ANALYZING PARTICULATE MATTERS VIA SURFACTANT-ASSISTED MICROFLUIDIC IONIC CURRENT SENSING WITH MACHINE LEARNING-DRIVEN IDENTIFICATION

Exposures to particulate matters (PMs) are one of important factors for human health, however, their risks are little known due to lacks of comprehensive sensing methods that can access their physicochemical properties. Here, we developed a microfluidics-based method to characterize size and compositions of PMs via combining surfactant-assisted single particle detection, presented on MicroTAS 2020 [1], with machine learning (ML)-driven identifications. Both of hydrophilic and hydrophobic particles were sensed and their electrical signals were discriminated with 98% accuracy at the single level. Our method will comprehensively sense real PMs to characterize their physicochemical properties, enabling to understand health risks.