Automatic segmentation of MR brain images in multiple sclerosis patients

R. T V Avula, Bradley J Erickson

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

Abstract

A totally automatic scheme for segmenting brain from extracranial tissues and to classify all intracranial voxels as CSF, gray matter(GM), white matter(WM) or abnormality such as multiple sclerosis (MS) lesions is presented in this paper. It is observed that in MR head images, if a tissue's intensity values are normalized, its relationship to the other tissues is essentially constant for a given type of image. Based on this approach, the subcutaneous fat surrounding the head is normalized to classify other tissues. Spatially registered 3 mm MR head image slices of T1 weighted, fast spin echo (dual echo T2 weighted and proton density(PD) weighted images) and fast Fluid Attenuated Inversion Recovery(FLAIR) sequences are used for segmentation. Subcutaneous fat surrounding the skull was identified based on intensity thresholding from T1 weighted images. A multiparametric space map was developed for CSF, GM and WM by normalizing each tissue with respect to the mean value of corresponding subcutaneous fat on each pulse sequence. To reduce the low frequency noise without blurring the fine morphological high frequency details, an anisotropic diffusion filter was applied to all images before segmentation. An initial slice by slice classification was followed by morphological operations to delete any bridges connecting extracranial segments. Finally 3-dimensional region growing of the segmented brain extracts GM, WM and pathology. The algorithm was tested on sequential scans of 10 patients with MS lesions. For well registered sequences, tissues and pathology have been accurately classified. This procedure does not require user input or image training data sets, and shows promise for automatic classification of brain and pathology.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsM.H. Loew, K.M. Hanson
Pages960-966
Number of pages7
Volume2710
DOIs
StatePublished - 1996
EventMedical Imaging 1996 Image Processing - Newport Beach, CA, United States
Duration: Feb 12 1996Feb 15 1996

Other

OtherMedical Imaging 1996 Image Processing
CountryUnited States
CityNewport Beach, CA
Period2/12/962/15/96

Fingerprint

brain
Brain
Tissue
Pathology
Oils and fats
fats
pathology
lesions
echoes
skull
normalizing
blurring
Image segmentation
abnormalities
Protons
education
recovery
Recovery
inversions
Fluids

Keywords

  • Anisotropic diffusion filter
  • Automatic Multiparametric Classification
  • Brain
  • FLAIR sequence
  • Multiple sclerosis lesions
  • Spatially registered MR images
  • Subcutaneous fat tissue

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Avula, R. T. V., & Erickson, B. J. (1996). Automatic segmentation of MR brain images in multiple sclerosis patients. In M. H. Loew, & K. M. Hanson (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 2710, pp. 960-966) https://doi.org/10.1117/12.237904

Automatic segmentation of MR brain images in multiple sclerosis patients. / Avula, R. T V; Erickson, Bradley J.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / M.H. Loew; K.M. Hanson. Vol. 2710 1996. p. 960-966.

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

Avula, RTV & Erickson, BJ 1996, Automatic segmentation of MR brain images in multiple sclerosis patients. in MH Loew & KM Hanson (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 2710, pp. 960-966, Medical Imaging 1996 Image Processing, Newport Beach, CA, United States, 2/12/96. https://doi.org/10.1117/12.237904
Avula RTV, Erickson BJ. Automatic segmentation of MR brain images in multiple sclerosis patients. In Loew MH, Hanson KM, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 2710. 1996. p. 960-966 https://doi.org/10.1117/12.237904
Avula, R. T V ; Erickson, Bradley J. / Automatic segmentation of MR brain images in multiple sclerosis patients. Proceedings of SPIE - The International Society for Optical Engineering. editor / M.H. Loew ; K.M. Hanson. Vol. 2710 1996. pp. 960-966
@inproceedings{c7dee52d995c4fe0860a3b5dec0ff9d1,
title = "Automatic segmentation of MR brain images in multiple sclerosis patients",
abstract = "A totally automatic scheme for segmenting brain from extracranial tissues and to classify all intracranial voxels as CSF, gray matter(GM), white matter(WM) or abnormality such as multiple sclerosis (MS) lesions is presented in this paper. It is observed that in MR head images, if a tissue's intensity values are normalized, its relationship to the other tissues is essentially constant for a given type of image. Based on this approach, the subcutaneous fat surrounding the head is normalized to classify other tissues. Spatially registered 3 mm MR head image slices of T1 weighted, fast spin echo (dual echo T2 weighted and proton density(PD) weighted images) and fast Fluid Attenuated Inversion Recovery(FLAIR) sequences are used for segmentation. Subcutaneous fat surrounding the skull was identified based on intensity thresholding from T1 weighted images. A multiparametric space map was developed for CSF, GM and WM by normalizing each tissue with respect to the mean value of corresponding subcutaneous fat on each pulse sequence. To reduce the low frequency noise without blurring the fine morphological high frequency details, an anisotropic diffusion filter was applied to all images before segmentation. An initial slice by slice classification was followed by morphological operations to delete any bridges connecting extracranial segments. Finally 3-dimensional region growing of the segmented brain extracts GM, WM and pathology. The algorithm was tested on sequential scans of 10 patients with MS lesions. For well registered sequences, tissues and pathology have been accurately classified. This procedure does not require user input or image training data sets, and shows promise for automatic classification of brain and pathology.",
keywords = "Anisotropic diffusion filter, Automatic Multiparametric Classification, Brain, FLAIR sequence, Multiple sclerosis lesions, Spatially registered MR images, Subcutaneous fat tissue",
author = "Avula, {R. T V} and Erickson, {Bradley J}",
year = "1996",
doi = "10.1117/12.237904",
language = "English (US)",
volume = "2710",
pages = "960--966",
editor = "M.H. Loew and K.M. Hanson",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Automatic segmentation of MR brain images in multiple sclerosis patients

