Abstract: Training a neural network for signal analysis in NMR spectra may use a plurality of computed NMR raw spectra, each raw spectrum being associated with a different NMR active molecule having a known number of protons (#P). Peaks of the raw spectra may be broadened to generate a broadened spectrum for each raw spectrum. For each broadened spectrum, its integral function may be computed to count the number of protons associated with peaks of the respective broadened spectrum. Signal intervals may be identified as intervals in the broadened spectrum where the integral function increases approximately by multiples of the value associated with a single proton so that the total number of counted protons matches the known number of protons (#P) of the associated molecule. The obtained spectra with associated labels for the identified signal intervals are provided as the training data set to the neural network.

Abstract: Training a neural network for signal analysis in NMR spectra may use a plurality of computed NMR raw spectra, each raw spectrum being associated with a different NMR active molecule having a known number of protons (#P). Peaks of the raw spectra may be broadened to generate a broadened spectrum for each raw spectrum. For each broadened spectrum, its integral function may be computed to count the number of protons associated with peaks of the respective broadened spectrum. Signal intervals may be identified as intervals in the broadened spectrum where the integral function increases approximately by multiples of the value associated with a single proton so that the total number of counted protons matches the known number of protons (#P) of the associated molecule. The obtained spectra with associated labels for the identified signal intervals are provided as the training data set to the neural network.