An objective comparison of cell-tracking algorithms

被引:359
作者
Ulman, Vladimir [1 ,29 ]
Maska, Martin [1 ]
Magnusson, Klas E. G. [2 ]
Ronneberger, Olaf [3 ,4 ,30 ]
Haubold, Carsten [5 ]
Harder, Nathalie [6 ,7 ,31 ]
Matula, Pavel [1 ]
Matula, Petr [1 ]
Svoboda, David [1 ]
Radojevic, Miroslav [8 ,9 ]
Smal, Ihor [8 ,9 ]
Rohr, Karl [6 ,7 ]
Jalden, Joakim [2 ]
Blau, Helen M. [10 ,11 ]
Dzyubachyk, Oleh [12 ]
Lelieveldt, Boudewijn [12 ,13 ]
Xiao, Pengdong [14 ,32 ]
Li, Yuexiang [15 ,33 ]
Cho, Siu-Yeung [16 ]
Dufour, Alexandre C. [17 ]
Olivo-Marin, Jean-Christophe [17 ]
Reyes-Aldasoro, Constantino C. [18 ]
Solis-Lemus, Jose A. [18 ]
Bensch, Robert [3 ,4 ]
Brox, Thomas [3 ,4 ]
Stegmaier, Johannes [19 ]
Mikut, Ralf [19 ]
Wolf, Steffen [5 ]
Hamprecht, Fred A. [5 ]
Esteves, Tiago [20 ,21 ]
Quelhas, Pedro [20 ]
Demirel, Omer [22 ]
Malmstrom, Lars [22 ]
Jug, Florian [23 ]
Tomancak, Pavel [23 ]
Meijering, Erik [8 ,9 ]
Munoz-Barrutia, Arrate [24 ,25 ]
Kozubek, Michal [1 ]
Ortiz-de-Solorzano, Carlos [26 ,27 ,28 ]
机构
[1] Masaryk Univ, Ctr Biomed Image Anal, Brno, Czech Republic
[2] KTH Royal Inst Technol, ACCESS Linnaeus Ctr, Stockholm, Sweden
[3] Univ Freiburg, Comp Sci Dept, Freiburg, Germany
[4] Univ Freiburg, BIOSS Ctr Biol Signaling Studies, Freiburg, Germany
[5] Heidelberg Univ, IWR, Heidelberg Collaboratory Image Proc, Heidelberg, Germany
[6] Heidelberg Univ, Biomed Comp Vision Grp, Dept Bioinformat & Funct Gen, BIOQUANT,IPMB, Heidelberg, Germany
[7] DKFZ, Heidelberg, Germany
[8] Erasmus MC, Biomed Imaging Grp Rotterdam, Dept Med Informat, Rotterdam, Netherlands
[9] Erasmus MC, Biomed Imaging Grp Rotterdam, Dept Radiol, Rotterdam, Netherlands
[10] Stanford Univ, Sch Med, Baxter Lab Stem Cell Biol, Dept Microbiol & Immunol, Stanford, CA USA
[11] Stanford Univ, Sch Med, Inst Stem Cell Biol & Regenerat Med, Stanford, CA USA
[12] Leiden Univ, Med Ctr, Dept Radiol, Div Image Proc, Leiden, Netherlands
[13] Delft Univ Technol, Intelligent Syst Dept, Delft, Netherlands
[14] A* Star, Inst Mol & Cell Biol, Singapore, Singapore
[15] Univ Nottingham, Dept Engn, Nottingham, England
[16] Univ Nottingham, Fac Engn, Ningbo, Zhejiang, Peoples R China
[17] Inst Pasteur, BioImage Anal Unit, Paris, France
[18] City Univ London, Sch Math Comp Sci & Engn, Res Ctr Biomed Engn, London, England
[19] Karlsruhe Inst Technol, Inst Appl Comp Sci, Grp Automated Image & Data Anal, Eggenstein Leopoldshafen, Germany
[20] Univ Porto, i3S, Oporto, Portugal
[21] Univ Porto, Fac Engn, Oporto, Portugal
[22] Univ Zurich, S3IT, Zurich, Switzerland
[23] Max Planck Inst Mol Cell Biol & Genet, Dresden, Germany
[24] Univ Carlos III Madrid, Bioengn & Aerosp Engn Dept, Getafe, Spain
[25] Inst Invest Sanitaria Gregorio Maranon, Madrid, Spain
[26] Univ Navarra, IDISNA, CIBERONC, Pamplona, Spain
[27] Univ Navarra, Program Solid Tumors & Biomarkers, Ctr Appl Med Res, Pamplona, Spain
[28] Univ Navarra, TECNUN Sch Engn, Bioengn Dept, San Sebastian, Spain
[29] Max Planck Inst Mol Cell Biol & Genet, Dresden, Germany
[30] DeepMind, London, England
[31] Definiens AG, Munich, Germany
[32] Natl Heart Ctr Singapore, Natl Heart Res Inst Singapore, Singapore, Singapore
[33] Shenzhen Univ, Coll Comp Sci & Software Engn, Comp Vision Inst, Shenzhen, Peoples R China
基金
美国国家卫生研究院; 瑞典研究理事会;
关键词
SEGMENTATION; NUCLEI; CLASSIFICATION; MICROSCOPY; IMAGE;
D O I
10.1038/nmeth.4473
中图分类号
Q5 [生物化学];
学科分类号
071010 ; 081704 ;
摘要
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.
