Automatic Segmentation of Abdominal Adipose Tissue in MRI

Thomas Hammershaimb Mosbech; Kasper Pilgaard; Allan Vaag; Rasmus Larsen

doi:10.1007/978-3-642-21227-7_47

Automatic Segmentation of Abdominal Adipose Tissue in MRI

Thomas Hammershaimb Mosbech, Kasper Pilgaard, Allan Vaag, Rasmus Larsen

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

This paper presents a method for automatically segmenting abdominal adipose tissue from 3-dimensional magnetic resonance images. We distinguish between three types of adipose tissue; visceral, deep subcutaneous and superficial subcutaneous. Images are pre-processed to remove the bias field effect of intensity in-homogeneities. This effect is estimated by a thin plate spline extended to fit two classes of automatically sampled intensity points in 3D. Adipose tissue pixels are labelled with fuzzy c-means clustering and locally determined thresholds. The visceral and subcutaneous adipose tissue are separated using deformable models, incorporating information from the clustering. The subcutaneous adipose tissue is subdivided into a deep and superficial part by means of dynamic programming applied to a spatial transformation of the image data. Regression analysis shows good correspondences between our results and total abdominal adipose tissue percentages assessed by dualemission X-ray absorptiometry (R2 = 0.86).

Original language	English
Title of host publication	Image Analysis : 17th Scandinavian Conference, SCIA 2011 - Ystad, Sweden, May 2011 - Proceedings
Publisher	Springer
Publication date	2011
Pages	501-511
ISBN (Print)	978-3-642-21226-0
ISBN (Electronic)	978-3-642-21227-7
DOIs	https://doi.org/10.1007/978-3-642-21227-7_47
Publication status	Published - 2011
Event	17th Scandinavian Conference on Image Analysis (SCIA) - Ystad Saltsjöbad, Ystad, Sweden Duration: 23 May 2011 → 27 May 2011 http://www.maths.lth.se/vision/scia2011/

Conference

Conference	17th Scandinavian Conference on Image Analysis (SCIA)
Location	Ystad Saltsjöbad
Country	Sweden
City	Ystad
Period	23/05/2011 → 27/05/2011
Internet address	http://www.maths.lth.se/vision/scia2011/

Series	Lecture Notes in Computer Science
Number	6688
ISSN	0302-9743

Keywords

Tissue classification
Abdominal adipose tissue
MRI
Bias field correction
Image processing

Access to Document

10.1007/978-3-642-21227-7_47

http://www.maths.lth.se/vision/scia2011/

Fingerprint Dive into the research topics of 'Automatic Segmentation of Abdominal Adipose Tissue in MRI'. Together they form a unique fingerprint.

Cite this

@inproceedings{04554c1be0b94a9d948a8a19030d1497,

title = "Automatic Segmentation of Abdominal Adipose Tissue in MRI",

abstract = "This paper presents a method for automatically segmenting abdominal adipose tissue from 3-dimensional magnetic resonance images. We distinguish between three types of adipose tissue; visceral, deep subcutaneous and superficial subcutaneous. Images are pre-processed to remove the bias field effect of intensity in-homogeneities. This effect is estimated by a thin plate spline extended to fit two classes of automatically sampled intensity points in 3D. Adipose tissue pixels are labelled with fuzzy c-means clustering and locally determined thresholds. The visceral and subcutaneous adipose tissue are separated using deformable models, incorporating information from the clustering. The subcutaneous adipose tissue is subdivided into a deep and superficial part by means of dynamic programming applied to a spatial transformation of the image data. Regression analysis shows good correspondences between our results and total abdominal adipose tissue percentages assessed by dualemission X-ray absorptiometry (R2 = 0.86).",

keywords = "Tissue classification, Abdominal adipose tissue, MRI, Bias field correction, Image processing",

author = "Mosbech, {Thomas Hammershaimb} and Kasper Pilgaard and Allan Vaag and Rasmus Larsen",

