OASIS is Automated Statistical Inference for Segmentation, with applications to multiple sclerosis lesion segmentation in MRI

Elizabeth M. Sweeney, Russell T. Shinohara, Navid Shiee, Farrah J. Mateen, Avni A. Chudgar, Jennifer L. Cuzzocreo, Peter Calabresi, Dzung L. Pham, Daniel S. Reich, Ciprian M Crainiceanu

Research output: Contribution to journalArticle

Abstract

Magnetic resonance imaging (MRI) can be used to detect lesions in the brains of multiple sclerosis (MS) patients and is essential for diagnosing the disease and monitoring its progression. In practice, lesion load is often quantified by either manual or semi-automated segmentation of MRI, which is time-consuming, costly, and associated with large inter- and intra-observer variability. We propose OASIS is Automated Statistical Inference for Segmentation (OASIS), an automated statistical method for segmenting MS lesions in MRI studies. We use logistic regression models incorporating multiple MRI modalities to estimate voxel-level probabilities of lesion presence. Intensity-normalized T1-weighted, T2-weighted, fluid-attenuated inversion recovery and proton density volumes from 131 MRI studies (98 MS subjects, 33 healthy subjects) with manual lesion segmentations were used to train and validate our model. Within this set, OASIS detected lesions with a partial area under the receiver operating characteristic curve for clinically relevant false positive rates of 1% and below of 0.59% (95% CI; [0.50%, 0.67%]) at the voxel level. An experienced MS neuroradiologist compared these segmentations to those produced by LesionTOADS, an image segmentation software that provides segmentation of both lesions and normal brain structures. For lesions, OASIS out-performed LesionTOADS in 74% (95% CI: [65%, 82%]) of cases for the 98 MS subjects. To further validate the method, we applied OASIS to 169 MRI studies acquired at a separate center. The neuroradiologist again compared the OASIS segmentations to those from LesionTOADS. For lesions, OASIS ranked higher than LesionTOADS in 77% (95% CI: [71%, 83%]) of cases. For a randomly selected subset of 50 of these studies, one additional radiologist and one neurologist also scored the images. Within this set, the neuroradiologist ranked OASIS higher than LesionTOADS in 76% (95% CI: [64%, 88%]) of cases, the neurologist 66% (95% CI: [52%, 78%]) and the radiologist 52% (95% CI: [38%, 66%]). OASIS obtains the estimated probability for each voxel to be part of a lesion by weighting each imaging modality with coefficient weights. These coefficients are explicit, obtained using standard model fitting techniques, and can be reused in other imaging studies. This fully automated method allows sensitive and specific detection of lesion presence and may be rapidly applied to large collections of images.

Original languageEnglish (US)
Pages (from-to)402-413
Number of pages12
JournalNeuroImage: Clinical
Volume2
Issue number1
DOIs
StatePublished - 2013

Fingerprint

Multiple Sclerosis
Magnetic Resonance Imaging
Logistic Models
poly(divinyl-co-N-vinylpyrrolidinone)
Observer Variation
Brain
ROC Curve
Protons
Healthy Volunteers
Software
Weights and Measures

Keywords

  • Brain
  • Lesion segmentation
  • Logistic regression
  • MRI
  • Multiple sclerosis
  • Statistical modeling

ASJC Scopus subject areas

  • Clinical Neurology
  • Radiology Nuclear Medicine and imaging
  • Cognitive Neuroscience
  • Neurology

Cite this

OASIS is Automated Statistical Inference for Segmentation, with applications to multiple sclerosis lesion segmentation in MRI. / Sweeney, Elizabeth M.; Shinohara, Russell T.; Shiee, Navid; Mateen, Farrah J.; Chudgar, Avni A.; Cuzzocreo, Jennifer L.; Calabresi, Peter; Pham, Dzung L.; Reich, Daniel S.; Crainiceanu, Ciprian M.

In: NeuroImage: Clinical, Vol. 2, No. 1, 2013, p. 402-413.