AU - Avula, R. T V

AU - Erickson, Bradley J

PY - 1996

Y1 - 1996

N2 - A totally automatic scheme for segmenting brain from extracranial tissues and to classify all intracranial voxels as CSF, gray matter(GM), white matter(WM) or abnormality such as multiple sclerosis (MS) lesions is presented in this paper. It is observed that in MR head images, if a tissue's intensity values are normalized, its relationship to the other tissues is essentially constant for a given type of image. Based on this approach, the subcutaneous fat surrounding the head is normalized to classify other tissues. Spatially registered 3 mm MR head image slices of T1 weighted, fast spin echo (dual echo T2 weighted and proton density(PD) weighted images) and fast Fluid Attenuated Inversion Recovery(FLAIR) sequences are used for segmentation. Subcutaneous fat surrounding the skull was identified based on intensity thresholding from T1 weighted images. A multiparametric space map was developed for CSF, GM and WM by normalizing each tissue with respect to the mean value of corresponding subcutaneous fat on each pulse sequence. To reduce the low frequency noise without blurring the fine morphological high frequency details, an anisotropic diffusion filter was applied to all images before segmentation. An initial slice by slice classification was followed by morphological operations to delete any bridges connecting extracranial segments. Finally 3-dimensional region growing of the segmented brain extracts GM, WM and pathology. The algorithm was tested on sequential scans of 10 patients with MS lesions. For well registered sequences, tissues and pathology have been accurately classified. This procedure does not require user input or image training data sets, and shows promise for automatic classification of brain and pathology.

AB - A totally automatic scheme for segmenting brain from extracranial tissues and to classify all intracranial voxels as CSF, gray matter(GM), white matter(WM) or abnormality such as multiple sclerosis (MS) lesions is presented in this paper. It is observed that in MR head images, if a tissue's intensity values are normalized, its relationship to the other tissues is essentially constant for a given type of image. Based on this approach, the subcutaneous fat surrounding the head is normalized to classify other tissues. Spatially registered 3 mm MR head image slices of T1 weighted, fast spin echo (dual echo T2 weighted and proton density(PD) weighted images) and fast Fluid Attenuated Inversion Recovery(FLAIR) sequences are used for segmentation. Subcutaneous fat surrounding the skull was identified based on intensity thresholding from T1 weighted images. A multiparametric space map was developed for CSF, GM and WM by normalizing each tissue with respect to the mean value of corresponding subcutaneous fat on each pulse sequence. To reduce the low frequency noise without blurring the fine morphological high frequency details, an anisotropic diffusion filter was applied to all images before segmentation. An initial slice by slice classification was followed by morphological operations to delete any bridges connecting extracranial segments. Finally 3-dimensional region growing of the segmented brain extracts GM, WM and pathology. The algorithm was tested on sequential scans of 10 patients with MS lesions. For well registered sequences, tissues and pathology have been accurately classified. This procedure does not require user input or image training data sets, and shows promise for automatic classification of brain and pathology.

KW - Anisotropic diffusion filter

KW - Automatic Multiparametric Classification

KW - Brain

KW - FLAIR sequence

KW - Multiple sclerosis lesions

KW - Spatially registered MR images

KW - Subcutaneous fat tissue

UR - http://www.scopus.com/inward/record.url?scp=0030316355&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030316355&partnerID=8YFLogxK

U2 - 10.1117/12.237904

DO - 10.1117/12.237904

M3 - Conference contribution

VL - 2710

SP - 960

EP - 966

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Loew, M.H.

A2 - Hanson, K.M.

ER -