Deep-learning-based multi-class segmentation for automated, non-invasive routine assessment of human pluripotent stem cell culture status

• Published in: Computers in biology and medicine Vol. 129; p. 104172
• Main Authors: Piotrowski, Tobias, Rippel, Oliver, Elanzew, Andreas, Nießing, Bastian, Stucken, Sebastian, Jung, Sven, König, Niels, Haupt, Simone, Stappert, Laura, Brüstle, Oliver, Schmitt, Robert, Jonas, Stephan
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.02.2021; Elsevier Limited
• Subjects: Algorithms; Automated cell culture; Automatic control; Automation; Cell analysis; Cell culture; Cell differentiation; Center of gravity; Clinical decision making; Colonies; Cultivation; Decision making; Deep learning; Drug screening; Human induced pluripotent stemcells(hiPSC); Human tissues; Image analysis; Image processing; Image segmentation; Inclusions; Internal Medicine; Mathematical models; Microscopy; Morphology; Multi class segmentation; Other; Parameters; Phase contrast; Pluripotency; Precision medicine; Roundness; Routine parameter calculation; Stem cells; Multi class segmentation; Deep-learning; Microscopy; Human induced pluripotent stemcells(hiPSC); Cell analysis; Routine parameter calculation; Automated cell culture
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2020.104172

Human induced pluripotent stem cells (hiPSCs) are capable of differentiating into a variety of human tissue cells. They offer new opportunities for personalized medicine and drug screening. This requires large quantities of high quality hiPSCs, obtainable only via automated cultivation. One of the major requirements of an automated cultivation is a regular, non-invasive analysis of the cell condition, e.g. by whole-well microscopy. However, despite the urgency of this requirement, there are currently no automatic, image-processing-based solutions for multi-class routine quantification of this nature. This paper describes a method to fully automate the cell state recognition based on phase contrast microscopy and deep-learning. This approach can be used for in process control during an automated hiPSC cultivation. The U-Net based algorithm is capable of segmenting important parameters of hiPSC colony formation and can discriminate between the classes hiPSC colony, single cells, differentiated cells and dead cells. The model achieves more accurate results for the classes hiPSC colonies, differentiated cells, single hiPSCs and dead cells than visual estimation by a skilled expert. Furthermore, parameters for each hiPSC colony are derived directly from the classification result such as roundness, size, center of gravity and inclusions of other cells. These parameters provide localized information about the cell state and enable well based treatment of the cell culture in automated processes. Thus, the model can be exploited for routine, non-invasive image analysis during an automated hiPSC cultivation. This facilitates the generation of high quality hiPSC derived products for biomedical purposes. [Display omitted] •Deep-learning-based cell type recognition for hiPSC.•Routine parameter calculation for adherent cell colonies.•Compared to experts, deep-learning has a higher performance in hiPSC routine analysis than visual inspection.•Enabling image analysis based multi parameter process control for hiPSC.