Research output: Contribution to journalArticle

Sweeney, Elizabeth M. ; Shinohara, Russell T. ; Shiee, Navid ; Mateen, Farrah J. ; Chudgar, Avni A. ; Cuzzocreo, Jennifer L. ; Calabresi, Peter ; Pham, Dzung L. ; Reich, Daniel S. ; Crainiceanu, Ciprian M. / OASIS is Automated Statistical Inference for Segmentation, with applications to multiple sclerosis lesion segmentation in MRI. In: NeuroImage: Clinical. 2013 ; Vol. 2, No. 1. pp. 402-413.
@article{bf0d267c58884852b10a19c6833a52fd,
title = "OASIS is Automated Statistical Inference for Segmentation, with applications to multiple sclerosis lesion segmentation in MRI",
abstract = "Magnetic resonance imaging (MRI) can be used to detect lesions in the brains of multiple sclerosis (MS) patients and is essential for diagnosing the disease and monitoring its progression. In practice, lesion load is often quantified by either manual or semi-automated segmentation of MRI, which is time-consuming, costly, and associated with large inter- and intra-observer variability. We propose OASIS is Automated Statistical Inference for Segmentation (OASIS), an automated statistical method for segmenting MS lesions in MRI studies. We use logistic regression models incorporating multiple MRI modalities to estimate voxel-level probabilities of lesion presence. Intensity-normalized T1-weighted, T2-weighted, fluid-attenuated inversion recovery and proton density volumes from 131 MRI studies (98 MS subjects, 33 healthy subjects) with manual lesion segmentations were used to train and validate our model. Within this set, OASIS detected lesions with a partial area under the receiver operating characteristic curve for clinically relevant false positive rates of 1{\%} and below of 0.59{\%} (95{\%} CI; [0.50{\%}, 0.67{\%}]) at the voxel level. An experienced MS neuroradiologist compared these segmentations to those produced by LesionTOADS, an image segmentation software that provides segmentation of both lesions and normal brain structures. For lesions, OASIS out-performed LesionTOADS in 74{\%} (95{\%} CI: [65{\%}, 82{\%}]) of cases for the 98 MS subjects. To further validate the method, we applied OASIS to 169 MRI studies acquired at a separate center. The neuroradiologist again compared the OASIS segmentations to those from LesionTOADS. For lesions, OASIS ranked higher than LesionTOADS in 77{\%} (95{\%} CI: [71{\%}, 83{\%}]) of cases. For a randomly selected subset of 50 of these studies, one additional radiologist and one neurologist also scored the images. Within this set, the neuroradiologist ranked OASIS higher than LesionTOADS in 76{\%} (95{\%} CI: [64{\%}, 88{\%}]) of cases, the neurologist 66{\%} (95{\%} CI: [52{\%}, 78{\%}]) and the radiologist 52{\%} (95{\%} CI: [38{\%}, 66{\%}]). OASIS obtains the estimated probability for each voxel to be part of a lesion by weighting each imaging modality with coefficient weights. These coefficients are explicit, obtained using standard model fitting techniques, and can be reused in other imaging studies. This fully automated method allows sensitive and specific detection of lesion presence and may be rapidly applied to large collections of images.",
keywords = "Brain, Lesion segmentation, Logistic regression, MRI, Multiple sclerosis, Statistical modeling",
author = "Sweeney, {Elizabeth M.} and Shinohara, {Russell T.} and Navid Shiee and Mateen, {Farrah J.} and Chudgar, {Avni A.} and Cuzzocreo, {Jennifer L.} and Peter Calabresi and Pham, {Dzung L.} and Reich, {Daniel S.} and Crainiceanu, {Ciprian M}",
year = "2013",
doi = "10.1016/j.nicl.2013.03.002",
language = "English (US)",
volume = "2",
pages = "402--413",
journal = "NeuroImage: Clinical",
issn = "2213-1582",
publisher = "Elsevier BV",
number = "1",

}

TY - JOUR

T1 - OASIS is Automated Statistical Inference for Segmentation, with applications to multiple sclerosis lesion segmentation in MRI

AU - Sweeney, Elizabeth M.

AU - Shinohara, Russell T.

AU - Shiee, Navid

AU - Mateen, Farrah J.

AU - Chudgar, Avni A.

AU - Cuzzocreo, Jennifer L.

AU - Calabresi, Peter

AU - Pham, Dzung L.

AU - Reich, Daniel S.

