Automated quantitative analysis of MCC-IMS spectra

The detection and quantification of Volatile Organic Compounds (VOCs) using a very sensitive ion mobility spectrometry (IMS) with added selectivity by pre-separation in Multi-Capillar Column (MCC) is a promising tool of direct metabolic profiling of human breath for medical diagnosis or therapy monitoring. Several algorithms for MCC-IMS individual peak detection have been proposed and described in the literature. However, some of the metabolites usually present in human breath show the overlapping of the peaks in the 3D-Chromatogram, making its detection and, especially quantitative analysis, extremely difficult and time consuming. In this work we present an algorithm for automatic quantitative analysis of the MCC-IMS spectral data. The proposed algorithm uses image processing techniques to detect spectral peaks and Levenberg-Marquardt optimization algorithm to fit a custom mathematical model to those detected peaks. From the produced model, VOCs features, such as position, amplitude and integral, are extracted and calculated using a priori calibration curves. For the test we use overlapping peaks of Isoprene and Acetone, as two of the most studied VOCs in the breath analysis, but determination of its exact concentration is still very important.