Multi-tissue Classification of Diffusion-Weighted Brain Images in Multiple System Atrophy Using Expectation Maximization Algorithm Initialized by Hierarchical Clustering

13th International Conference on Biomedical Engineering, ICBME 2008

Chia-Feng Lu, Po-Shan Wang, Bing-Wen Soong, Yen-Chun Chou, Hsiao-Chien Li, Yu-Te Wu, AMTI; BES Technology; DELSYS; VICON; INSTRON; et al

Research output: Contribution to conferenceOther

1 Citation (Scopus)

Abstract

Multiple system atrophy (MSA) is a well-known neurodegenerative disorders that present parkinsonism syndrome and autonomic dysfunction. Patients with MSA who have the combination of parkinsonism and cerebellar ataxia are referred to as MSA-C. Brain diffusion-weighted imaging (DWI) offers the potential for objective criteria in the diagnosis of MSA. We aim to develop an automatic method to segment out the abnormal whole brain area in MSA-C patients based on the 13-direction DWI raw data. The whole brain DWI raw data of fifteen normal subjects and nine MSA-C patients were analyzed. In this study, we proposed a novel method to perform tissue segmentation directly based on the directional information of the DWI images, rather than using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC) as in the previous literatures. 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. Our results demonstrate that the HC-EM is effective in multi-tissue classification, namely, the cerebrospinal fluid, gray matter, and several areas of white matters, on the DWI raw data. The segmented patterns and the corresponding intensities of thirteen directions of the cerebellum in MSA-C patients showed the decrease of the anisotropy, which were evidently different from the results in normal subjects.
Original languageEnglish
Pages722-725
Number of pages4
DOIs
Publication statusPublished - 2009
Externally publishedYes

Fingerprint

Biomedical engineering
Brain
Tissue
Imaging techniques
Anisotropy
Cerebrospinal fluid

Keywords

  • diffusion-weighted imaging
  • expectation maximization algorithm
  • hierarchical clustering
  • Multiple system atrophy
  • Apparent diffusion coefficient
  • Automatic segmentations
  • Diffusion weighted imaging
  • Directional information
  • Expectation-maximization algorithms
  • Hier-archical clustering
  • Multiple system atrophies
  • Neurodegenerative disorders
  • Anisotropy
  • Biomedical engineering
  • Brain
  • Brain mapping
  • Cerebrospinal fluid
  • Clustering algorithms
  • Diffusion
  • Maximum principle
  • Neurodegenerative diseases
  • Tissue
  • Image segmentation

Cite this

Multi-tissue Classification of Diffusion-Weighted Brain Images in Multiple System Atrophy Using Expectation Maximization Algorithm Initialized by Hierarchical Clustering : 13th International Conference on Biomedical Engineering, ICBME 2008. / Lu, Chia-Feng; Wang, Po-Shan; Soong, Bing-Wen; Chou, Yen-Chun; Li, Hsiao-Chien; Wu, Yu-Te; al, AMTI; BES Technology; DELSYS; VICON; INSTRON; et.

2009. 722-725.

