Segmentation of diffusion-weighted brain images using expectation maximization algorithm initialized by hierarchical clustering.

Chia-Feng Lu, Po-Shan Wang, Yen-Chun Chou, Hsiao-Chien Li, Bing-Wen Soong, Yu-Te Wu

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Tissue segmentation based on diffusion-weighted images (DWI) provides complementary information of tissue contrast to the structural MRI for facilitating the tissue segmentation. In the previous literatures, DWI-based brain tissue segmentation was carried out using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC). However, the information of directions of neural fibers was very limited in the parametric images. To fully utilize the directional information, we propose a novel method to perform tissue segmentation directly on the DWI raw image data. Specifically, a hierarchical clustering (HC) technique was first applied on the down-sampled data to initialize the model parameters for each tissue cluster followed by automatic segmentation using the expectation maximization (EM) algorithm. The whole brain DWI raw data of five normal subjects were analyzed. The results demonstrated that HC-EM is effective in multi-tissue classification on DWI raw data.
Original languageEnglish
Pages (from-to)5502-5505
Number of pages4
JournalConference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference
Publication statusPublished - 2008
Externally publishedYes

Fingerprint

Brain
Tissue
Magnetic resonance imaging
Anisotropy
Fibers

Keywords

  • aged
  • algorithm
  • article
  • artificial intelligence
  • automated pattern recognition
  • brain
  • cluster analysis
  • computer assisted diagnosis
  • diffusion weighted imaging
  • female
  • histology
  • human
  • image enhancement
  • male
  • methodology
  • middle aged
  • reproducibility
  • sensitivity and specificity
  • statistical model
  • three dimensional imaging
  • very elderly
  • Aged
  • Aged, 80 and over
  • Algorithms
  • Artificial Intelligence
  • Brain
  • Cluster Analysis
  • Diffusion Magnetic Resonance Imaging
  • Female
  • Humans
  • Image Enhancement
  • Image Interpretation, Computer-Assisted
  • Imaging, Three-Dimensional
  • Likelihood Functions
  • Male
  • Middle Aged
  • Pattern Recognition, Automated
  • Reproducibility of Results
  • Sensitivity and Specificity

Cite this

@article{28d0ea7dc9544ed8964170e83e3d8ee4,
title = "Segmentation of diffusion-weighted brain images using expectation maximization algorithm initialized by hierarchical clustering.",
abstract = "Tissue segmentation based on diffusion-weighted images (DWI) provides complementary information of tissue contrast to the structural MRI for facilitating the tissue segmentation. In the previous literatures, DWI-based brain tissue segmentation was carried out using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC). However, the information of directions of neural fibers was very limited in the parametric images. To fully utilize the directional information, we propose a novel method to perform tissue segmentation directly on the DWI raw image data. Specifically, a hierarchical clustering (HC) technique was first applied on the down-sampled data to initialize the model parameters for each tissue cluster followed by automatic segmentation using the expectation maximization (EM) algorithm. The whole brain DWI raw data of five normal subjects were analyzed. The results demonstrated that HC-EM is effective in multi-tissue classification on DWI raw data.",
keywords = "aged, algorithm, article, artificial intelligence, automated pattern recognition, brain, cluster analysis, computer assisted diagnosis, diffusion weighted imaging, female, histology, human, image enhancement, male, methodology, middle aged, reproducibility, sensitivity and specificity, statistical model, three dimensional imaging, very elderly, Aged, Aged, 80 and over, Algorithms, Artificial Intelligence, Brain, Cluster Analysis, Diffusion Magnetic Resonance Imaging, Female, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Imaging, Three-Dimensional, Likelihood Functions, Male, Middle Aged, Pattern Recognition, Automated, Reproducibility of Results, Sensitivity and Specificity",
author = "Chia-Feng Lu and Po-Shan Wang and Yen-Chun Chou and Hsiao-Chien Li and Bing-Wen Soong and Yu-Te Wu",
note = "Export Date: 31 March 2016 通訊地址: Lu, C.F.",
year = "2008",
language = "English",
pages = "5502--5505",
journal = "Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference",
issn = "1557-170X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Segmentation of diffusion-weighted brain images using expectation maximization algorithm initialized by hierarchical clustering.

AU - Lu, Chia-Feng

AU - Wang, Po-Shan

AU - Chou, Yen-Chun

AU - Li, Hsiao-Chien

AU - Soong, Bing-Wen

AU - Wu, Yu-Te

N1 - Export Date: 31 March 2016 通訊地址: Lu, C.F.

PY - 2008

Y1 - 2008

N2 - Tissue segmentation based on diffusion-weighted images (DWI) provides complementary information of tissue contrast to the structural MRI for facilitating the tissue segmentation. In the previous literatures, DWI-based brain tissue segmentation was carried out using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC). However, the information of directions of neural fibers was very limited in the parametric images. To fully utilize the directional information, we propose a novel method to perform tissue segmentation directly on the DWI raw image data. Specifically, a hierarchical clustering (HC) technique was first applied on the down-sampled data to initialize the model parameters for each tissue cluster followed by automatic segmentation using the expectation maximization (EM) algorithm. The whole brain DWI raw data of five normal subjects were analyzed. The results demonstrated that HC-EM is effective in multi-tissue classification on DWI raw data.

AB - Tissue segmentation based on diffusion-weighted images (DWI) provides complementary information of tissue contrast to the structural MRI for facilitating the tissue segmentation. In the previous literatures, DWI-based brain tissue segmentation was carried out using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC). However, the information of directions of neural fibers was very limited in the parametric images. To fully utilize the directional information, we propose a novel method to perform tissue segmentation directly on the DWI raw image data. Specifically, a hierarchical clustering (HC) technique was first applied on the down-sampled data to initialize the model parameters for each tissue cluster followed by automatic segmentation using the expectation maximization (EM) algorithm. The whole brain DWI raw data of five normal subjects were analyzed. The results demonstrated that HC-EM is effective in multi-tissue classification on DWI raw data.

KW - aged

KW - algorithm

KW - article

KW - artificial intelligence

KW - automated pattern recognition

KW - brain

KW - cluster analysis

KW - computer assisted diagnosis

KW - diffusion weighted imaging

KW - female

KW - histology

KW - human

KW - image enhancement

KW - male

KW - methodology

KW - middle aged

KW - reproducibility

KW - sensitivity and specificity

KW - statistical model

KW - three dimensional imaging

KW - very elderly

KW - Aged

KW - Aged, 80 and over

KW - Algorithms

KW - Artificial Intelligence

KW - Brain

KW - Cluster Analysis

KW - Diffusion Magnetic Resonance Imaging

KW - Female

KW - Humans

KW - Image Enhancement

KW - Image Interpretation, Computer-Assisted

KW - Imaging, Three-Dimensional

KW - Likelihood Functions

KW - Male

KW - Middle Aged

KW - Pattern Recognition, Automated

KW - Reproducibility of Results

KW - Sensitivity and Specificity

UR - http://www.scopus.com/inward/record.url?scp=84903855344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84903855344&partnerID=8YFLogxK

M3 - Article

SP - 5502

EP - 5505

JO - Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference

JF - Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference

SN - 1557-170X

ER -