Scientific publications

An objective comparison of cell-tracking algorithms

Dec 1, 2017 | Magazine: Nature Methods

Vladimír Ulman  1 , Martin Maška  1 , Klas E G Magnusson  2 , Olaf Ronneberger  3 , Carsten Haubold  4 , Nathalie Harder  5 , Pavel Matula  1 , Petr Matula  1 , David Svoboda  1 , Miroslav Radojevic  6 , Ihor Smal  6 , Karl Rohr  5 , Joakim Jaldén  2 , Helen M Blau  7 , Oleh Dzyubachyk  8 , Boudewijn Lelieveldt  8   9 , Pengdong Xiao  10 , Yuexiang Li  11 , Siu-Yeung Cho  12 , Alexandre C Dufour  13 , Jean-Christophe Olivo-Marin  13 , Constantino C Reyes-Aldasoro  14 , Jose A Solis-Lemus  14 , Robert Bensch  3 , Thomas Brox  3 , Johannes Stegmaier  15 , Ralf Mikut  15 , Steffen Wolf  4 , Fred A Hamprecht  4 , Tiago Esteves  16   17 , Pedro Quelhas  16 , Ömer Demirel  18 , Lars Malmström  18 , Florian Jug  19 , Pavel Tomancak  19 , Erik Meijering  6 , Arrate Muñoz-Barrutia  20   21 , Michal Kozubek  1 , Carlos Ortiz-de-Solorzano  22   23


Abstract

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.

CITA DEL ARTÍCULO  Nat Methods. 2017 Dec;14(12):1141-1152.  doi: 10.1038/nmeth.4473.  Epub 2017 Oct 30