Journal article
Evaluation of a New Method for Automated Detection of Left Ventricular Boundaries in Time Series of Magnetic Resonance Images Using an Active Appearance Motion Model
Journal of cardiovascular magnetic resonance, Vol.6(3), pp.609-617
2004
DOI: 10.1081/JCMR-120038082
PMID: 15347125
Abstract
The purpose of this study was the evaluation of a computer algorithm for the automated detection of endocardial and epicardial boundaries of the left ventricle in time series of short-axis magnetic resonance images based on an Active Appearance Motion Model (AAMM). In 20 short-axis MR examinations, manual contours were defined in multiple temporal frames (from end-diastole to end-systole) in multiple slices from base to apex. Using a leave-one-out procedure, the image data and contours were used to build 20 different AAMMs giving a statistical description of the ventricular shape, gray value appearance, and cardiac motion patterns in the training set. Automated contour detection was performed by iteratively deforming the AAMM within statistically allowed limits until an optimal match was found between the deformed AAMM and the underlying image data of the left-out subject. Global ventricular function results derived from automatically detected contours were compared with results obtained from manually traced boundaries. The AAMM contour detection method was successful in 17 of 20 studies. The three failures were excluded from further statistical analysis. Automated contour detection resulted in small, but statistically nonsignificant, underestimations of ventricular volumes and mass: differences for end-diastolic volume were 0.3% ± 12.0%, for end-systolic volume 2.0% ± 23.4% and for left ventricular myocardial mass 0.73% ± 14.9% (mean ± SD). An excellent agreement was observed in the ejection fraction: difference of 0.1% ± 6.7%. In conclusion, the presented fully automated contour detection method provides assessment of quantitative global function that is comparable to manual analysis.
Details
- Title: Subtitle
- Evaluation of a New Method for Automated Detection of Left Ventricular Boundaries in Time Series of Magnetic Resonance Images Using an Active Appearance Motion Model
- Creators
- Rob J van der Geest - 1Department of Radiology, Division of Image Processing (LKEB), Leiden University Medical Center, Mailstop 1 C2-S, PO Box 9600, Leiden, 2300 RC, The NetherlandsBoudewijn P. F Lelieveldt - 1Department of Radiology, Division of Image Processing (LKEB), Leiden University Medical Center, Mailstop 1 C2-S, PO Box 9600, Leiden, 2300 RC, The NetherlandsEmmanuelle Angelié - 1Department of Radiology, Division of Image Processing (LKEB), Leiden University Medical Center, Mailstop 1 C2-S, PO Box 9600, Leiden, 2300 RC, The NetherlandsMikhail Danilouchkine - 1Department of Radiology, Division of Image Processing (LKEB), Leiden University Medical Center, Mailstop 1 C2-S, PO Box 9600, Leiden, 2300 RC, The NetherlandsCory Swingen - 2Department of Radiology, University of Minnesota, Minneapolis, Minnesota, 55455, USAM Sonka - 3Department of Electrical and Computer Engineering, University of Iowa, Iowa City, Iowa, 52242, USAJohan H. C Reiber - 1Department of Radiology, Division of Image Processing (LKEB), Leiden University Medical Center, Mailstop 1 C2-S, PO Box 9600, Leiden, 2300 RC, The Netherlands
- Resource Type
- Journal article
- Publication Details
- Journal of cardiovascular magnetic resonance, Vol.6(3), pp.609-617
- DOI
- 10.1081/JCMR-120038082
- PMID
- 15347125
- NLM abbreviation
- J Cardiovasc Magn Reson
- ISSN
- 1097-6647
- eISSN
- 1532-429X
- Publisher
- Informa UK Ltd
- Language
- English
- Date published
- 2004
- Academic Unit
- Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Radiation Oncology; Fraternal Order of Eagles Diabetes Research Center; Injury Prevention Research Center; Ophthalmology and Visual Sciences
- Record Identifier
- 9984047877002771
Metrics
10 Record Views