Robust unsupervised tissue classification in MR images

Dzung L. Pham; Jerry L. Prince

Robust unsupervised tissue classification in MR images

Dzung L. Pham^*, Jerry L. Prince

^*Corresponding author for this work

Radiology and Bioengineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

10 Scopus citations

Abstract

A general framework for performing robust, unsupervised tissue classification in magnetic resonance images is presented. Tissue classification is formulated as an estimation problem based on an imaging model. Prior models are used within the estimation problem to compensate for noise and intensity inhomogeneity artifacts. From this framework, approaches based on K-means clustering, clustering via the expectation-maximization algorithm, and fuzzy clustering can be derived. The performance of the different types of approaches are evaluated using both simulated and real neuroimaging data.

Original language	English
Title of host publication	2004 2nd IEEE International Symposium on Biomedical Imaging
Subtitle of host publication	Macro to Nano
Pages	109-112
Number of pages	4
State	Published - 2004
Event	2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano - Arlington, VA, United States Duration: 15 Apr 2004 → 18 Apr 2004

Publication series

Name	2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano
Volume	1

Conference

Conference	2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano
Country/Territory	United States
City	Arlington, VA
Period	15/04/04 → 18/04/04

Cite this

@inproceedings{c4ab690f88554d27a700914edb0000a4,

title = "Robust unsupervised tissue classification in MR images",

abstract = "A general framework for performing robust, unsupervised tissue classification in magnetic resonance images is presented. Tissue classification is formulated as an estimation problem based on an imaging model. Prior models are used within the estimation problem to compensate for noise and intensity inhomogeneity artifacts. From this framework, approaches based on K-means clustering, clustering via the expectation-maximization algorithm, and fuzzy clustering can be derived. The performance of the different types of approaches are evaluated using both simulated and real neuroimaging data.",

author = "Pham, {Dzung L.} and Prince, {Jerry L.}",

year = "2004",

language = "English",

isbn = "0780383885",

series = "2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano",

pages = "109--112",

booktitle = "2004 2nd IEEE International Symposium on Biomedical Imaging",

note = "2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano ; Conference date: 15-04-2004 Through 18-04-2004",

}

Pham, DL & Prince, JL 2004, Robust unsupervised tissue classification in MR images. in 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano. 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano, vol. 1, pp. 109-112, 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano, Arlington, VA, United States, 15/04/04.

Robust unsupervised tissue classification in MR images. / Pham, Dzung L.; Prince, Jerry L.
2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano. 2004. p. 109-112 (2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Robust unsupervised tissue classification in MR images

AU - Pham, Dzung L.

AU - Prince, Jerry L.

PY - 2004

Y1 - 2004

N2 - A general framework for performing robust, unsupervised tissue classification in magnetic resonance images is presented. Tissue classification is formulated as an estimation problem based on an imaging model. Prior models are used within the estimation problem to compensate for noise and intensity inhomogeneity artifacts. From this framework, approaches based on K-means clustering, clustering via the expectation-maximization algorithm, and fuzzy clustering can be derived. The performance of the different types of approaches are evaluated using both simulated and real neuroimaging data.

AB - A general framework for performing robust, unsupervised tissue classification in magnetic resonance images is presented. Tissue classification is formulated as an estimation problem based on an imaging model. Prior models are used within the estimation problem to compensate for noise and intensity inhomogeneity artifacts. From this framework, approaches based on K-means clustering, clustering via the expectation-maximization algorithm, and fuzzy clustering can be derived. The performance of the different types of approaches are evaluated using both simulated and real neuroimaging data.

UR - http://www.scopus.com/inward/record.url?scp=17144391496&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:17144391496

SN - 0780383885

SN - 9780780383883

T3 - 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano

SP - 109

EP - 112

BT - 2004 2nd IEEE International Symposium on Biomedical Imaging

T2 - 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano

Y2 - 15 April 2004 through 18 April 2004

ER -