Peak detection algorithm based on second derivative properties for two dimensional ion mobility spectrometry signals

Abstract

In this paper we propose a novel peak detection algorithm for 2-dimensional analytical data. The proposed algorithm utilizes the properties of the second derivative and curvature of (regular) surfaces to perform peak detection. Raw data used in this study for performance demonstration were obtained by a hyphenated system called gas-chromatographic column ion mobility spectrometer (GC-IMS). GC-IMS is a two stage technique for separation of gas-phased (organic) compounds. Despite the good performance of the MCC-IMS separation in general, there are still unsatisfactory cases where the substances overlap and the recorded signals nearly merge. Frequently used peak detection algorithm for 2-dimensional data, like the watershed algorithm, do not perform well in those cases. Preliminary empirical results show good peak detection performance of the proposed algorithm. Furthermore the results indicate that the proposed algorithm is capable to solve the problem of peak detection even in cases of strong peak overlapping.