YingLi Lu1, Perry Radau1, Kim A.
Connelly1,2, Alexander Dick3, Graham A. Wright1
1Imaging Research, Sunnybrook Health
Sciences Centre, Toronto, ON, Canada; 2Cardiology, St Michael's
Hospital, Toronto, ON, Canada; 3Cardiology, Sunnybrook Health
Sciences Centre, Toronto, ON, Canada
Purpose
of this study is to develop a fully automatic left ventricle segmentation
method from cine short-axis MR images and evaluate it on a large data set of
147 subjects grouped by pathology. Advantages of this method include that it:
1) does not require manually drawn contours; 2) provides not only endocardial
and epicardial contours, but also papillary muscles and trabeculations
contours; 3) introduces a roundness measure that automatically locates the
left ventricle; 4) simplifies the epicardial contour segmentation by mapping
the pixels from Cartesian to approximately polar coordinates.