Cardiac material markers from tagged MR images

Area of Science:

• Biomedical Imaging
• Cardiovascular Mechanics
• Medical Image Analysis

Background:

• Tagged magnetic resonance imaging (MRI) shows potential for non-invasive heart motion analysis.
• Current methods require high accuracy to replace invasive implanted markers.
• Tracking sparse material points is crucial for detailed cardiac motion assessment.

Purpose of the Study:

• To develop and validate a new method for accurate motion estimation of sparse material points using tagged MRI.
• To enable tagged MRI to serve as a gold standard for non-invasive cardiac motion tracking.
• To ensure compatibility of generated data with existing implanted marker analysis applications.

Main Methods:

• Utilizes standard, parallel-tagged MR images for motion estimation.
• Employs thin-plate splines to estimate tag surfaces.
• Determines material marker positions at tag surface intersections via an iterative alternating projections algorithm.
• Provides a proof of convergence for the algorithm.

Main Results:

• Successfully tracks material markers with high accuracy.
• Demonstrates compatibility with established marker data analysis pipelines.
• Achieves an RMS error of approximately 0.2 mm in a simulated left ventricle model under typical conditions.
• Includes detailed visualization and numerical results from a human volunteer study.

Conclusions:

• The presented method offers accurate and reliable non-invasive tracking of cardiac motion using tagged MRI.
• This technique advances the potential of tagged MRI to replace invasive methods for cardiac motion analysis.
• The approach is robust and validated through simulation and in vivo data.