Related Experiment Video
Updated: Aug 8, 2026
07:21
Magnetic Resonance Derived Myocardial Strain Assessment Using Feature Tracking
Published on: February 12, 2011
Myocardial motion estimation in tagged MR sequences by using alphaMI-based non rigid registration
E Oubel1, C Tobon-Gomez, A O Hero
1Computational Imaging Laboratory, Pompeu Fabra University, Barcelona, Spain. estanislao.oubel@upf.edu
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|May 12, 2006
Summary
This study explores using high-dimensional Haar wavelet features with alphaMI for cardiac deformation analysis via non-rigid registration, achieving comparable results to traditional methods.
Area of Science:
- Medical Imaging
- Biomedical Engineering
- Cardiovascular Research
Background:
- Tagged Magnetic Resonance Imaging (MRI) is the standard for myocardial motion and strain analysis.
- Non-rigid registration methods, particularly Normalized Mutual Information (NMI)-based, accurately retrieve cardiac deformation fields.
- Current methods face limitations in incorporating higher-dimensional features for enhanced myocardial deformation estimation.
Purpose of the Study:
- To investigate the feasibility of using high-dimensional Haar wavelet features with alphaMI for non-rigid image registration in myocardial motion and strain analysis.
- To evaluate the performance of this novel approach compared to traditional NMI-based registration.
- To introduce Entropic Spanning Graphs (ESGs) as a method for estimating alphaMI with high-dimensional feature vectors.
Main Methods:
- Utilized alphaMI with high-dimensional Haar wavelet feature vectors (WFVs) for non-rigid image registration.
- Employed Entropic Spanning Graphs (ESGs) to estimate alphaMI for WFVs, overcoming limitations of histogram-based methods.
- Compared the performance of the alphaMI-based registration with high-dimensional features against traditional NMI-based registration.
Main Results:
- Demonstrated the feasibility of implementing high-dimensional Haar wavelet features with alphaMI for myocardial deformation estimation.
- Observed no significant performance degradation compared to traditional NMI-based registration.
- Successfully applied ESGs for alphaMI estimation of WFVs in the context of non-rigid registration.
Conclusions:
- The integration of high-dimensional Haar wavelet features with alphaMI is a viable approach for myocardial deformation analysis.
- Entropic Spanning Graphs represent a novel and effective tool for estimating alphaMI with complex feature vectors in medical image registration.
- This study pioneers the use of ESGs for non-rigid registration, opening new avenues for advanced cardiac motion analysis.
