IMAGO – Image-based cell sorting

Sorting cells according to their microscopic appearance using smart image analysis 

A key technology in modern biomedicine is segregating heterogeneous cell populations. The IMAGO research project is taking cell sorting to a whole new level: in the future, it will be possible to sort cells according to spatially resolved characteristics. This is about characteristics that describe a cell’s inner appearance, such as the position of cell organelles or the dis- tribution of proteins.

IMAGO uses smart image analysis methods as well as neural networks that learn the cell characteristics upon which sorting decisions will be made. If successful, this could overcome the previous limitations of conventional cell sorting methods and open new doors in biomedical research – for example in the field of immunotherapy.

At Fraunhofer IIS, the working group on Medical Image Analysis is responsible for developing the AI that makes the sorting decision. This group is develo-ping a user-adaptable AI that will work for many sorting use cases as well as an application software for future users. As part of Fraunhofer’s PREPARE program, IMAGO is to achieve market readiness in collaboration with the Bioanalytics and Bioprocesses (BB) branch of the Fraunhofer Institute for Cell Therapy and Immunology IZI, the Fraunhofer Institute for Applied Optics and Precision Engineering IOF and the Charité university hospital in Berlin. 

Schematische Darstellung der bildbasierten Aussortierung von Zellpaaren
© Fraunhofer IIS, Erik Rümpel
Schematische Darstellung der bildbasierten Aussortierung von Zellpaaren