Automated Segmentation of Cardiac Magnetic Resonance Images

Mikkel Bille Stegmann; Jens Chr. Nilsson; Bjørn A. Grønning

Automated Segmentation of Cardiac Magnetic Resonance Images

Mikkel Bille Stegmann, Jens Chr. Nilsson, Bjørn A. Grønning

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

45 Downloads (Pure)

Abstract

Magnetic resonance imaging (MRI) has been shown to be an accurate and precise technique to assess cardiac volumes and function in a non-invasive manner and is generally considered to be the current gold-standard for cardiac imaging [1]. Measurement of ventricular volumes, muscle mass and function is based on determination of the left-ventricular endocardial and epicardial borders. Since manual border detection is laborious, automated segmentation is highly desirable as a fast, objective and reproducible alternative. Automated segmentation will thus enhance comparability between and within cardiac studies and increase accuracy by allowing acquisition of thinner MRI-slices. This abstract demonstrates that statistical models of shape and appearance, namely the deformable models: Active Appearance Models, can successfully segment cardiac MRIs.

Original language	English
Title of host publication	Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK
Number of pages	827
Publisher	ISMRM
Publication date	2001
Publication status	Published - 2001
Event	Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK - Berkeley, CA, USA Duration: 1 Jan 2001 → …

Conference

Conference	Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK
City	Berkeley, CA, USA
Period	01/01/2001 → …

Access to Document

Imm119

http://www.imm.dtu.dk/pubdb/p.php?119

Cite this

@inproceedings{2b0923f4420745268e6683c4aabd9876,

title = "Automated Segmentation of Cardiac Magnetic Resonance Images",

abstract = "Magnetic resonance imaging (MRI) has been shown to be an accurate and precise technique to assess cardiac volumes and function in a non-invasive manner and is generally considered to be the current gold-standard for cardiac imaging [1]. Measurement of ventricular volumes, muscle mass and function is based on determination of the left-ventricular endocardial and epicardial borders. Since manual border detection is laborious, automated segmentation is highly desirable as a fast, objective and reproducible alternative. Automated segmentation will thus enhance comparability between and within cardiac studies and increase accuracy by allowing acquisition of thinner MRI-slices. This abstract demonstrates that statistical models of shape and appearance, namely the deformable models: Active Appearance Models, can successfully segment cardiac MRIs.",

author = "Stegmann, {Mikkel Bille} and Nilsson, {Jens Chr.} and Gr{\o}nning, {Bj{\o}rn A.}",

year = "2001",

language = "English",

booktitle = "Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK",

publisher = "ISMRM",

note = "Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK ; Conference date: 01-01-2001",

}

Stegmann, MB, Nilsson, JC & Grønning, BA 2001, Automated Segmentation of Cardiac Magnetic Resonance Images. in Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK. ISMRM, Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK, Berkeley, CA, USA, 01/01/2001. <http://www.imm.dtu.dk/pubdb/p.php?119>

TY - GEN

T1 - Automated Segmentation of Cardiac Magnetic Resonance Images

AU - Stegmann, Mikkel Bille

AU - Nilsson, Jens Chr.

AU - Grønning, Bjørn A.

PY - 2001

Y1 - 2001

N2 - Magnetic resonance imaging (MRI) has been shown to be an accurate and precise technique to assess cardiac volumes and function in a non-invasive manner and is generally considered to be the current gold-standard for cardiac imaging [1]. Measurement of ventricular volumes, muscle mass and function is based on determination of the left-ventricular endocardial and epicardial borders. Since manual border detection is laborious, automated segmentation is highly desirable as a fast, objective and reproducible alternative. Automated segmentation will thus enhance comparability between and within cardiac studies and increase accuracy by allowing acquisition of thinner MRI-slices. This abstract demonstrates that statistical models of shape and appearance, namely the deformable models: Active Appearance Models, can successfully segment cardiac MRIs.

AB - Magnetic resonance imaging (MRI) has been shown to be an accurate and precise technique to assess cardiac volumes and function in a non-invasive manner and is generally considered to be the current gold-standard for cardiac imaging [1]. Measurement of ventricular volumes, muscle mass and function is based on determination of the left-ventricular endocardial and epicardial borders. Since manual border detection is laborious, automated segmentation is highly desirable as a fast, objective and reproducible alternative. Automated segmentation will thus enhance comparability between and within cardiac studies and increase accuracy by allowing acquisition of thinner MRI-slices. This abstract demonstrates that statistical models of shape and appearance, namely the deformable models: Active Appearance Models, can successfully segment cardiac MRIs.

M3 - Article in proceedings

BT - Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK

PB - ISMRM

T2 - Proc. International Society of Magnetic Resonance In Medicine - ISMRM 2001, Glasgow, Scotland, UK

Y2 - 1 January 2001

ER -