Segmentation, tracking and lineage analysis of yeast cellsin bright field microscopy images

Time lapse microscopy images are an important support in quantitative biology. Gene circuit dynamics can be precisely estimated at a single-cell level but automatic cell segmentation and tracking are required due to the large number of cells under study and the large amount of images to be analyzed. Here we present a solution for segmentation, tracking and lineage analysis of yeast cells in bright field, phase contrast microscopy images. The solution is designed to be applied with little effort by biologists thanks to the robust global linking segmentation approach and a pattern recognition-based false-positive detection system. Performance evaluation methods are also introduced and used for a reliable evaluation of our method. Moreover, we show here that our method achieves competitive performances with existing methods without time-consuming optimal parameters search.