2020/10/22 by Dennis Bähr, Bähr, Dennis, Dennis Eschweiler +9
Biochemistry, Genetics and Molecular Biology · Computer Science · Engineering · #AI in cancer detection #Cell Behavior (q-bio.CB) #Cell Image Analysis Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Biological sciences #FOS: Computer and information sciences #FOS: Electrical engineering #Image Processing Techniques and Applications #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2010.12011
openalex publication_date 2020/10/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Automatic analysis of spatio-temporal microscopy images is inevitable for\nstate-of-the-art research in the life sciences. Recent developments in deep\nlearning provide powerful tools for automatic analyses of such image data, but\nheavily depend on the amount and quality of provided training data to perform\nwell. To this end, we developed a new method for realistic generation of\nsynthetic 2D+t microscopy image data of fluorescently labeled cellular nuclei.\nThe method combines spatiotemporal statistical shape models of different cell\ncycle stages with a conditional GAN to generate time series of cell populations\nand provides instance-level control of cell cycle stage and the fluorescence\nintensity of generated cells. We show the effect of the GAN conditioning and\ncreate a set of synthetic images that can be readily used for training and\nbenchmarking of cell segmentation and tracking approaches.\n