PARTICLE SIZE DETERMINATION VIA SUPERVISED MACHINE LEARNING IN MICROFLUIDIC IMPEDANCE SPECTROSCOPY

Impedance flow cytometers based on coplanar electrodes often have a significant signal dependency on the particle position. In this work we show that supervised machine learning can be employed to accurately predict the particle size of monodisperse polystyrene beads in an inhomogeneous electric field. This approach offers accurate results for the presented irregular signal shape (due to sensor geometry, particle position, and electrode alignment) without the need for signal template fitting and a compensation function.