Dynamic Imaging and Characterization of Volatile Aerosols in E-Cigarette Emissions Using Deep Learning-Based Holographic Microscopy

Various volatile aerosols have been associated with adverse health effects; however, characterization of these is challenging due to their dynamic nature. Here we present a method that directly measures the volatility particulate matter (PM) using computational microscopy and deep learning. This was applied generated by electronic cigarettes (e-cigs), which vaporize liquid mixture (e-liquid) mainly consists propylene glycol (PG), vegetable glycerin (VG), nicotine, flavoring compounds. E-cig were recorded field-portable microscope, an impaction-based air sampler. A lensless digital holographic microscope inside this mobile device continuously records inline holograms collected particles. learning-based algorithm used automatically reconstruct microscopic images e-cig particles from holograms, rapidly quantify volatility. To evaluate effects e-liquid composition on aerosol dynamics, measured flavorless, nicotine-free e-liquids various PG/VG volumetric ratios, revealing negative correlation between particles' ratio VG in e-liquid. For given composition, addition nicotine dominated evaporation dynamics aforementioned no longer observed. We also revealed additives significantly decrease aerosol. The presented measurement technique might provide new insights can be characterize PM.