GLISTR

Glioma image segmentation and registration

Ali Gooya, Kilian M. Pohl, Michel Bilello, Luigi Cirillo, George Biros, Elias R. Melhem, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

We present a generative approach for simultaneously registering a probabilistic atlas of a healthy population to brain magnetic resonance (MR) scans showing glioma and segmenting the scans into tumor as well as healthy tissue labels. The proposed method is based on the expectation maximization (EM) algorithm that incorporates a glioma growth model for atlas seeding, a process which modifies the original atlas into one with tumor and edema adapted to best match a given set of patient's images. The modified atlas is registered into the patient space and utilized for estimating the posterior probabilities of various tissue labels. EM iteratively refines the estimates of the posterior probabilities of tissue labels, the deformation field and the tumor growth model parameters. Hence, in addition to segmentation, the proposed method results in atlas registration and a low-dimensional description of the patient scans through estimation of tumor model parameters. We validate the method by automatically segmenting 10 MR scans and comparing the results to those produced by clinical experts and two state-of-the-art methods. The resulting segmentations of tumor and edema outperform the results of the reference methods, and achieve a similar accuracy from a second human rater. We additionally apply the method to 122 patients scans and report the estimated tumor model parameters and their relations with segmentation and registration results. Based on the results from this patient population, we construct a statistical atlas of the glioma by inverting the estimated deformation fields to warp the tumor segmentations of patients scans into a common space.

Original languageEnglish (US)
Article number6266750
Pages (from-to)1941-1954
Number of pages14
JournalIEEE Transactions on Medical Imaging
Volume31
Issue number10
DOIs
StatePublished - 2012
Externally publishedYes

Fingerprint

Image registration
Image segmentation
Atlases
Glioma
Tumors
Neoplasms
Labels
Magnetic resonance
Tissue
Edema
Magnetic Resonance Spectroscopy
Growth
Population
Brain

Keywords

  • Diffusion-reaction model
  • expectation maximization (EM) algorithm
  • glioma atlas
  • joint segmentation-registration

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Radiological and Ultrasound Technology
  • Software

Cite this

Gooya, A., Pohl, K. M., Bilello, M., Cirillo, L., Biros, G., Melhem, E. R., & Davatzikos, C. (2012). GLISTR: Glioma image segmentation and registration. IEEE Transactions on Medical Imaging, 31(10), 1941-1954. [6266750]. https://doi.org/10.1109/TMI.2012.2210558

GLISTR : Glioma image segmentation and registration. / Gooya, Ali; Pohl, Kilian M.; Bilello, Michel; Cirillo, Luigi; Biros, George; Melhem, Elias R.; Davatzikos, Christos.

In: IEEE Transactions on Medical Imaging, Vol. 31, No. 10, 6266750, 2012, p. 1941-1954.

Research output: Contribution to journalArticle

Gooya, A, Pohl, KM, Bilello, M, Cirillo, L, Biros, G, Melhem, ER & Davatzikos, C 2012, 'GLISTR: Glioma image segmentation and registration', IEEE Transactions on Medical Imaging, vol. 31, no. 10, 6266750, pp. 1941-1954. https://doi.org/10.1109/TMI.2012.2210558
Gooya A, Pohl KM, Bilello M, Cirillo L, Biros G, Melhem ER et al. GLISTR: Glioma image segmentation and registration. IEEE Transactions on Medical Imaging. 2012;31(10):1941-1954. 6266750. https://doi.org/10.1109/TMI.2012.2210558
Gooya, Ali ; Pohl, Kilian M. ; Bilello, Michel ; Cirillo, Luigi ; Biros, George ; Melhem, Elias R. ; Davatzikos, Christos. / GLISTR : Glioma image segmentation and registration. In: IEEE Transactions on Medical Imaging. 2012 ; Vol. 31, No. 10. pp. 1941-1954.
@article{2732d79d00354d439ab8c5465d4dccac,
title = "GLISTR: Glioma image segmentation and registration",
abstract = "We present a generative approach for simultaneously registering a probabilistic atlas of a healthy population to brain magnetic resonance (MR) scans showing glioma and segmenting the scans into tumor as well as healthy tissue labels. The proposed method is based on the expectation maximization (EM) algorithm that incorporates a glioma growth model for atlas seeding, a process which modifies the original atlas into one with tumor and edema adapted to best match a given set of patient's images. The modified atlas is registered into the patient space and utilized for estimating the posterior probabilities of various tissue labels. EM iteratively refines the estimates of the posterior probabilities of tissue labels, the deformation field and the tumor growth model parameters. Hence, in addition to segmentation, the proposed method results in atlas registration and a low-dimensional description of the patient scans through estimation of tumor model parameters. We validate the method by automatically segmenting 10 MR scans and comparing the results to those produced by clinical experts and two state-of-the-art methods. The resulting segmentations of tumor and edema outperform the results of the reference methods, and achieve a similar accuracy from a second human rater. We additionally apply the method to 122 patients scans and report the estimated tumor model parameters and their relations with segmentation and registration results. Based on the results from this patient population, we construct a statistical atlas of the glioma by inverting the estimated deformation fields to warp the tumor segmentations of patients scans into a common space.",
keywords = "Diffusion-reaction model, expectation maximization (EM) algorithm, glioma atlas, joint segmentation-registration",
author = "Ali Gooya and Pohl, {Kilian M.} and Michel Bilello and Luigi Cirillo and George Biros and Melhem, {Elias R.} and Christos Davatzikos",
year = "2012",
doi = "10.1109/TMI.2012.2210558",
language = "English (US)",
volume = "31",
pages = "1941--1954",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "10",

