Principal Warps: Thin-Plate Splines and the Decomposition of Deformations
IEEE Transactions on Pattern Analysis and Machine Intelligence
Maintaining Multi-Modality through Mixture Tracking
ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking
IEEE Transactions on Signal Processing
Trans-dimensional MCMC methods for fully automatic motion analysis in tagged MRI
MICCAI'11 Proceedings of the 14th international conference on Medical image computing and computer-assisted intervention - Volume Part I
Hi-index | 0.01 |
Myocardial tagging using magnetic resonance imaging (MRI) is a well-known noninvasive method for studying regional heart dynamics. While it offers great potential for quantitative analysis of a variety of kinematic and kinetic parameters, its clinical use has so far been limited, mainly due to mediocre performance of existing tag tracking algorithms under poor imaging conditions. In this paper we propose a new approach to tracking of MRI tag intersections. It is based on a Bayesian estimation framework, implemented by means of particle filtering, and combines information about heart dynamics, the imaging process, and tag appearance. Since at any time point it optimally incorporates all available information, it can be expected to be more robust and accurate. This is demonstrated by results of preliminary experiments on image sequences from (small) animal imaging studies.