| 摘要 |
A method for cardiac segmentation includes providing a plurality of digitized medical images, each image comprising a plurality of intensities corresponding to a domain of points on an 2-dimensional grid, the plurality of images forming a sequence of cardiac images acquired at different time points, filtering noise from the sequence of images, approximating the endocardium with a first circle, approximating the epicardium with a second circle, finding candidate points for the endocardium, finding candidate points for the epicardium, and interpolating gaps between points of the endocardium and between points of the epicardium wherein the myocardium is segmented.
|
