An EM Framework for Segmentation of Tissue Mixtures from Medical Images

Zhengrong Liang; Xiang Li; Daria Eremina; Lihong Li

An EM Framework for Segmentation of Tissue Mixtures from Medical Images

Zhengrong Liang
, Xiang Li
, Daria Eremina
, Lihong Li

Radiology

Stony Brook University

Research output: Contribution to journal › Conference article › peer-review

26 Scopus citations

Abstract

Image segmentation plays a major role in quantitative image analysis and computer aided detection (CAD) and diagnosis (CADx) for clinical applications. Conventional segmentation assigns a single label to each voxel, neglecting the partial volume (PV) effect. This work presents an EM (Expectation Maximization) framework for segmentation of tissue mixture in each voxel. Image data and tissue mixture models, EM algorithm for mixture quantification, prior model for regularization on the mixtures, and multi-spectral MR (magnetic resonance) data characterization are described in details. Preliminary results from CT (computed tomography) and MR images are reported to demonstrate its potential for clinical use.

Original language	English
Pages (from-to)	682-685
Number of pages	4
Journal	Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
Volume	1
State	Published - 2003
Event	A New Beginning for Human Health: Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Cancun, Mexico Duration: Sep 17 2003 → Sep 21 2003

Keywords

EM algorithm
Image segmentation
Maximum a posteriori (MAP) probability
Tissue mixture

Cite this

@article{4a3f373e79e14003938d5dc85f7ee157,

title = "An EM Framework for Segmentation of Tissue Mixtures from Medical Images",

abstract = "Image segmentation plays a major role in quantitative image analysis and computer aided detection (CAD) and diagnosis (CADx) for clinical applications. Conventional segmentation assigns a single label to each voxel, neglecting the partial volume (PV) effect. This work presents an EM (Expectation Maximization) framework for segmentation of tissue mixture in each voxel. Image data and tissue mixture models, EM algorithm for mixture quantification, prior model for regularization on the mixtures, and multi-spectral MR (magnetic resonance) data characterization are described in details. Preliminary results from CT (computed tomography) and MR images are reported to demonstrate its potential for clinical use.",

keywords = "EM algorithm, Image segmentation, Maximum a posteriori (MAP) probability, Tissue mixture",

author = "Zhengrong Liang and Xiang Li and Daria Eremina and Lihong Li",

year = "2003",

language = "English",

volume = "1",

pages = "682--685",

journal = "Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings",

issn = "0589-1019",

note = "A New Beginning for Human Health: Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society ; Conference date: 17-09-2003 Through 21-09-2003",

}

TY - JOUR

T1 - An EM Framework for Segmentation of Tissue Mixtures from Medical Images

AU - Liang, Zhengrong

AU - Li, Xiang

AU - Eremina, Daria

AU - Li, Lihong

PY - 2003

Y1 - 2003

N2 - Image segmentation plays a major role in quantitative image analysis and computer aided detection (CAD) and diagnosis (CADx) for clinical applications. Conventional segmentation assigns a single label to each voxel, neglecting the partial volume (PV) effect. This work presents an EM (Expectation Maximization) framework for segmentation of tissue mixture in each voxel. Image data and tissue mixture models, EM algorithm for mixture quantification, prior model for regularization on the mixtures, and multi-spectral MR (magnetic resonance) data characterization are described in details. Preliminary results from CT (computed tomography) and MR images are reported to demonstrate its potential for clinical use.

AB - Image segmentation plays a major role in quantitative image analysis and computer aided detection (CAD) and diagnosis (CADx) for clinical applications. Conventional segmentation assigns a single label to each voxel, neglecting the partial volume (PV) effect. This work presents an EM (Expectation Maximization) framework for segmentation of tissue mixture in each voxel. Image data and tissue mixture models, EM algorithm for mixture quantification, prior model for regularization on the mixtures, and multi-spectral MR (magnetic resonance) data characterization are described in details. Preliminary results from CT (computed tomography) and MR images are reported to demonstrate its potential for clinical use.

KW - EM algorithm

KW - Image segmentation

KW - Maximum a posteriori (MAP) probability

KW - Tissue mixture

UR - https://www.scopus.com/pages/publications/1542302599

M3 - Conference article

AN - SCOPUS:1542302599

SN - 0589-1019

VL - 1

SP - 682

EP - 685

JO - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

JF - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

T2 - A New Beginning for Human Health: Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Y2 - 17 September 2003 through 21 September 2003

ER -

An EM Framework for Segmentation of Tissue Mixtures from Medical Images

Abstract

Keywords

Other files and links

Fingerprint

Cite this