QUANTITATIVE-ANALYSIS OF VOLATILE ORGANIC-COMPOUNDS USING ION MOBILITY SPECTROMETRY AND CASCADE CORRELATION NEURAL NETWORKS

Ion mobility spectrometry (IMS) has a limited linear range. Nonlinear calibration methods, such as neural networks are ideally suited for IM S due to their capability of modeling complex systems. Many neural net works suffer from long training times and overfitting. Cascade correla tion neural networks (CCN) are interesting, because they train at fast rates. Another advantage of CCNs is that they automatically configure their own topology (number of layers and number of units in each laye r). By using a the decay parameter in training neural networks, reprod ucible and general models may be obtained at the cost of longer traini ng times. CCN networks were trained to furnish both quantitative and q ualitative prediction for a complex IMS data set (229 spectra, 200 inp ut points, and 15 output classes). The advantage of rapid training is that replicate neural networks may be obtained. The precision of repli cated network predictions appears to provide a measure of accuracy. Pa rtial least-squares regression (PLS) is used as a comparative method. The CCN with decay rates an order of magnitude larger than learning ra te achieves significantly better results than those obtained from an o ptimal PLS model.