AU - Crainiceanu, Ciprian M

PY - 2013

Y1 - 2013

N2 - Magnetic resonance imaging (MRI) can be used to detect lesions in the brains of multiple sclerosis (MS) patients and is essential for diagnosing the disease and monitoring its progression. In practice, lesion load is often quantified by either manual or semi-automated segmentation of MRI, which is time-consuming, costly, and associated with large inter- and intra-observer variability. We propose OASIS is Automated Statistical Inference for Segmentation (OASIS), an automated statistical method for segmenting MS lesions in MRI studies. We use logistic regression models incorporating multiple MRI modalities to estimate voxel-level probabilities of lesion presence. Intensity-normalized T1-weighted, T2-weighted, fluid-attenuated inversion recovery and proton density volumes from 131 MRI studies (98 MS subjects, 33 healthy subjects) with manual lesion segmentations were used to train and validate our model. Within this set, OASIS detected lesions with a partial area under the receiver operating characteristic curve for clinically relevant false positive rates of 1% and below of 0.59% (95% CI; [0.50%, 0.67%]) at the voxel level. An experienced MS neuroradiologist compared these segmentations to those produced by LesionTOADS, an image segmentation software that provides segmentation of both lesions and normal brain structures. For lesions, OASIS out-performed LesionTOADS in 74% (95% CI: [65%, 82%]) of cases for the 98 MS subjects. To further validate the method, we applied OASIS to 169 MRI studies acquired at a separate center. The neuroradiologist again compared the OASIS segmentations to those from LesionTOADS. For lesions, OASIS ranked higher than LesionTOADS in 77% (95% CI: [71%, 83%]) of cases. For a randomly selected subset of 50 of these studies, one additional radiologist and one neurologist also scored the images. Within this set, the neuroradiologist ranked OASIS higher than LesionTOADS in 76% (95% CI: [64%, 88%]) of cases, the neurologist 66% (95% CI: [52%, 78%]) and the radiologist 52% (95% CI: [38%, 66%]). OASIS obtains the estimated probability for each voxel to be part of a lesion by weighting each imaging modality with coefficient weights. These coefficients are explicit, obtained using standard model fitting techniques, and can be reused in other imaging studies. This fully automated method allows sensitive and specific detection of lesion presence and may be rapidly applied to large collections of images.

AB - Magnetic resonance imaging (MRI) can be used to detect lesions in the brains of multiple sclerosis (MS) patients and is essential for diagnosing the disease and monitoring its progression. In practice, lesion load is often quantified by either manual or semi-automated segmentation of MRI, which is time-consuming, costly, and associated with large inter- and intra-observer variability. We propose OASIS is Automated Statistical Inference for Segmentation (OASIS), an automated statistical method for segmenting MS lesions in MRI studies. We use logistic regression models incorporating multiple MRI modalities to estimate voxel-level probabilities of lesion presence. Intensity-normalized T1-weighted, T2-weighted, fluid-attenuated inversion recovery and proton density volumes from 131 MRI studies (98 MS subjects, 33 healthy subjects) with manual lesion segmentations were used to train and validate our model. Within this set, OASIS detected lesions with a partial area under the receiver operating characteristic curve for clinically relevant false positive rates of 1% and below of 0.59% (95% CI; [0.50%, 0.67%]) at the voxel level. An experienced MS neuroradiologist compared these segmentations to those produced by LesionTOADS, an image segmentation software that provides segmentation of both lesions and normal brain structures. For lesions, OASIS out-performed LesionTOADS in 74% (95% CI: [65%, 82%]) of cases for the 98 MS subjects. To further validate the method, we applied OASIS to 169 MRI studies acquired at a separate center. The neuroradiologist again compared the OASIS segmentations to those from LesionTOADS. For lesions, OASIS ranked higher than LesionTOADS in 77% (95% CI: [71%, 83%]) of cases. For a randomly selected subset of 50 of these studies, one additional radiologist and one neurologist also scored the images. Within this set, the neuroradiologist ranked OASIS higher than LesionTOADS in 76% (95% CI: [64%, 88%]) of cases, the neurologist 66% (95% CI: [52%, 78%]) and the radiologist 52% (95% CI: [38%, 66%]). OASIS obtains the estimated probability for each voxel to be part of a lesion by weighting each imaging modality with coefficient weights. These coefficients are explicit, obtained using standard model fitting techniques, and can be reused in other imaging studies. This fully automated method allows sensitive and specific detection of lesion presence and may be rapidly applied to large collections of images.

KW - Brain

KW - Lesion segmentation

KW - Logistic regression

KW - MRI

KW - Multiple sclerosis

KW - Statistical modeling

UR - http://www.scopus.com/inward/record.url?scp=84875897566&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875897566&partnerID=8YFLogxK

U2 - 10.1016/j.nicl.2013.03.002

DO - 10.1016/j.nicl.2013.03.002

M3 - Article

VL - 2

SP - 402

EP - 413

JO - NeuroImage: Clinical

JF - NeuroImage: Clinical

SN - 2213-1582

IS - 1

ER -