Electronic Nose for Indoor Mold Detection and Identification

Indoor mold infestations lead to adverse effects on air quality and thus pose significant health risks to humans. Traditional methods for mold detection and identification are time-consuming and costly. In this study, the application of an electronic nose as a highly reliable tool for detecting and identifying mold is explored. Two common indoor mold species, Stachybotrys chartarum and Chaetomium globosum, each separately grown on two different substrates, are investigated. Our e-nose uses vaporliquid-solid-grown, UV-activated SnO2 nanowires as the chemiresistive sensing material. Linear discriminant analysis (LDA) is
used for classification. Moreover, novelty detection is enabled by default using decision boundaries. While the conventional LDA only shows mediocre classification results, improved versions can achieve an average F1-score of 98.37%. Therefore, our results demonstrate that the e-nose can not only detect but also identify different mold genera, and thus making a significant step toward fast, objective, and cost-effective indoor air quality monitoring.