Left ventricular motion reconstruction with a prolate spheroidal B-spline model

Area of Science:

• Cardiovascular imaging
• Biomedical engineering
• Medical physics

Background:

• Tagged cardiac magnetic resonance (MR) imaging enables non-invasive assessment of left ventricular (LV) wall motion.
• Accurate three-dimensional (3D) analysis of myocardial deformation from tagged MR data requires sophisticated modeling of tag line behavior.

Purpose of the Study:

• To introduce and evaluate a novel 3D myocardial displacement and strain reconstruction method using a prolate spheroidal B-spline (PSB) deformation model.
• To compare the performance of the PSB model against existing methods for LV deformation analysis.

Main Methods:

• Development of a 3D deformation model based on B-splines defined in prolate spheroidal coordinates, tailored to the LV geometry.
• Implementation of the PSB reconstruction algorithm for analyzing tagged MR image data.
• Validation of the PSB method using a previously published dataset for direct comparison with other techniques.

Main Results:

• The prolate spheroidal coordinate system provides a better fit to the LV wall shape compared to Cartesian or cylindrical models.
• The PSB model ensures smoothness of deformation and enables strain computation at the cardiac apex.
• Head-to-head comparison demonstrated the accurate reconstruction capabilities of the PSB method.

Conclusions:

• The PSB method accurately reconstructs myocardial deformation and strain from tagged MR images.
• The PSB approach offers significant advantages over existing LV deformation reconstruction techniques.
• This method enhances the quantitative analysis of cardiac mechanics using tagged MR imaging.