화학공학소재연구정보센터
Biotechnology and Bioengineering, Vol.97, No.6, 1489-1500, 2007
An in situ probe for on-line monitoring of cell density and viability on the basis of dark field microscopy in conjunction with image processing and supervised machine learning
Fermentation industries would benefit from on-line monitoring of important parameters describing cell growth such as cell density and variability during fermentation processes. For this purpose, an in situ probe has been developed, which utilizes a dark field illumination unit to obtain high contrast images with an integrated CCD camera. To test the probe, brewer's yeast Saccharomyces cerevisiae is chosen as the target microorganism. Images of the yeast cells in the bioreactors are captured, processed, and analyzed automatically by means of mechatronics, image processing, and machine learning. Two support vector machine based classifiers are used for separating cells from background, and for distinguishing live from dead cells afterwards. The evaluation of the in situ experiments showed strong correlation between results obtained by the probe and those by widely accepted standard methods. Thus, the in situ probe has been proved to be a feasible device for on-line monitoring of both cell density and viability with accuracy and stability.