Unified approach for multiple sclerosis lesion segmentation on brain MRI

Balasrinivasa R Sajja, Sushmita Datta, Renjie He, Meghana Mehta, Rakesh K. Gupta, Jerry S. Wolinsky, Ponnada A. Narayana

Research output: Contribution to journalArticle

85 Citations (Scopus)

Abstract

The presence of large number of false lesion classification on segmented brain MR images is a major problem in the accurate determination of lesion volumes in multiple sclerosis (MS) brains. In order to minimize the false lesion classifications, a strategy that combines parametric and nonparametric techniques is developed and implemented. This approach uses the information from the proton density (PD)- and T2-weighted and fluid attenuation inversion recovery (FLAIR) images. This strategy involves CSF and lesion classification using the Parzen window classifier. Image processing, morphological operations, and ratio maps of PD- and T2-weighted images are used for minimizing false positives. Contextual information is exploited for minimizing the false negative lesion classifications using hidden Markov random field-expectation maximization (HMRF-EM) algorithm. Lesions are delineated using fuzzy connectivity. The performance of this algorithm is quantitatively evaluated on 23 MS patients. Similarity index, percentages of over, under, and correct estimations of lesions are computed by spatially comparing the results of present procedure with expert manual segmentation. The automated processing scheme detected 80% of the manually segmented lesions in the case of low lesion load and 93% of the lesions in those cases with high lesion load.

Original languageEnglish (US)
Pages (from-to)142-151
Number of pages10
JournalAnnals of biomedical engineering
Volume34
Issue number1
DOIs
StatePublished - Jan 1 2006

Fingerprint

Magnetic resonance imaging
Brain
Protons
Image processing
Classifiers
Recovery
Fluids
Processing

Keywords

  • Expectation maximization
  • Feature classification
  • Hidden Markov random field
  • MRI
  • Multiple sclerosis
  • Segmentation

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Sajja, B. R., Datta, S., He, R., Mehta, M., Gupta, R. K., Wolinsky, J. S., & Narayana, P. A. (2006). Unified approach for multiple sclerosis lesion segmentation on brain MRI. Annals of biomedical engineering, 34(1), 142-151. https://doi.org/10.1007/s10439-005-9009-0

Unified approach for multiple sclerosis lesion segmentation on brain MRI. / Sajja, Balasrinivasa R; Datta, Sushmita; He, Renjie; Mehta, Meghana; Gupta, Rakesh K.; Wolinsky, Jerry S.; Narayana, Ponnada A.

In: Annals of biomedical engineering, Vol. 34, No. 1, 01.01.2006, p. 142-151.

Research output: Contribution to journalArticle

Sajja, BR, Datta, S, He, R, Mehta, M, Gupta, RK, Wolinsky, JS & Narayana, PA 2006, 'Unified approach for multiple sclerosis lesion segmentation on brain MRI', Annals of biomedical engineering, vol. 34, no. 1, pp. 142-151. https://doi.org/10.1007/s10439-005-9009-0
Sajja, Balasrinivasa R ; Datta, Sushmita ; He, Renjie ; Mehta, Meghana ; Gupta, Rakesh K. ; Wolinsky, Jerry S. ; Narayana, Ponnada A. / Unified approach for multiple sclerosis lesion segmentation on brain MRI. In: Annals of biomedical engineering. 2006 ; Vol. 34, No. 1. pp. 142-151.
@article{d9438bcfda1b4507b79a39aaff3114bf,
title = "Unified approach for multiple sclerosis lesion segmentation on brain MRI",
abstract = "The presence of large number of false lesion classification on segmented brain MR images is a major problem in the accurate determination of lesion volumes in multiple sclerosis (MS) brains. In order to minimize the false lesion classifications, a strategy that combines parametric and nonparametric techniques is developed and implemented. This approach uses the information from the proton density (PD)- and T2-weighted and fluid attenuation inversion recovery (FLAIR) images. This strategy involves CSF and lesion classification using the Parzen window classifier. Image processing, morphological operations, and ratio maps of PD- and T2-weighted images are used for minimizing false positives. Contextual information is exploited for minimizing the false negative lesion classifications using hidden Markov random field-expectation maximization (HMRF-EM) algorithm. Lesions are delineated using fuzzy connectivity. The performance of this algorithm is quantitatively evaluated on 23 MS patients. Similarity index, percentages of over, under, and correct estimations of lesions are computed by spatially comparing the results of present procedure with expert manual segmentation. The automated processing scheme detected 80{\%} of the manually segmented lesions in the case of low lesion load and 93{\%} of the lesions in those cases with high lesion load.",
keywords = "Expectation maximization, Feature classification, Hidden Markov random field, MRI, Multiple sclerosis, Segmentation",
author = "Sajja, {Balasrinivasa R} and Sushmita Datta and Renjie He and Meghana Mehta and Gupta, {Rakesh K.} and Wolinsky, {Jerry S.} and Narayana, {Ponnada A.}",
year = "2006",
month = "1",
day = "1",
doi = "10.1007/s10439-005-9009-0",
language = "English (US)",
volume = "34",
pages = "142--151",
journal = "Annals of Biomedical Engineering",
issn = "0090-6964",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Unified approach for multiple sclerosis lesion segmentation on brain MRI

