(249g) Smart Imaging Analysis for the On Line Observation of Bioprocesses At the Example of the Microalgae Crypteconinium Cohnii

Docosahexaenoic acid (DHA) is a polyunsatured fatty acid and belongs to the group of omega-3-fatty-acids. Due to its beneficial effect on human health DHA is a valuable natural product and is commercially manufactured from the microalgae Crypthecodinium cohnii [1]. The dinoflagellate grows in a certain culture medium heterotrophically to high cell concentrations.Under controlled and monitored conditions the microalgae produces a significant fraction of pure DHA [2]. The organism is very sensitive to various environmental conditions which make high yield production of DHA challenging. Detailed process understanding and exact process control is essential to accurately adjust process parameters to the requirements of the organism.Achieving certain cell densities and starting the DHA production phase, the process is switched from a nutrient-excess state to a nutrient-limiting state. Based on the morphologic and physiologic changes of the cells, the optimal modes how to switch between these two process stages can be defined. The integration of an optical detection method will be beneficial, when cell densities and different morphological states can be displayed and recognized automatically [3].

This study analyzes the influence of flash light and media contamination of an invasive photo-optical microscope through changes of optical density and DHA productivity. Cell concentration is measured automatically with the developed in-situ probe and compared to ex-situ cell counting methods. Additionally feature extraction with focus on changes in cell sizes, cell shapes and cell walls is performed. This information provides the basis for the development of automated feature recognition by a newly developed image analysis software.

 [1] Wynn, J., P. Behrens, A. Sundararajan, J. Hansen and A. Kirk 2010. Production of Single Cell Oils by Dinoflagellates, Single Cell Oils. Z. Cohen and C. Ratledge.

[2] Mendes et al. 2008, Journal of Applied Phycology 21: 199-214

[3] Maaß et al. 2012, Computers and Chemical Engineering, DOI10.1016/j.compchemeng.2012.05.014