Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images: Method and Validation

Lihong Li, Hongbing Lu, Xiang Li, Wei Huang, Luminita (Alina) Tudorica, Chris Christodoulou, Lauren Krupp, Zhengrong Liang

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

Abstract

We present a fully automatic mixture-based algorithm for segmentation of brain tissues (white and gray matters - WM and GM), cerebral spinal fluid (CSF) and brain lesion to quantitatively analyze multiple sclerosis. The method performs intensity-based tissue classification using multispectral magnetic resonance (MR) images based on a stochastic model. With the existence of white Gaussian noise and spatially invariant blurring in acquired MR images, a Karhunen-Loeve (K-L) domain Wiener filter is applied for an accurate noise reduction and resolution restoration on blurred and noisy images to minimize the partial volume effect (PVE), which is a major limiting factor for the quantitative analysis. Following that, we utilize a Markov random field Gibbs model to integrate the local spatial information into the established expectation-maximization model-fitting algorithm. Each voxel is then classified by a mixture-based maximum a posterior (MAP) criterion, indicating its probabilities of belonging to each class, i.e., each voxel is labeled as a mixel with different tissue percentages, leading to further minimization of the PVE. The volumes of WM, GM and CSF are extracted from the mixture-based segmentation and the corresponding brain atrophies are computed. In this study, we have investigated the accuracy and repeatability of the algorithm with inclusion of noise analysis and point spread function for image resolution enhancement. Experimental results on both phantom and healthy volunteer studies are presented.

Original languageEnglish (US)
Title of host publicationIEEE Nuclear Science Symposium and Medical Imaging Conference
EditorsS. Metzler
Pages1361-1365
Number of pages5
Volume2
StatePublished - 2002
Externally publishedYes
Event2002 IEEE Nuclear Science Symposium Conference Record - Norfolk, VA, United States
Duration: Nov 10 2002Nov 16 2002

Other

Other2002 IEEE Nuclear Science Symposium Conference Record
CountryUnited States
CityNorfolk, VA
Period11/10/0211/16/02

Fingerprint

Volumetric analysis
Magnetic resonance
Brain
Tissue
Fluids
Optical transfer function
Stochastic models
Image resolution
Noise abatement
Restoration
Chemical analysis

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Industrial and Manufacturing Engineering

Cite this

Li, L., Lu, H., Li, X., Huang, W., Tudorica, L. A., Christodoulou, C., ... Liang, Z. (2002). Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images: Method and Validation. In S. Metzler (Ed.), IEEE Nuclear Science Symposium and Medical Imaging Conference (Vol. 2, pp. 1361-1365)

Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images : Method and Validation. / Li, Lihong; Lu, Hongbing; Li, Xiang; Huang, Wei; Tudorica, Luminita (Alina); Christodoulou, Chris; Krupp, Lauren; Liang, Zhengrong.

IEEE Nuclear Science Symposium and Medical Imaging Conference. ed. / S. Metzler. Vol. 2 2002. p. 1361-1365.

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

Li, L, Lu, H, Li, X, Huang, W, Tudorica, LA, Christodoulou, C, Krupp, L & Liang, Z 2002, Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images: Method and Validation. in S Metzler (ed.), IEEE Nuclear Science Symposium and Medical Imaging Conference. vol. 2, pp. 1361-1365, 2002 IEEE Nuclear Science Symposium Conference Record, Norfolk, VA, United States, 11/10/02.
Li L, Lu H, Li X, Huang W, Tudorica LA, Christodoulou C et al. Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images: Method and Validation. In Metzler S, editor, IEEE Nuclear Science Symposium and Medical Imaging Conference. Vol. 2. 2002. p. 1361-1365
Li, Lihong ; Lu, Hongbing ; Li, Xiang ; Huang, Wei ; Tudorica, Luminita (Alina) ; Christodoulou, Chris ; Krupp, Lauren ; Liang, Zhengrong. / Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images : Method and Validation. IEEE Nuclear Science Symposium and Medical Imaging Conference. editor / S. Metzler. Vol. 2 2002. pp. 1361-1365
@inproceedings{689add3e326f41ff9fca9a510e8ecec1,
title = "Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images: Method and Validation",
abstract = "We present a fully automatic mixture-based algorithm for segmentation of brain tissues (white and gray matters - WM and GM), cerebral spinal fluid (CSF) and brain lesion to quantitatively analyze multiple sclerosis. The method performs intensity-based tissue classification using multispectral magnetic resonance (MR) images based on a stochastic model. With the existence of white Gaussian noise and spatially invariant blurring in acquired MR images, a Karhunen-Loeve (K-L) domain Wiener filter is applied for an accurate noise reduction and resolution restoration on blurred and noisy images to minimize the partial volume effect (PVE), which is a major limiting factor for the quantitative analysis. Following that, we utilize a Markov random field Gibbs model to integrate the local spatial information into the established expectation-maximization model-fitting algorithm. Each voxel is then classified by a mixture-based maximum a posterior (MAP) criterion, indicating its probabilities of belonging to each class, i.e., each voxel is labeled as a mixel with different tissue percentages, leading to further minimization of the PVE. The volumes of WM, GM and CSF are extracted from the mixture-based segmentation and the corresponding brain atrophies are computed. In this study, we have investigated the accuracy and repeatability of the algorithm with inclusion of noise analysis and point spread function for image resolution enhancement. Experimental results on both phantom and healthy volunteer studies are presented.",
author = "Lihong Li and Hongbing Lu and Xiang Li and Wei Huang and Tudorica, {Luminita (Alina)} and Chris Christodoulou and Lauren Krupp and Zhengrong Liang",
year = "2002",
language = "English (US)",
volume = "2",
pages = "1361--1365",
editor = "S. Metzler",
booktitle = "IEEE Nuclear Science Symposium and Medical Imaging Conference",

