Structure Prior Constrained Estimation of Human Cardiac Diffusion Tensors.

Area of Science:

• Medical Imaging
• Biophysics
• Cardiovascular Research

Background:

• Diffusion Magnetic Resonance Imaging (dMRI) is crucial for non-invasive assessment of tissue microstructure.
• Accurate estimation of cardiac diffusion tensor (DT) is vital for understanding heart function and disease.
• Current dMRI methods for cardiac DT estimation face challenges with limited diffusion gradient directions.

Purpose of the Study:

• To develop and validate a novel method for improving the accuracy of human cardiac diffusion tensor (DT) estimation.
• To enhance DT estimation in diffusion magnetic resonance imaging (dMRI) using a reduced number of diffusion gradient directions.
• To provide a more robust and reliable approach for cardiac microstructure analysis.

Main Methods:

• A Structure Prior Constrained (SPC) method was developed, incorporating novel regularizers into a nonlinear least squares estimator.
• Regularizers were designed to penalize dissimilarity between neighboring DTs and deviations from prior fiber orientations.
• A unique numerical solution was implemented to ensure positive definite DT estimations.

Main Results:

• The SPC method demonstrated accurate DT estimation in ex vivo human cardiac data across most voxels.
• Performance evaluation showed the SPC method outperformed state-of-the-art techniques in estimating key diffusion parameters.
• Significant improvements were observed in the accuracy of principal eigenvector, second eigenvector, helix angle, transverse angle, fractional anisotropy, and mean diffusivity.

Conclusions:

• The Structure Prior Constrained (SPC) method offers a practical and dependable alternative for human cardiac DT estimation.
• The SPC method achieves high accuracy in cardiac DT estimation using dMRI with few diffusion gradient directions.
• This advancement facilitates more precise non-invasive analysis of cardiac tissue microstructure.