Fusing speed and phase information for vascular segmentation of phase contrast MR angiograms

Albert C S Chung, J. Alison Noble, Paul Summers

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

This paper presents a statistical approach to aggregating speed and phase (directional) information for vascular segmentation of phase contrast magnetic resonance angiograms (PC-MRA). Rather than relying on speed information alone, as done by others and in our own work, we demonstrate that including phase information as a priori knowledge in a Markov random field (MRF) model can improve the quality of segmentation. This is particularly true in the region within an aneurysm where there is a heterogeneous intensity pattern and significant vascular signal loss. We propose to use a Maxwell-Gaussian mixture density to model the background signal distribution and combine this with a uniform distribution for modelling vascular signal to give a Maxwell-Gaussian-uniform (MGU) mixture model of image intensity. The MGU model parameters are estimated by the modified expectation-maximisation (EM) algorithm. In addition, it is shown that the Maxwell-Gaussian mixture distribution (a) models the background signal more accurately than a Maxwell distribution, (b) exhibits a better fit to clinical data and (c) gives fewer false positive voxels (misclassified vessel voxels) in segmentation. The new segmentation algorithm is tested on an aneurysm phantom data set and two clinical data sets. The experimental results show that the proposed method can provide a better quality of segmentation when both speed and phase information are utilised.

Original languageEnglish
Pages (from-to)109-128
Number of pages20
JournalMedical Image Analysis
Volume6
Issue number2
DOIs
Publication statusPublished - Jun 2002

Fingerprint

Blood Vessels
Angiography
Aneurysm
Normal Distribution
Magnetic Resonance Spectroscopy
Magnetic resonance
Datasets

Keywords

  • Expectation-maximisation (EM) algorithm
  • Flow coherence
  • Magnetic resonance angiography (MRA)
  • Markov random fields (MRF)
  • MR signal modelling and statistical segmentation

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Fusing speed and phase information for vascular segmentation of phase contrast MR angiograms. / Chung, Albert C S; Noble, J. Alison; Summers, Paul.

In: Medical Image Analysis, Vol. 6, No. 2, 06.2002, p. 109-128.

Research output: Contribution to journalArticle

@article{cc87feedb369426bb7a5b93cca1cbb2b,
title = "Fusing speed and phase information for vascular segmentation of phase contrast MR angiograms",
abstract = "This paper presents a statistical approach to aggregating speed and phase (directional) information for vascular segmentation of phase contrast magnetic resonance angiograms (PC-MRA). Rather than relying on speed information alone, as done by others and in our own work, we demonstrate that including phase information as a priori knowledge in a Markov random field (MRF) model can improve the quality of segmentation. This is particularly true in the region within an aneurysm where there is a heterogeneous intensity pattern and significant vascular signal loss. We propose to use a Maxwell-Gaussian mixture density to model the background signal distribution and combine this with a uniform distribution for modelling vascular signal to give a Maxwell-Gaussian-uniform (MGU) mixture model of image intensity. The MGU model parameters are estimated by the modified expectation-maximisation (EM) algorithm. In addition, it is shown that the Maxwell-Gaussian mixture distribution (a) models the background signal more accurately than a Maxwell distribution, (b) exhibits a better fit to clinical data and (c) gives fewer false positive voxels (misclassified vessel voxels) in segmentation. The new segmentation algorithm is tested on an aneurysm phantom data set and two clinical data sets. The experimental results show that the proposed method can provide a better quality of segmentation when both speed and phase information are utilised.",
keywords = "Expectation-maximisation (EM) algorithm, Flow coherence, Magnetic resonance angiography (MRA), Markov random fields (MRF), MR signal modelling and statistical segmentation",
author = "Chung, {Albert C S} and Noble, {J. Alison} and Paul Summers",
year = "2002",
month = "6",
doi = "10.1016/S1361-8415(02)00057-9",
language = "English",
volume = "6",
pages = "109--128",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",
number = "2",

}

TY - JOUR

T1 - Fusing speed and phase information for vascular segmentation of phase contrast MR angiograms

AU - Chung, Albert C S

AU - Noble, J. Alison

AU - Summers, Paul

PY - 2002/6

Y1 - 2002/6

N2 - This paper presents a statistical approach to aggregating speed and phase (directional) information for vascular segmentation of phase contrast magnetic resonance angiograms (PC-MRA). Rather than relying on speed information alone, as done by others and in our own work, we demonstrate that including phase information as a priori knowledge in a Markov random field (MRF) model can improve the quality of segmentation. This is particularly true in the region within an aneurysm where there is a heterogeneous intensity pattern and significant vascular signal loss. We propose to use a Maxwell-Gaussian mixture density to model the background signal distribution and combine this with a uniform distribution for modelling vascular signal to give a Maxwell-Gaussian-uniform (MGU) mixture model of image intensity. The MGU model parameters are estimated by the modified expectation-maximisation (EM) algorithm. In addition, it is shown that the Maxwell-Gaussian mixture distribution (a) models the background signal more accurately than a Maxwell distribution, (b) exhibits a better fit to clinical data and (c) gives fewer false positive voxels (misclassified vessel voxels) in segmentation. The new segmentation algorithm is tested on an aneurysm phantom data set and two clinical data sets. The experimental results show that the proposed method can provide a better quality of segmentation when both speed and phase information are utilised.

AB - This paper presents a statistical approach to aggregating speed and phase (directional) information for vascular segmentation of phase contrast magnetic resonance angiograms (PC-MRA). Rather than relying on speed information alone, as done by others and in our own work, we demonstrate that including phase information as a priori knowledge in a Markov random field (MRF) model can improve the quality of segmentation. This is particularly true in the region within an aneurysm where there is a heterogeneous intensity pattern and significant vascular signal loss. We propose to use a Maxwell-Gaussian mixture density to model the background signal distribution and combine this with a uniform distribution for modelling vascular signal to give a Maxwell-Gaussian-uniform (MGU) mixture model of image intensity. The MGU model parameters are estimated by the modified expectation-maximisation (EM) algorithm. In addition, it is shown that the Maxwell-Gaussian mixture distribution (a) models the background signal more accurately than a Maxwell distribution, (b) exhibits a better fit to clinical data and (c) gives fewer false positive voxels (misclassified vessel voxels) in segmentation. The new segmentation algorithm is tested on an aneurysm phantom data set and two clinical data sets. The experimental results show that the proposed method can provide a better quality of segmentation when both speed and phase information are utilised.

KW - Expectation-maximisation (EM) algorithm

KW - Flow coherence

KW - Magnetic resonance angiography (MRA)

KW - Markov random fields (MRF)

KW - MR signal modelling and statistical segmentation

UR - http://www.scopus.com/inward/record.url?scp=0036624536&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036624536&partnerID=8YFLogxK

U2 - 10.1016/S1361-8415(02)00057-9

DO - 10.1016/S1361-8415(02)00057-9

M3 - Article

C2 - 12044999

AN - SCOPUS:0036624536

VL - 6

SP - 109

EP - 128

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

IS - 2

ER -