}

TY - GEN

T1 - Volumetric Analysis of Multiple Sclerosis Using Multispectral MR Images

T2 - Method and Validation

AU - Li, Lihong

AU - Lu, Hongbing

AU - Li, Xiang

AU - Huang, Wei

AU - Tudorica, Luminita (Alina)

AU - Christodoulou, Chris

AU - Krupp, Lauren

AU - Liang, Zhengrong

PY - 2002

Y1 - 2002

N2 - We present a fully automatic mixture-based algorithm for segmentation of brain tissues (white and gray matters - WM and GM), cerebral spinal fluid (CSF) and brain lesion to quantitatively analyze multiple sclerosis. The method performs intensity-based tissue classification using multispectral magnetic resonance (MR) images based on a stochastic model. With the existence of white Gaussian noise and spatially invariant blurring in acquired MR images, a Karhunen-Loeve (K-L) domain Wiener filter is applied for an accurate noise reduction and resolution restoration on blurred and noisy images to minimize the partial volume effect (PVE), which is a major limiting factor for the quantitative analysis. Following that, we utilize a Markov random field Gibbs model to integrate the local spatial information into the established expectation-maximization model-fitting algorithm. Each voxel is then classified by a mixture-based maximum a posterior (MAP) criterion, indicating its probabilities of belonging to each class, i.e., each voxel is labeled as a mixel with different tissue percentages, leading to further minimization of the PVE. The volumes of WM, GM and CSF are extracted from the mixture-based segmentation and the corresponding brain atrophies are computed. In this study, we have investigated the accuracy and repeatability of the algorithm with inclusion of noise analysis and point spread function for image resolution enhancement. Experimental results on both phantom and healthy volunteer studies are presented.

AB - We present a fully automatic mixture-based algorithm for segmentation of brain tissues (white and gray matters - WM and GM), cerebral spinal fluid (CSF) and brain lesion to quantitatively analyze multiple sclerosis. The method performs intensity-based tissue classification using multispectral magnetic resonance (MR) images based on a stochastic model. With the existence of white Gaussian noise and spatially invariant blurring in acquired MR images, a Karhunen-Loeve (K-L) domain Wiener filter is applied for an accurate noise reduction and resolution restoration on blurred and noisy images to minimize the partial volume effect (PVE), which is a major limiting factor for the quantitative analysis. Following that, we utilize a Markov random field Gibbs model to integrate the local spatial information into the established expectation-maximization model-fitting algorithm. Each voxel is then classified by a mixture-based maximum a posterior (MAP) criterion, indicating its probabilities of belonging to each class, i.e., each voxel is labeled as a mixel with different tissue percentages, leading to further minimization of the PVE. The volumes of WM, GM and CSF are extracted from the mixture-based segmentation and the corresponding brain atrophies are computed. In this study, we have investigated the accuracy and repeatability of the algorithm with inclusion of noise analysis and point spread function for image resolution enhancement. Experimental results on both phantom and healthy volunteer studies are presented.

UR - http://www.scopus.com/inward/record.url?scp=0142179061&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0142179061&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0142179061

VL - 2

SP - 1361

EP - 1365

BT - IEEE Nuclear Science Symposium and Medical Imaging Conference

A2 - Metzler, S.

ER -