- Poster presentation
- Open access
- Published:
Validation and development of a new automatic algorithm for time resolved segmentation of the left ventricle in magnetic resonance imaging
Journal of Cardiovascular Magnetic Resonance volume 17, Article number: P68 (2015)
Background
Manually delineating the left ventricle (LV) is considered the clinical standard for quantification of cardiovascular magnetic resonance (CMR) images, despite being time consuming and observer dependent. Previous automatic methods generally do not account for the long-axis motion, which is a major contributor to the stroke volume. Therefore, the aim of this study was to develop and validate an automatic algorithm for time-resolved segmentation covering the whole LV, including basal slices moving out of the imaging plane during systole.
Methods
Ninety subjects were imaged using a routinely applied cine balanced steady state free precession (bSSFP) sequence (training set n=40, test set n=50).
The automatic algorithm uses deformable model with Expectation-Maximization (EM), followed by automatic removal of papillary muscles and detection of the outflow tract.
The reference standard was manual delineation and second observer analysis was performed in a subset of patients (n=25).
Results
Typical automatic segmentation and reference manual delineation is shown in three selected slices in Figure 1. The differences between automatic segmentation and reference manual delineation were EDV -2.4±12.8 ml, ESV 9.1±12.3 ml, EF -5.4±4.7 % and LVM -3.5±19.5 g (Figure 2, n=49 after exclusion due to severe artefacts in 1 patient). In the second observer sub set differences between second observer manual delineation and reference manual delineation were EDV 10.5 ± 4.4 ml, ESV 5.1 ± 4.7 ml, EF -0.4 ± 2.3 % and LVM -7.4 ± 8.9 g compared to the differences between automatic segmentation and the reference manual delineation which were EDV -1.0 ± 9.6 ml, ESV 9.2 ± 10.0, EF -4.9 ± 4.0 % and LVM -10.5 ± 17.5 g.
Conclusions
The proposed automatic LV segmentation algorithm reached accuracy comparable to that found between observers, taking automatic segmentation one step closer to clinical routine. The algorithm and all images with manual delineations will be made available for benchmarking.
Funding
This study has been funded by the Swedish Research Council, The Swedish Heart and Lung Foundation, The Medical Faculty of Lund University, Sweden, and Region of Scania, Sweden.
Author information
Authors and Affiliations
Rights and permissions
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
About this article
Cite this article
Tufvesson, J., Hedström, E., Steding-Ehrenborg, K. et al. Validation and development of a new automatic algorithm for time resolved segmentation of the left ventricle in magnetic resonance imaging. J Cardiovasc Magn Reson 17 (Suppl 1), P68 (2015). https://doi.org/10.1186/1532-429X-17-S1-P68
Published:
DOI: https://doi.org/10.1186/1532-429X-17-S1-P68