Towards in vivo diffusion tensor MRI on human heart using edge-preserving regularization.

Area of Science:

• Cardiovascular Imaging
• Biomedical Engineering
• Medical Physics

Background:

• Diffusion Tensor MRI (DT-MRI) is crucial for assessing cardiac tissue microstructure.
• Noise sensitivity in DT-MRI acquisition protocols can limit image quality and accuracy.
• Optimizing acquisition parameters is essential for reliable in vivo cardiac DT-MRI.

Purpose of the Study:

• To evaluate noise sensitivity in various DT-MRI protocols for ex vivo human hearts.
• To compare the accuracy of different acquisition parameters for estimating principal diffusion directions.
• To introduce a novel regularization method to improve DT-MRI quality and reduce acquisition time.

Main Methods:

• Investigated sixteen human ex vivo hearts using various DT-MRI acquisition protocols.
• Compared protocols with differing numbers of excitations and diffusion sensitizing directions.
• Developed and applied a 3-D edge-preserving regularization method on diffusion-weighted images.

Main Results:

• Noise sensitivity decreased with increased excitations and diffusion directions.
• Higher numbers of excitations and directions correlated with longer acquisition times.
• The regularization method maintained principal diffusion direction field quality while minimizing acquisition time.

Conclusions:

• Acquisition time in cardiac DT-MRI is directly related to noise reduction.
• The proposed regularization technique enhances DT-MRI feasibility for in vivo human heart imaging.
• This method allows for reduced scan times without compromising diagnostic quality.