论文信息 - Left Ventricular Segmentation Challenge from Cardiac MRI: A Collation Study

Left Ventricular Segmentation Challenge from Cardiac MRI: A Collation Study

This paper presents collated results from the left ventricular (LV) cardiac MRI segmentation challenge as part of STACOM'11. Clinical cases from patients with myocardial infarction (100 test and 100 validation cases) were randomly selected from the Cardiac Atlas Project (CAP) database. Two independent sets of expert (manual) segmentation from different sources that are available from the CAP database were included in this study. Automated segmentations from five groups were contributed in the challenge. The total number of cases with segmentations from all seven raters was 18. For these cases, a ground truth "consensus" segmentation was estimated based on all raters using an Expectation-Maximization (EM) method (the STAPLE algorithm).

[1] Alistair A. Young,et al. The Cardiac Atlas Project—an imaging database for computational modeling and statistical atlases of the heart , 2011, Bioinform..

[2] W. J. Hedley,et al. Left ventricular mass and volume: fast calculation with guide-point modeling on MR images. , 2000, Radiology.

[3] Xiaoguang Lu,et al. Automatic Segmentation of the Myocardium in Cine MR Images Using Deformable Registration , 2011, STACOM.

[4] Carissa G. Fonseca,et al. Rationale and Design for the Defibrillators to Reduce Risk by Magnetic Resonance Imaging Evaluation (DETERMINE) Trial , 2009, Journal of cardiovascular electrophysiology.

[5] William M. Wells,et al. Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation , 2004, IEEE Transactions on Medical Imaging.

[6] Ayman M. Khalifa,et al. Myocardial Segmentation Using Contour-Constrained Optical Flow Tracking , 2011, STACOM.

[7] Nicholas Ayache,et al. Layered Spatio-temporal Forests for Left Ventricle Segmentation from 4D Cardiac MRI Data , 2011, STACOM.