TOMOGRAPHIC RECONSTRUCTION OF DIFFUSION PROPAGATORS FROM DW-MRI USING OPTIMAL SAMPLING LATTICES.

Area of Science:

• Medical Imaging
• Applied Mathematics
• Signal Processing

Background:

• Diffusion Weighted Magnetic Resonance Imaging (DWMRI) is crucial for understanding water diffusion in biological tissues.
• Reconstructing the diffusion propagator accurately is essential for quantitative analysis in DWMRI.
• Current methods often require a large number of samples, impacting acquisition time and efficiency.

Purpose of the Study:

• To investigate the use of optimal sampling lattices for improved tomographic reconstruction of the diffusion propagator in DWMRI.
• To evaluate the trade-off between the number of samples and reconstruction accuracy using optimal sampling geometries.
• To leverage information-theoretic advantages of sphere packing lattices for multidimensional signal sampling.

Main Methods:

• Exploitation of optimal sampling lattices derived from sphere packing principles.
• Application of a tomographic reconstruction approach, building upon the Pickalov and Basser method.
• Comparative analysis of diffusion propagator reconstructions using both Cartesian and optimal sampling geometries.
• Validation with synthetic and real DWMRI data sets.

Main Results:

• Optimal sampling lattices significantly increase the accuracy of tomographic reconstruction for the diffusion propagator.
• The proposed optimal sampling geometry allows for a reduction in the number of required samples without compromising reconstruction accuracy.
• Demonstrated superior performance of optimal sampling compared to traditional Cartesian sampling in DWMRI reconstruction.
• Validation of findings on both simulated and experimental DWMRI data.

Conclusions:

• Optimal sampling lattices offer a powerful approach to enhance the accuracy and efficiency of diffusion propagator reconstruction in DWMRI.
• The information-theoretic benefits of sphere packing lattices are effectively integrated into tomographic reconstruction for DWMRI.
• This methodology presents a promising direction for optimizing DWMRI data acquisition and analysis.