}

TY - JOUR

T1 - GLISTR

T2 - Glioma image segmentation and registration

AU - Gooya, Ali

AU - Pohl, Kilian M.

AU - Bilello, Michel

AU - Cirillo, Luigi

AU - Biros, George

AU - Melhem, Elias R.

AU - Davatzikos, Christos

PY - 2012

Y1 - 2012

N2 - We present a generative approach for simultaneously registering a probabilistic atlas of a healthy population to brain magnetic resonance (MR) scans showing glioma and segmenting the scans into tumor as well as healthy tissue labels. The proposed method is based on the expectation maximization (EM) algorithm that incorporates a glioma growth model for atlas seeding, a process which modifies the original atlas into one with tumor and edema adapted to best match a given set of patient's images. The modified atlas is registered into the patient space and utilized for estimating the posterior probabilities of various tissue labels. EM iteratively refines the estimates of the posterior probabilities of tissue labels, the deformation field and the tumor growth model parameters. Hence, in addition to segmentation, the proposed method results in atlas registration and a low-dimensional description of the patient scans through estimation of tumor model parameters. We validate the method by automatically segmenting 10 MR scans and comparing the results to those produced by clinical experts and two state-of-the-art methods. The resulting segmentations of tumor and edema outperform the results of the reference methods, and achieve a similar accuracy from a second human rater. We additionally apply the method to 122 patients scans and report the estimated tumor model parameters and their relations with segmentation and registration results. Based on the results from this patient population, we construct a statistical atlas of the glioma by inverting the estimated deformation fields to warp the tumor segmentations of patients scans into a common space.

AB - We present a generative approach for simultaneously registering a probabilistic atlas of a healthy population to brain magnetic resonance (MR) scans showing glioma and segmenting the scans into tumor as well as healthy tissue labels. The proposed method is based on the expectation maximization (EM) algorithm that incorporates a glioma growth model for atlas seeding, a process which modifies the original atlas into one with tumor and edema adapted to best match a given set of patient's images. The modified atlas is registered into the patient space and utilized for estimating the posterior probabilities of various tissue labels. EM iteratively refines the estimates of the posterior probabilities of tissue labels, the deformation field and the tumor growth model parameters. Hence, in addition to segmentation, the proposed method results in atlas registration and a low-dimensional description of the patient scans through estimation of tumor model parameters. We validate the method by automatically segmenting 10 MR scans and comparing the results to those produced by clinical experts and two state-of-the-art methods. The resulting segmentations of tumor and edema outperform the results of the reference methods, and achieve a similar accuracy from a second human rater. We additionally apply the method to 122 patients scans and report the estimated tumor model parameters and their relations with segmentation and registration results. Based on the results from this patient population, we construct a statistical atlas of the glioma by inverting the estimated deformation fields to warp the tumor segmentations of patients scans into a common space.

KW - Diffusion-reaction model

KW - expectation maximization (EM) algorithm

KW - glioma atlas

KW - joint segmentation-registration

UR - http://www.scopus.com/inward/record.url?scp=84867071291&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867071291&partnerID=8YFLogxK

U2 - 10.1109/TMI.2012.2210558

DO - 10.1109/TMI.2012.2210558

M3 - Article

VL - 31

SP - 1941

EP - 1954

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 10

M1 - 6266750

ER -