year = "2011",

doi = "10.1007/978-3-642-21227-7_47",

language = "English",

isbn = "978-3-642-21226-0",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

number = "6688",

pages = "501--511",

booktitle = "Image Analysis",

note = "17th Scandinavian Conference on Image Analysis (SCIA) ; Conference date: 23-05-2011 Through 27-05-2011",

url = "http://www.maths.lth.se/vision/scia2011/",

}

Mosbech, TH, Pilgaard, K, Vaag, A & Larsen, R 2011, Automatic Segmentation of Abdominal Adipose Tissue in MRI. in Image Analysis: 17th Scandinavian Conference, SCIA 2011 - Ystad, Sweden, May 2011 - Proceedings. Springer, Lecture Notes in Computer Science, no. 6688, pp. 501-511, 17th Scandinavian Conference on Image Analysis (SCIA), Ystad, Sweden, 23/05/2011. https://doi.org/10.1007/978-3-642-21227-7_47

Automatic Segmentation of Abdominal Adipose Tissue in MRI. / Mosbech, Thomas Hammershaimb; Pilgaard, Kasper; Vaag, Allan; Larsen, Rasmus.

Image Analysis: 17th Scandinavian Conference, SCIA 2011 - Ystad, Sweden, May 2011 - Proceedings. Springer, 2011. p. 501-511 (Lecture Notes in Computer Science; No. 6688).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Automatic Segmentation of Abdominal Adipose Tissue in MRI

AU - Mosbech, Thomas Hammershaimb

AU - Pilgaard, Kasper

AU - Vaag, Allan

AU - Larsen, Rasmus

PY - 2011

Y1 - 2011

N2 - This paper presents a method for automatically segmenting abdominal adipose tissue from 3-dimensional magnetic resonance images. We distinguish between three types of adipose tissue; visceral, deep subcutaneous and superficial subcutaneous. Images are pre-processed to remove the bias field effect of intensity in-homogeneities. This effect is estimated by a thin plate spline extended to fit two classes of automatically sampled intensity points in 3D. Adipose tissue pixels are labelled with fuzzy c-means clustering and locally determined thresholds. The visceral and subcutaneous adipose tissue are separated using deformable models, incorporating information from the clustering. The subcutaneous adipose tissue is subdivided into a deep and superficial part by means of dynamic programming applied to a spatial transformation of the image data. Regression analysis shows good correspondences between our results and total abdominal adipose tissue percentages assessed by dualemission X-ray absorptiometry (R2 = 0.86).

AB - This paper presents a method for automatically segmenting abdominal adipose tissue from 3-dimensional magnetic resonance images. We distinguish between three types of adipose tissue; visceral, deep subcutaneous and superficial subcutaneous. Images are pre-processed to remove the bias field effect of intensity in-homogeneities. This effect is estimated by a thin plate spline extended to fit two classes of automatically sampled intensity points in 3D. Adipose tissue pixels are labelled with fuzzy c-means clustering and locally determined thresholds. The visceral and subcutaneous adipose tissue are separated using deformable models, incorporating information from the clustering. The subcutaneous adipose tissue is subdivided into a deep and superficial part by means of dynamic programming applied to a spatial transformation of the image data. Regression analysis shows good correspondences between our results and total abdominal adipose tissue percentages assessed by dualemission X-ray absorptiometry (R2 = 0.86).

KW - Tissue classification

KW - Abdominal adipose tissue

KW - MRI

KW - Bias field correction

KW - Image processing

U2 - 10.1007/978-3-642-21227-7_47

DO - 10.1007/978-3-642-21227-7_47

M3 - Article in proceedings

SN - 978-3-642-21226-0

T3 - Lecture Notes in Computer Science

SP - 501

EP - 511

BT - Image Analysis

PB - Springer

T2 - 17th Scandinavian Conference on Image Analysis (SCIA)

Y2 - 23 May 2011 through 27 May 2011

ER -