Segmenting and tracking moving cells in time-lapse sequences is a challenging task, required for many applications in both scientific and industrial settings. Properly characterizing how cells change their shapes and move as they interact with their surrounding environment is key to understanding the mechanobiology of cell migration and its multiple implications in both normal tissue development and many diseases.
In this challenge, we objectively compare and evaluate state-of-the-art whole-cell and nucleus segmentation and tracking methods using both real and computer-generated (2D and 3D) time-lapse microscopy videos of cells and nuclei. With over a decade-long history and three detailed analyses of its results published in Bioinformatics 2014, Nature Methods 2017, and Nature Methods 2023, the Cell Tracking Challenge has become a reference in cell segmentation and tracking algorithm development.
This ongoing benchmarking initiative calls for segmentation-and-tracking and segmentation-only submissions to the Cell Tracking Benchmark and the Cell Segmentation Benchmark, respectively, with evaluation running on a monthly basis. These benchmarks share a diverse repository of annotated datasets that are freely to download. To submit new results for evaluation within any of them, you need to register first.
The repository of annotated datasets along with the gathered pool of segmentation-and-tracking and segmentation-only methods constitutes a unique resource for further studies and analyses. In addition to our own research, we are open to common collaborative projects led by other groups with the involvement of selected challenge organizers who can perform evaluations using the non-public reference annotations of the test datasets. In case of interest, please consult your intention with one of the Steering Committee members who will discuss the feasibility of your proposal and the availability of our human and/or technical resources with the other Steering Committee members.