Machine learning and signal processing assisted differential mobility spectrometry (DMS) data analysis for chemical identification

Abstract

The convolutional neural algorithm outperforms previously reported algorithms, and MSC approach needs minimal data for chemical identification.

Insights into modifiers effects in differential mobility spectrometry: A data science approach for metabolomics and peptidomics

“…Bissonnette et al 22 used first‐principles kinetics‐based model, together with MobCal‐MPI and DFT/DLPNO‐CCSD(T) to study binary solvent mixtures in DMS, with conclusion that the differential mobility of ions is predominantly influenced by the solvent binding energies with a secondary contribution from solvent size. Chakrabort et al 23 implemented convolutional neural networks and signal processing techniques like magnitude‐squared coherence in DMS data analysis, achieving high accuracy in identifying pure chemicals and their mixtures, offering an efficient approach for chemical identification in various applications.…”