Improved data efficiency for NMR diffusion-relaxation processing

Area of Science:

• Nuclear Magnetic Resonance (NMR) Spectroscopy
• Materials Science
• Chemical Engineering

Background:

• Two-dimensional diffusion and transverse (T2) NMR relaxation measurements are crucial for research and industrial applications.
• Converting these measurements into D-T2 maps typically involves inverse integral transformations.
• A key challenge arises from coupled diffusion and relaxation kernels when data is acquired without pulsed field gradients, complicating D-T2 estimation.

Purpose of the Study:

• To present novel data processing methods for D-T2 estimation that avoid the data and signal energy wastage associated with conventional time-offset techniques.
• To enable the use of all available data in D-T2 mapping, thereby improving accuracy and feature discovery.
• To address the limitations of current methods in handling coupled diffusion and relaxation kernels.

Main Methods:

• Development of two new data processing techniques based on the linear operator of the forward integral transformation.
• Implementation of a data compression method.
• Application of the fast iterative soft thresholding algorithm, termed FLINT, for D-T2 estimation.

Main Results:

• Both presented methods successfully utilize all available data, unlike traditional approaches that discard signal energy.
• The methods demonstrate improved accuracy in D-T2 estimation when applied to simulated data.
• Enhanced discovery of significant features is observed in measured data processed with these new techniques.

Conclusions:

• The novel data processing methods offer a significant advancement for D-T2 mapping from NMR relaxation measurements.
• These techniques provide a more efficient and accurate way to extract diffusion and relaxation information, particularly when pulsed field gradients are not used.
• The FLINT algorithm and data compression approach represent valuable tools for researchers and industrial practitioners utilizing NMR spectroscopy.