A study on multi-exponential inversion of nuclear magnetic resonance relaxation data using deep learning

Luo, Gang; Xiao, Lizhi; Luo, Sihui; Liao, Guangzhi; Shao, R.

doi:10.1016/j.jmr.2022.107358

Cited by 7 publications

(3 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The foundational architecture of DLEMLR consists of a classic transformer network featuring multilayer perception (MLP) 31,32 and multihead attention (MHA). 33,34 Compared to the widely applied convolution layer, MLP has strong representation ability crucial for domain transformation tasks in Laplace NMR reconstruction, which has been proven essential for T 2 estimation 29 and DOSY reconstruction. 30 Transformers excel in processing sequential data, but they face challenges when handling multidimensional data.…”

Section: ■ Methodsmentioning

confidence: 99%

“…Figure shows the flowchart of DLEMLR, a deep-learning-based Laplace NMR processing method designed to uncover fundamental patterns and relationships between input data and the target spectrum, facilitating accurate predictions for both new and previously unseen data. The foundational architecture of DLEMLR consists of a classic transformer network featuring multilayer perception (MLP) , and multihead attention (MHA). , Compared to the widely applied convolution layer, MLP has strong representation ability crucial for domain transformation tasks in Laplace NMR reconstruction, which has been proven essential for T 2 estimation and DOSY reconstruction …”

Section: Methodsmentioning

confidence: 99%

“…Deep learning is a powerful artificial intelligence technique that is successfully applied to various tasks of NMR data processing, e.g., spectral reconstruction, , spectral denoising, − and automated peak picking. − Besides, deep learning provides new insight into addressing the intricate problem of Laplace NMR processing. , Compared with traditional Laplace processing methods, deep-learning-based methods require no deliberate parameter adjustment, making it not only more robust to different experimental conditions but also capable of reducing computational time. While existing studies have effectively demonstrated the potential of deep learning in handling Laplace NMR data, so far, it has only been limited to 1D scenarios.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Two-Dimensional Laplace NMR Reconstruction through Deep Learning Enhancement

Chen,

Fang,

Zhang

et al. 2024

J. Am. Chem. Soc.

View full text Add to dashboard Cite

Laplace NMR is a powerful tool for studying molecular dynamics and spin interactions, providing diffusion and relaxation information that complements Fourier NMR used for composition determination and structure elucidation. However, Laplace NMR demands sophisticated signal processing algorithms such as inverse Laplace transform (ILT). Due to the inherently illposed nature of ILT problems, it is generally challenging to perform satisfactory Laplace NMR processing and reconstruction, particularly for two-dimensional Laplace NMR. Herein, we propose a proof-of-concept approach that blends a physics-informed strategy with data-driven deep learning for two-dimensional Laplace NMR reconstruction. This approach integrates prior knowledge of mathematical and physical laws governing multidimensional decay signals by constructing a forward process model to simulate relationships among different decay factors. Benefiting from a noniterative neural network algorithm that automatically acquires prior information from synthetic data during training, this approach avoids tedious parameter tuning and enhances user friendliness. Experimental results demonstrate the practical effectiveness of this approach. As an advanced and impactful technique, this approach brings a fresh perspective to multidimensional Laplace NMR inversion.

show abstract