Automatic Construction of 3D Statistical Deformation Models Using Non-rigid Registration
MICCAI '01 Proceedings of the 4th International Conference on Medical Image Computing and Computer-Assisted Intervention
Robust Real-Time Face Detection
International Journal of Computer Vision
Physically-Constrained Diffeomorphic Demons for the Estimation of 3D Myocardium Strain from Cine-MRI
FIMH '09 Proceedings of the 5th International Conference on Functional Imaging and Modeling of the Heart
Haar-like features with optimally weighted rectangles for rapid object detection
Pattern Recognition
Robust segmentation of brain structures in MRI
ISBI'09 Proceedings of the Sixth IEEE international conference on Symposium on Biomedical Imaging: From Nano to Macro
Hi-index | 0.00 |
Tagged magnetic resonance (MR) imaging is unique in its ability to noninvasively image the motion and deformation of the heart. However, it is difficult to identify and quantify structures of interest in the cardiac anatomy since the tags obscure the anatomy. In this paper, we present a novel and fully automated technique based on nonrigid image registration for the analysis of myocardial motion using both tagged and untagged MR images. The novel aspect of our technique is its simultaneous usage of complementary information from both tagged and untagged images. No manual intervention is required to obtain the segmentation of the end-diastolic images. To estimate the motion within the myocardium, we register a sequence of images taken during systole to a set of reference images taken at end-diastole, maximizing a spatial weighted similarity measure between the images. We use short-axis and long-axis images of the heart as well as tagged and untagged images to estimate a fully four-dimensional motion field within the myocardium. We have evaluated the proposed approach on 8 patients both in terms of robustness, accuracy and consistency of the motion tracking. The proposed method is significantly more consistent than motion tracking on tagged MR images only.