AU - Sajja, Balasrinivasa R

AU - Datta, Sushmita

AU - He, Renjie

AU - Mehta, Meghana

AU - Gupta, Rakesh K.

AU - Wolinsky, Jerry S.

AU - Narayana, Ponnada A.

PY - 2006/1/1

Y1 - 2006/1/1

N2 - The presence of large number of false lesion classification on segmented brain MR images is a major problem in the accurate determination of lesion volumes in multiple sclerosis (MS) brains. In order to minimize the false lesion classifications, a strategy that combines parametric and nonparametric techniques is developed and implemented. This approach uses the information from the proton density (PD)- and T2-weighted and fluid attenuation inversion recovery (FLAIR) images. This strategy involves CSF and lesion classification using the Parzen window classifier. Image processing, morphological operations, and ratio maps of PD- and T2-weighted images are used for minimizing false positives. Contextual information is exploited for minimizing the false negative lesion classifications using hidden Markov random field-expectation maximization (HMRF-EM) algorithm. Lesions are delineated using fuzzy connectivity. The performance of this algorithm is quantitatively evaluated on 23 MS patients. Similarity index, percentages of over, under, and correct estimations of lesions are computed by spatially comparing the results of present procedure with expert manual segmentation. The automated processing scheme detected 80% of the manually segmented lesions in the case of low lesion load and 93% of the lesions in those cases with high lesion load.

AB - The presence of large number of false lesion classification on segmented brain MR images is a major problem in the accurate determination of lesion volumes in multiple sclerosis (MS) brains. In order to minimize the false lesion classifications, a strategy that combines parametric and nonparametric techniques is developed and implemented. This approach uses the information from the proton density (PD)- and T2-weighted and fluid attenuation inversion recovery (FLAIR) images. This strategy involves CSF and lesion classification using the Parzen window classifier. Image processing, morphological operations, and ratio maps of PD- and T2-weighted images are used for minimizing false positives. Contextual information is exploited for minimizing the false negative lesion classifications using hidden Markov random field-expectation maximization (HMRF-EM) algorithm. Lesions are delineated using fuzzy connectivity. The performance of this algorithm is quantitatively evaluated on 23 MS patients. Similarity index, percentages of over, under, and correct estimations of lesions are computed by spatially comparing the results of present procedure with expert manual segmentation. The automated processing scheme detected 80% of the manually segmented lesions in the case of low lesion load and 93% of the lesions in those cases with high lesion load.

KW - Expectation maximization

KW - Feature classification

KW - Hidden Markov random field

KW - MRI

KW - Multiple sclerosis

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=33645153542&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645153542&partnerID=8YFLogxK

U2 - 10.1007/s10439-005-9009-0

DO - 10.1007/s10439-005-9009-0

M3 - Article

VL - 34

SP - 142

EP - 151

JO - Annals of Biomedical Engineering

JF - Annals of Biomedical Engineering

SN - 0090-6964

IS - 1

ER -