PEAK TRACKING WITH A NEURAL NETWORK FOR SPECTRAL RECOGNITION

A peak tracking method based on a simulated feed-forward neural network with back-propagation is presented. The network uses the normalized UV spectra and peak areas measured in one chromatogram for peak recognition. It suffices to train the network with only one set of spectra recorded in one chromatogram of a sample to recognize the sample peaks in other chromatograms recorded in different mobile phases. The peak recognition method was used in a mixture design-based optimization of the separation of eight sulphonamides, some of which have very similar spectra. Even peaks in a cluster of four to five overlapping peaks could be recognized.