Gaussian mixture model-based classification of dynamic contrast enhanced MRI data for identifying diverse tumor microenvironments

Preliminary results

S. H. Han, E. Ackerstaff, Radka Stoyanova, S. Carlin, W. Huang, J. A. Koutcher, J. K. Kim, G. Cho, G. Jang, H. Cho

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

Tumor hypoxia develops heterogeneously, affects radiation sensitivity and the development of metastases. Prognostic information derived from the in vivo characterization of the spatial distribution of hypoxic areas in solid tumors can be of value for radiation therapy planning and for monitoring the early treatment response. Tumor hypoxia is caused by an imbalance between the supply and consumption of oxygen. The tumor oxygen supply is inherently linked to its vasculature and perfusion which can be evaluated by dynamic contrast enhanced (DCE-) MRI using the contrast agent Gd-DTPA. Thus, we hypothesize that DCE-MRI data may provide surrogate information regarding tumor hypoxia. In this study, DCE-MRI data from a rat prostate tumor model were analysed with a Gaussian mixture model (GMM)-based classification to identify perfused, hypoxic and necrotic areas for a total of ten tumor slices from six rats, of which one slice was used as training data for GMM classifications. The results of pattern recognition analyzes were validated by comparison to corresponding Akep maps defining the perfused area (0.84±0.09 overlap), hematoxylin and eosin (H&E)-stained tissue sections defining necrosis (0.64±0.15 overlap) and pimonidazole-stained sections defining hypoxia (0.72±0.17 overlap), respectively. Our preliminary data indicate the feasibility of a GMM-based classification to identify tumor hypoxia, necrosis and perfusion/permeability from non-invasively acquired, in vivo DCE-MRI data alone, possibly obviating the need for invasive procedures, such as biopsies, or exposure to radioactivity, such as positron emission tomography (PET) exams.

Original languageEnglish
Pages (from-to)519-532
Number of pages14
JournalNMR in Biomedicine
Volume26
Issue number5
DOIs
StatePublished - May 1 2013

Fingerprint

Tumor Microenvironment
Magnetic resonance imaging
Tumors
Neoplasms
Necrosis
Perfusion
Gadolinium DTPA
Radiation Tolerance
Hematoxylin
Eosine Yellowish-(YS)
Oxygen Consumption
Positron-Emission Tomography
Radioactivity
Contrast Media
Rats
Prostate
Permeability
Radiotherapy
Oxygen
Neoplasm Metastasis

Keywords

  • DCE-MRI
  • Gaussian mixture model
  • Hypoxia
  • Preclinical prostate model
  • Tumor microenvironments

ASJC Scopus subject areas

  • Spectroscopy
  • Molecular Medicine
  • Radiology Nuclear Medicine and imaging

Cite this

Gaussian mixture model-based classification of dynamic contrast enhanced MRI data for identifying diverse tumor microenvironments : Preliminary results. / Han, S. H.; Ackerstaff, E.; Stoyanova, Radka; Carlin, S.; Huang, W.; Koutcher, J. A.; Kim, J. K.; Cho, G.; Jang, G.; Cho, H.

In: NMR in Biomedicine, Vol. 26, No. 5, 01.05.2013, p. 519-532.

Research output: Contribution to journalArticle

Han, S. H. ; Ackerstaff, E. ; Stoyanova, Radka ; Carlin, S. ; Huang, W. ; Koutcher, J. A. ; Kim, J. K. ; Cho, G. ; Jang, G. ; Cho, H. / Gaussian mixture model-based classification of dynamic contrast enhanced MRI data for identifying diverse tumor microenvironments : Preliminary results. In: NMR in Biomedicine. 2013 ; Vol. 26, No. 5. pp. 519-532.
@article{5bb5296688e047338c1cdb27ed44942e,
title = "Gaussian mixture model-based classification of dynamic contrast enhanced MRI data for identifying diverse tumor microenvironments: Preliminary results",
abstract = "Tumor hypoxia develops heterogeneously, affects radiation sensitivity and the development of metastases. Prognostic information derived from the in vivo characterization of the spatial distribution of hypoxic areas in solid tumors can be of value for radiation therapy planning and for monitoring the early treatment response. Tumor hypoxia is caused by an imbalance between the supply and consumption of oxygen. The tumor oxygen supply is inherently linked to its vasculature and perfusion which can be evaluated by dynamic contrast enhanced (DCE-) MRI using the contrast agent Gd-DTPA. Thus, we hypothesize that DCE-MRI data may provide surrogate information regarding tumor hypoxia. In this study, DCE-MRI data from a rat prostate tumor model were analysed with a Gaussian mixture model (GMM)-based classification to identify perfused, hypoxic and necrotic areas for a total of ten tumor slices from six rats, of which one slice was used as training data for GMM classifications. The results of pattern recognition analyzes were validated by comparison to corresponding Akep maps defining the perfused area (0.84±0.09 overlap), hematoxylin and eosin (H&E)-stained tissue sections defining necrosis (0.64±0.15 overlap) and pimonidazole-stained sections defining hypoxia (0.72±0.17 overlap), respectively. Our preliminary data indicate the feasibility of a GMM-based classification to identify tumor hypoxia, necrosis and perfusion/permeability from non-invasively acquired, in vivo DCE-MRI data alone, possibly obviating the need for invasive procedures, such as biopsies, or exposure to radioactivity, such as positron emission tomography (PET) exams.",
keywords = "DCE-MRI, Gaussian mixture model, Hypoxia, Preclinical prostate model, Tumor microenvironments",
author = "Han, {S. H.} and E. Ackerstaff and Radka Stoyanova and S. Carlin and W. Huang and Koutcher, {J. A.} and Kim, {J. K.} and G. Cho and G. Jang and H. Cho",
year = "2013",
month = "5",
day = "1",
doi = "10.1002/nbm.2888",
language = "English",
volume = "26",
pages = "519--532",
journal = "NMR in Biomedicine",
issn = "0952-3480",
publisher = "John Wiley and Sons Ltd",
number = "5",