Research output: Contribution to conferenceOther

@conference{7fe1117a9a774e42adf4d9bc6f7ad22b,
title = "Multi-tissue Classification of Diffusion-Weighted Brain Images in Multiple System Atrophy Using Expectation Maximization Algorithm Initialized by Hierarchical Clustering: 13th International Conference on Biomedical Engineering, ICBME 2008",
abstract = "Multiple system atrophy (MSA) is a well-known neurodegenerative disorders that present parkinsonism syndrome and autonomic dysfunction. Patients with MSA who have the combination of parkinsonism and cerebellar ataxia are referred to as MSA-C. Brain diffusion-weighted imaging (DWI) offers the potential for objective criteria in the diagnosis of MSA. We aim to develop an automatic method to segment out the abnormal whole brain area in MSA-C patients based on the 13-direction DWI raw data. The whole brain DWI raw data of fifteen normal subjects and nine MSA-C patients were analyzed. In this study, we proposed a novel method to perform tissue segmentation directly based on the directional information of the DWI images, rather than using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC) as in the previous literatures. 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. Our results demonstrate that the HC-EM is effective in multi-tissue classification, namely, the cerebrospinal fluid, gray matter, and several areas of white matters, on the DWI raw data. The segmented patterns and the corresponding intensities of thirteen directions of the cerebellum in MSA-C patients showed the decrease of the anisotropy, which were evidently different from the results in normal subjects.",
keywords = "diffusion-weighted imaging, expectation maximization algorithm, hierarchical clustering, Multiple system atrophy, Apparent diffusion coefficient, Automatic segmentations, Diffusion weighted imaging, Directional information, Expectation-maximization algorithms, Hier-archical clustering, Multiple system atrophies, Neurodegenerative disorders, Anisotropy, Biomedical engineering, Brain, Brain mapping, Cerebrospinal fluid, Clustering algorithms, Diffusion, Maximum principle, Neurodegenerative diseases, Tissue, Image segmentation",
author = "Chia-Feng Lu and Po-Shan Wang and Bing-Wen Soong and Yen-Chun Chou and Hsiao-Chien Li and Yu-Te Wu and al, {AMTI; BES Technology; DELSYS; VICON; INSTRON; et}",
note = "會議代碼: 101867 Export Date: 31 March 2016 通訊地址: Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, No.155, Sec.2, Linong Street, Taipei, Taiwan 參考文獻: Liu, T., Li, H., Wong, K., Tarokh, A., Guo, L., Wonga, T.C., Brain tissue segmentation based on DWI data (2007) NeuroImage, 38, pp. 114-123; Wishart, D., An algorithm for hierarchical classifications (1969) Biometrics, 25, pp. 165-170; Wu, Y.T., Chou, Y.C., Guo, W.Y., Yeh, T.C., Hsieh, J.C., Classification of spatiotemporal hemodynamics from brain perfusion MR images using expectation-Maximization estimation with finite mixture of multivariate Gaussian distributions (2007) Magnetic Resonance in Medicine, 57, pp. 181-191; Schwarz, G., Estimating the dimension of a model (1978) Ann. Statist., 6, pp. 461-464; Otsu, N., A threshold selection method from gray-level histogram (1979) IEEE Transactions on Systems, Man, and Cybernetics; Jiang, H., Kim, J., Pearlson, G.D., Mori, S., DtiStudio: Resource program for diffusion tensor computation and fiber bundle tracking (2006) Computer Methods and Programs in Biomedicine, 81, pp. 106-116",
year = "2009",
doi = "10.1007/978-3-540-92841-6_177",
language = "English",
pages = "722--725",

}

TY - CONF

T1 - Multi-tissue Classification of Diffusion-Weighted Brain Images in Multiple System Atrophy Using Expectation Maximization Algorithm Initialized by Hierarchical Clustering

T2 - 13th International Conference on Biomedical Engineering, ICBME 2008

AU - Lu, Chia-Feng

AU - Wang, Po-Shan

AU - Soong, Bing-Wen

AU - Chou, Yen-Chun

AU - Li, Hsiao-Chien

AU - Wu, Yu-Te

AU - al, AMTI; BES Technology; DELSYS; VICON; INSTRON; et

N1 - 會議代碼: 101867 Export Date: 31 March 2016 通訊地址: Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, No.155, Sec.2, Linong Street, Taipei, Taiwan 參考文獻: Liu, T., Li, H., Wong, K., Tarokh, A., Guo, L., Wonga, T.C., Brain tissue segmentation based on DWI data (2007) NeuroImage, 38, pp. 114-123; Wishart, D., An algorithm for hierarchical classifications (1969) Biometrics, 25, pp. 165-170; Wu, Y.T., Chou, Y.C., Guo, W.Y., Yeh, T.C., Hsieh, J.C., Classification of spatiotemporal hemodynamics from brain perfusion MR images using expectation-Maximization estimation with finite mixture of multivariate Gaussian distributions (2007) Magnetic Resonance in Medicine, 57, pp. 181-191; Schwarz, G., Estimating the dimension of a model (1978) Ann. Statist., 6, pp. 461-464; Otsu, N., A threshold selection method from gray-level histogram (1979) IEEE Transactions on Systems, Man, and Cybernetics; Jiang, H., Kim, J., Pearlson, G.D., Mori, S., DtiStudio: Resource program for diffusion tensor computation and fiber bundle tracking (2006) Computer Methods and Programs in Biomedicine, 81, pp. 106-116

