Impedimetric biofilm characterization with microelectrode arrays using equivalent electrical circuit features and ensemble classifiers

Electrochemical impedance spectroscopy has proven to be a promising technique for detecting bacterial biofilms. However, its potential for microbial identification has yet to be thoroughly investigated. In this work, we explore the classification of bacterial biofilms at both the strain and species level using commercial microelectrode arrays. Here, we built predictive decision tree ensemble classifiers based on impedance spectroscopy measurements and the extracted equivalent electrical circuit model features. We evaluated this experimental and modelling strategy for classifying a selection of bacterial (sub)species' biofilms relevant to food processing and healthcare. The inclusion of equivalent electrical circuit features in the models consistently improves the classification performance by 2% to 11%. At the same time, a further robust improvement of 3% to 9% is achieved through the use of microelectrode arrays as opposed to single electrodes.