Acceleration of multi-dimensional propagator measurements with compressed sensing

Area of Science:

• Materials Science
• Biophysics
• Chemistry

Background:

• Nuclear Magnetic Resonance (NMR) is crucial for probing microstructures in anisotropic materials like liquid crystals, polymers, and tissues.
• Diffusion propagator measurements reveal internal structures, but multi-dimensional analysis is often time-prohibitive due to extensive data sampling requirements.

Purpose of the Study:

• To demonstrate the effectiveness of compressed sensing for accelerating multi-dimensional NMR diffusion propagator measurements.
• To assess the impact of reduced sampling on the accuracy of reconstructing microscopic anisotropy in materials.

Main Methods:

• Implementing a compressed sensing (CS) acquisition scheme for NMR diffusion propagator measurements.
• Comparing results from incompletely sampled data (down to 64x sub-sampling) with full sampling.
• Evaluating the reconstruction accuracy of the diffusion propagator for anisotropic diffusion.

Main Results:

• Compressed sensing significantly reduces the required data sampling for diffusion propagator measurements.
• Experimental acquisition time for 3D experiments was reduced by a factor of 32 (from ~80 days to ~2.5 days).
• Reconstruction of the diffusion propagator showed minimal loss of information despite substantial sub-sampling.

Conclusions:

• Compressed sensing is an effective technique for accelerating NMR experiments on anisotropic materials.
• This acceleration allows for more efficient investigation of microscopic anisotropy and internal structures.
• The study validates CS as a viable method for reducing experimental time without compromising scientific outcomes.