Automated reconstruction of whole-embryo cell lineages by learning from sparse annotations
Caroline Malin-Mayor, P. B. Hirsch, Léo Guignard, Katie McDole, Yinan Wan, William C. Lemon, Dagmar Kainmueller, Philipp Keller, Stephan Preibisch, Jan Funke
Abstract
We present a method to automatically identify and track nuclei in time-lapse microscopy recordings of entire developing embryos. The method combines deep learning and global optimization. On a mouse dataset, it reconstructs 75.8% of cell lineages spanning 1 h, as compared to 31.8% for the competing method. Our approach improves understanding of where and when cell fate decisions are made in developing embryos, tissues, and organs.
Topics & Concepts
EmbryoArtificial intelligenceComputer scienceDeep learningBiologyComputational biologyMachine learningCell biologyCell Image Analysis TechniquesMolecular Biology Techniques and ApplicationsSingle-cell and spatial transcriptomics