TY - JOUR
T1 - Direct deduction of chemical class from NMR spectra
AU - Kuhn, Stefan
AU - Cobas, Carlos
AU - Barba, Agustin
AU - Colreavy-Donnelly, Simon
AU - Caraffini, Fabio
AU - Borges, Ricardo Moreira
N1 - Publisher Copyright:
© 2023 Elsevier Inc.
PY - 2023/3
Y1 - 2023/3
N2 - 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.
AB - 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.
KW - Chemical classification
KW - Convolutional neural network
KW - Deep learning
KW - Image processing
KW - NMR
UR - http://www.scopus.com/inward/record.url?scp=85146724605&partnerID=8YFLogxK
U2 - 10.1016/j.jmr.2023.107381
DO - 10.1016/j.jmr.2023.107381
M3 - Article
C2 - 36706464
AN - SCOPUS:85146724605
SN - 1090-7807
VL - 348
SP - 107381
JO - Journal of Magnetic Resonance
JF - Journal of Magnetic Resonance
M1 - 107381
ER -