}

TY - JOUR

T1 - Gaussian mixture model-based classification of dynamic contrast enhanced MRI data for identifying diverse tumor microenvironments

T2 - Preliminary results

AU - Han, S. H.

AU - Ackerstaff, E.

AU - Stoyanova, Radka

AU - Carlin, S.

AU - Huang, W.

AU - Koutcher, J. A.

AU - Kim, J. K.

AU - Cho, G.

AU - Jang, G.

AU - Cho, H.

PY - 2013/5/1

Y1 - 2013/5/1

N2 - Tumor hypoxia develops heterogeneously, affects radiation sensitivity and the development of metastases. Prognostic information derived from the in vivo characterization of the spatial distribution of hypoxic areas in solid tumors can be of value for radiation therapy planning and for monitoring the early treatment response. Tumor hypoxia is caused by an imbalance between the supply and consumption of oxygen. The tumor oxygen supply is inherently linked to its vasculature and perfusion which can be evaluated by dynamic contrast enhanced (DCE-) MRI using the contrast agent Gd-DTPA. Thus, we hypothesize that DCE-MRI data may provide surrogate information regarding tumor hypoxia. In this study, DCE-MRI data from a rat prostate tumor model were analysed with a Gaussian mixture model (GMM)-based classification to identify perfused, hypoxic and necrotic areas for a total of ten tumor slices from six rats, of which one slice was used as training data for GMM classifications. The results of pattern recognition analyzes were validated by comparison to corresponding Akep maps defining the perfused area (0.84±0.09 overlap), hematoxylin and eosin (H&E)-stained tissue sections defining necrosis (0.64±0.15 overlap) and pimonidazole-stained sections defining hypoxia (0.72±0.17 overlap), respectively. Our preliminary data indicate the feasibility of a GMM-based classification to identify tumor hypoxia, necrosis and perfusion/permeability from non-invasively acquired, in vivo DCE-MRI data alone, possibly obviating the need for invasive procedures, such as biopsies, or exposure to radioactivity, such as positron emission tomography (PET) exams.

AB - Tumor hypoxia develops heterogeneously, affects radiation sensitivity and the development of metastases. Prognostic information derived from the in vivo characterization of the spatial distribution of hypoxic areas in solid tumors can be of value for radiation therapy planning and for monitoring the early treatment response. Tumor hypoxia is caused by an imbalance between the supply and consumption of oxygen. The tumor oxygen supply is inherently linked to its vasculature and perfusion which can be evaluated by dynamic contrast enhanced (DCE-) MRI using the contrast agent Gd-DTPA. Thus, we hypothesize that DCE-MRI data may provide surrogate information regarding tumor hypoxia. In this study, DCE-MRI data from a rat prostate tumor model were analysed with a Gaussian mixture model (GMM)-based classification to identify perfused, hypoxic and necrotic areas for a total of ten tumor slices from six rats, of which one slice was used as training data for GMM classifications. The results of pattern recognition analyzes were validated by comparison to corresponding Akep maps defining the perfused area (0.84±0.09 overlap), hematoxylin and eosin (H&E)-stained tissue sections defining necrosis (0.64±0.15 overlap) and pimonidazole-stained sections defining hypoxia (0.72±0.17 overlap), respectively. Our preliminary data indicate the feasibility of a GMM-based classification to identify tumor hypoxia, necrosis and perfusion/permeability from non-invasively acquired, in vivo DCE-MRI data alone, possibly obviating the need for invasive procedures, such as biopsies, or exposure to radioactivity, such as positron emission tomography (PET) exams.

KW - DCE-MRI

KW - Gaussian mixture model

KW - Hypoxia

KW - Preclinical prostate model

KW - Tumor microenvironments

UR - http://www.scopus.com/inward/record.url?scp=84876802376&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876802376&partnerID=8YFLogxK

U2 - 10.1002/nbm.2888

DO - 10.1002/nbm.2888

M3 - Article

VL - 26

SP - 519

EP - 532

JO - NMR in Biomedicine

JF - NMR in Biomedicine

SN - 0952-3480

IS - 5

ER -