PY - 2009

Y1 - 2009

N2 - Multiple system atrophy (MSA) is a well-known neurodegenerative disorders that present parkinsonism syndrome and autonomic dysfunction. Patients with MSA who have the combination of parkinsonism and cerebellar ataxia are referred to as MSA-C. Brain diffusion-weighted imaging (DWI) offers the potential for objective criteria in the diagnosis of MSA. We aim to develop an automatic method to segment out the abnormal whole brain area in MSA-C patients based on the 13-direction DWI raw data. The whole brain DWI raw data of fifteen normal subjects and nine MSA-C patients were analyzed. In this study, we proposed a novel method to perform tissue segmentation directly based on the directional information of the DWI images, rather than using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC) as in the previous literatures. 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. Our results demonstrate that the HC-EM is effective in multi-tissue classification, namely, the cerebrospinal fluid, gray matter, and several areas of white matters, on the DWI raw data. The segmented patterns and the corresponding intensities of thirteen directions of the cerebellum in MSA-C patients showed the decrease of the anisotropy, which were evidently different from the results in normal subjects.

AB - Multiple system atrophy (MSA) is a well-known neurodegenerative disorders that present parkinsonism syndrome and autonomic dysfunction. Patients with MSA who have the combination of parkinsonism and cerebellar ataxia are referred to as MSA-C. Brain diffusion-weighted imaging (DWI) offers the potential for objective criteria in the diagnosis of MSA. We aim to develop an automatic method to segment out the abnormal whole brain area in MSA-C patients based on the 13-direction DWI raw data. The whole brain DWI raw data of fifteen normal subjects and nine MSA-C patients were analyzed. In this study, we proposed a novel method to perform tissue segmentation directly based on the directional information of the DWI images, rather than using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC) as in the previous literatures. 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. Our results demonstrate that the HC-EM is effective in multi-tissue classification, namely, the cerebrospinal fluid, gray matter, and several areas of white matters, on the DWI raw data. The segmented patterns and the corresponding intensities of thirteen directions of the cerebellum in MSA-C patients showed the decrease of the anisotropy, which were evidently different from the results in normal subjects.

KW - diffusion-weighted imaging

KW - expectation maximization algorithm

KW - hierarchical clustering

KW - Multiple system atrophy

KW - Apparent diffusion coefficient

KW - Automatic segmentations

KW - Diffusion weighted imaging

KW - Directional information

KW - Expectation-maximization algorithms

KW - Hier-archical clustering

KW - Multiple system atrophies

KW - Neurodegenerative disorders

KW - Anisotropy

KW - Biomedical engineering

KW - Brain

KW - Brain mapping

KW - Cerebrospinal fluid

KW - Clustering algorithms

KW - Diffusion

KW - Maximum principle

KW - Neurodegenerative diseases

KW - Tissue

KW - Image segmentation

UR - http://www.scopus.com/inward/record.url?eid=2-s2.0-84891914161&partnerID=40&md5=e82d38776f34b974fbc10bb7433a1848

UR - https://www.scopus.com/results/citedbyresults.uri?sort=plf-f&cite=2-s2.0-84891914161&src=s&imp=t&sid=79c36b2c23fa07a7effac516cebf74ea&sot=cite&sdt=a&sl=0&origin=recordpage&editSaveSearch=&txGid=569b0695be6b6583c73b09fdc6a02a6a

U2 - 10.1007/978-3-540-92841-6_177

DO - 10.1007/978-3-540-92841-6_177

M3 - Other

SP - 722

EP - 725

ER -