Multi-view active appearance models for consistent segmentation of multiple standard views: Application to long- and short-axis cardiac MR images

Abstract

This paper describes a multi-view Active Appearance Model (AAM) for coherent segmentation of multiple cardiac views. Cootes' AAM framework was adapted by considering shapes and intensities from multiple views as single shape and intensity samples, while eliminating trivial difference in object pose in different views. This way, the coherence in organ shape and intensities between different views is modeled and utilized during image search. In a leave-one-out validation study on a combination of four-chamber, two-chamber and short-axis cardiac MR views from 29 patients, low border positioning errors were found comparing manual and computer detected borders, whereas no statistically significant difference in contour area was present. The presented validation shows that the method combines a high robustness with clinically acceptable accuracy, and therefore is a promising tool to further automate the integral quantitative analysis of cardiac MR patient examinations. © 2003, Elsevier Science B.V. All rights reserved.