Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

Direct deduction of chemical class from NMR spectra.

In Journal of magnetic resonance (San Diego, Calif. : 1997)

This paper presents a proof-of-concept method for classifying chemical compounds directly from NMR data without performing structure elucidation. This can help to reduce the time in finding good structure candidates, as in most cases matching must be done by a human engineer, or at the very least a process for matching must be meaningfully interpreted by one. The method identified as suitable for classification is a convolutional neural network (CNN). Other methods, including clustering and image registration, have not been found to be suitable for the task in a comparative analysis. The result shows that deep learning can offer solutions to spectral interpretation problems.

Kuhn Stefan, Cobas Carlos, Barba Agustin, Colreavy-Donnelly Simon, Caraffini Fabio, Borges Ricardo Moreira

2023-Jan-21

Chemical classification, Convolutional neural network, Deep learning, Image processing, NMR