Spectrum Reconstruction for Laplace NMR: From Handcraft Regularization to Deep Learning

Abstract

Laplace NMR can provide information on diffusion coefficients or relaxation time, serving as a powerful technology for studying molecular structure, dynamics, and interactions in samples. Generally, the applicability of Laplace NMR is subject to the performance of signal processing and reconstruction algorithms associated with the inverse Laplace transform. In this paper, we first discuss the ill-posed nature of the spectrum reconstruction problem for Laplace NMR, then revisit the classic regularization-based reconstruction algorithms and introduce the state-of-the-art deep-learning-based methods. In conclusion, the advantages and disadvantages of these algorithms are summarized, and future improvements for Laplace NMR signal processing methods are prospected.

Keywords

• laplace nmr
• diffusion nmr
• dosy (diffusion ordered spectroscopy)
• ilt (inverse laplace transform)
• deep learning