引用
收藏
页码:1141 / +
页数:15
相关论文
共 42 条
[1]  
Amat F, 2014, NAT METHODS, V11, P951, DOI [10.1038/NMETH.3036, 10.1038/nmeth.3036]
[2]  
Bensch R., 2015, P 2015 IEEE INT S BI, P1120
[3]  
Bise R, 2011, I S BIOMED IMAGING, P1004, DOI 10.1109/ISBI.2011.5872571
[4]   Flow-Based Cytometric Analysis of Cell Cycle via Simulated Cell Populations [J].
Brown, M. Rowan ;
Summers, Huw D. ;
Rees, Paul ;
Smith, Paul J. ;
Chappell, Sally C. ;
Errington, Rachel J. .
PLOS COMPUTATIONAL BIOLOGY, 2010, 6 (04)
[5]   Microscopic imaging techniques for drug discovery [J].
Bullen, Andrew .
NATURE REVIEWS DRUG DISCOVERY, 2008, 7 (01) :54-67
[6]   Automated segmentation, classification, and tracking of cancer cell nuclei in time-lapse microscopy [J].
Chen, XW ;
Zhou, XB ;
Wong, STC .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2006, 53 (04) :762-766
[7]   Objective comparison of particle tracking methods [J].
Chenouard, Nicolas ;
Smal, Ihor ;
de Chaumont, Fabrice ;
Maska, Martin ;
Sbalzarini, Ivo F. ;
Gong, Yuanhao ;
Cardinale, Janick ;
Carthel, Craig ;
Coraluppi, Stefano ;
Winter, Mark ;
Cohen, Andrew R. ;
Godinez, William J. ;
Rohr, Karl ;
Kalaidzidis, Yannis ;
Liang, Liang ;
Duncan, James ;
Shen, Hongying ;
Xu, Yingke ;
Magnusson, Klas E. G. ;
Jalden, Joakim ;
Blau, Helen M. ;
Paul-Gilloteaux, Perrine ;
Roudot, Philippe ;
Kervrann, Charles ;
Waharte, Francois ;
Tinevez, Jean-Yves ;
Shorte, Spencer L. ;
Willemse, Joost ;
Celler, Katherine ;
van Wezel, Gilles P. ;
Dan, Han-Wei ;
Tsai, Yuh-Show ;
Ortiz de Solorzano, Carlos ;
Olivo-Marin, Jean-Christophe ;
Meijering, Erik .
NATURE METHODS, 2014, 11 (03) :281-U247
[8]  
Cicconet Marcelo, 2013, International Conference on Computer Vision Theory and Applications (VISAPP 2013). Proceedings, P669
[9]   Quantitative 3D analysis of complex single border cell behaviors in coordinated collective cell migration [J].
Cliffe, Adam ;
Doupe, David P. ;
Sung, HsinHo ;
Lim, Isaac Kok Hwee ;
Ong, Kok Haur ;
Cheng, Li ;
Yu, Weimiao .
NATURE COMMUNICATIONS, 2017, 8
[10]   Segmentation of confocal microscope images of cell nuclei in thick tissue sections [J].
de Solórzano, CO ;
Rodriguez, EG ;
Jones, A ;
Pinkel, D ;
Gray, JW ;
Sudar, D ;
Lockett, SJ .
JOURNAL OF MICROSCOPY, 1999, 193 :212-226