Attention-Based Fusion of Infrared and Mass Spectra for Identification of Chemical Threats BY Functional Group Prediction

Résumé fourni par la source

ApproachDetermining molecular structure through optical spectroscopy (Infrared, Raman) or Mass Spectrometry is challenging, requiring expert-level knowledge.Machine-Learning based approaches have gained interest as a solution to automating molecular identification, but still struggle to decode the subtleties contained within spectral data.To address this issue, we turn to sensor fusionbased approaches, leveraging multi-modal ML models capable of combining both IR and mass spectra to identify functional groups.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.