Quantitative determination of mold growth and inhibition by multispectral imaging

Quantifying mold growth is of great relevance and interest in microbiology. However, predictive modeling of filamentous fungal growth has been hampered by the lack of an appropriate and accurate method for quantification. An adequate, rapid and objective method will allow studying the effect of many different parameters and conditions on mold growth patterns and can thus provide valuable insight and knowledge. This study outlines a new approach for quantifying mold growth by providing an accurate tool for measuring different segments of mold colonies. The method is based on clustering multispectral images by k-means, an unsupervised and simple clustering algorithm. In order to demonstrate the efficiency of the new approach, three different sample sets were analyzed by the developed method, with the objective of quantifying mold growth and size of the colony segments of Penicillium mold. The results verify the ability of the proposed method to quantify mold growth and colony composition (relative size of the white and green segments) accurately. This provides a robust measure for interpreting inhibition activity against mold in different samples and makes a quantitative comparison possible. Among the virtues of the method are: 1) the ability to quantify very small differences in the size of colonies which cannot be easily discriminated by visual inspection, 2) the ability to quantify mold growth on transparent as well as on opaque media (e.g. milk), and 3) no prior assumptions for the shape and multiplicity of colonies. The accuracy and non-destructive characteristic of the method allow dynamic quantification of mold growth which can be very valuable in predictive microbiology and in studies related to biopreservation of food products.