88% compared to the manual segmentations, and have 7% improvement over the conventional AAM.

AB - Segmenting prostate from MR images is important yet challenging. Due to non-Gaussian distribution of prostate appearances in MR images, the popular active appearance model (AAM) has its limited performance. Although the newly developed sparse dictionary learning method[1, 2] can model the image appearance in a non-parametric fashion, the learned dictionaries still lack the discriminative power between prostate and non-prostate tissues, which is critical for accurate prostate segmentation. In this paper, we propose to integrate deformable model with a novel learning scheme, namely the Distributed Discriminative Dictionary (DDD) learning, which can capture image appearance in a non-parametric and discriminative fashion. In particular, three strategies are designed to boost the tissue discriminative power of DDD. First, minimum Redundancy Maximum Relevance (mRMR) feature selection is performed to constrain the dictionary learning in a discriminative feature space. Second, linear discriminant analysis (LDA) is employed to assemble residuals from different dictionaries for optimal separation between prostate and non-prostate tissues. Third, instead of learning the global dictionaries, we learn a set of local dictionaries for the local regions (each with small appearance variations) along prostate boundary, thus achieving better tissue differentiation locally. In the application stage, DDDs will provide the appearance cues to robustly drive the deformable model onto the prostate boundary. Experiments on 50 MR prostate images show that our method can yield a Dice Ratio of 88% compared to the manual segmentations, and have 7% improvement over the conventional AAM.

KW - deformable segmentation

KW - magnetic resonance image

KW - Prostate segmentation

KW - sparse dictionary learning

UR - http://www.scopus.com/inward/record.url?scp=84881633490&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881633490&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2013.6556613

DO - 10.1109/ISBI.2013.6556613

M3 - Conference contribution

AN - SCOPUS:84881633490

SN - 9781467364546

SP - 868

EP - 871

BT - Proceedings - International Symposium on Biomedical Imaging

ER -