An objective comparison of cell-tracking algorithms

• Published in: Nature methods Vol. 14; no. 12; pp. 1141 - 1152
• Main Authors: Ulman, Vladimír, Maška, Martin, Magnusson, Klas E G, Ronneberger, Olaf, Haubold, Carsten, Harder, Nathalie, Matula, Pavel, Matula, Petr, Svoboda, David, Radojevic, Miroslav, Smal, Ihor, Rohr, Karl, Jaldén, Joakim, Blau, Helen M, Dzyubachyk, Oleh, Lelieveldt, Boudewijn, Xiao, Pengdong, Li, Yuexiang, Cho, Siu-Yeung, Dufour, Alexandre C, Olivo-Marin, Jean-Christophe, Reyes-Aldasoro, Constantino C, Solis-Lemus, Jose A, Bensch, Robert, Brox, Thomas, Stegmaier, Johannes, Mikut, Ralf, Wolf, Steffen, Hamprecht, Fred A, Esteves, Tiago, Quelhas, Pedro, Demirel, Ömer, Malmström, Lars, Jug, Florian, Tomancak, Pavel, Meijering, Erik, Muñoz-Barrutia, Arrate, Kozubek, Michal, Ortiz-de-Solorzano, Carlos
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.12.2017; Nature Publishing Group
• Subjects: 14/19; 14/63; 631/114/1564; 631/80/84; Algorithms; analysis; Benchmarking; Bioinformatics; Biological Microscopy; Biological Techniques; Biomedical Engineering/Biotechnology; Cell Line; Cell Tracking - methods; Computational biology; Humans; Image Interpretation, Computer-Assisted; Innovations; Life Sciences; Mathematical research; Proteomics; Segmentation; Tracking
• ISSN: 1548-7091; 1548-7105; 1548-7105
• DOI: 10.1038/nmeth.4473

This analysis describes the results of three Cell Tracking Challenge editions for examining the performance of cell segmentation and tracking algorithms and provides practical feedback for users and developers. We present a combined report on the results of three editions of the Cell Tracking Challenge, an ongoing initiative aimed at promoting the development and objective evaluation of cell segmentation and tracking algorithms. With 21 participating algorithms and a data repository consisting of 13 data sets from various microscopy modalities, the challenge displays today's state-of-the-art methodology in the field. We analyzed the challenge results using performance measures for segmentation and tracking that rank all participating methods. We also analyzed the performance of all of the algorithms in terms of biological measures and practical usability. Although some methods scored high in all technical aspects, none obtained fully correct solutions. We found that methods that either take prior information into account using learning strategies or analyze cells in a global spatiotemporal video context performed better than other methods under the segmentation and tracking scenarios included in the challenge.