We developed a simple and rapid method based on the combination of image analysis and k-means clustering to selectively enumerate cocci and bacilli from among lactic acid bacteria (LAB) in commercially prepared yogurt. We used our previously reported method for recovering only LAB without non-microbial substances from commercial yogurt, and found that the shape and light intensity of LAB cell images taken by optical microscopy were factors that could distinguish cocci and bacilli, allowing the selective enumeration of LAB. Also, k-means clustering was executed on a dataset of the mean light intensity and aspect ratio of each LAB obtained by image analysis, and each LAB in the image could be automatically assigned to either the cocci or bacilli group. The results obtained by this automated method were in good agreement with those obtained by manually counting the LAB under a microscope, with an overall error within 10%. In addition, this method could provide results within a few hours, which is approximately 1/32 of the time required for the conventional colony-counting method.
Keywords: K-means clustering; Lactic acid bacteria; Microscopy; Selective enumeration; Yogurt.
© 2022. The Author(s), under exclusive licence to The Japan Society for Analytical Chemistry.