Network flow integer programming to track elliptical cells in time-lapse sequences

• Verfasst von: Türetken, Engin [VerfasserIn]
• Verfasst von: Haubold, Carsten [VerfasserIn]
• Titel: Network flow integer programming to track elliptical cells in time-lapse sequences
• Verf.angabe: Engin Türetken, Xinchao Wang, Carlos J. Becker, Carsten Haubold, and Pascal Fua
• Jahr: 2017
• Jahr des Originals: 2016
• Umfang: 10 S.
• Teil: volume:36
• Teil: year:2017
• Teil: number:4
• Teil: pages:942-951
• Teil: extent:10
• Fussnoten: Article was published on 15 December 2016 ; Gesehen am 17.09.2018
• Titel Quelle: Enthalten in: Institute of Electrical and Electronics EngineersIEEE transactions on medical imaging
• Ort Quelle: New York, NY : Institute of Electrical and Electronics Engineers, 1982
• Jahr Quelle: 2017
• Band/Heft Quelle: 36(2017), 4, Seite 942-951
• ISSN Quelle: 1558-254X
• DOI: doi:10.1109/TMI.2016.2640859
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: Algorithms
• Sach-SW: Cell tracking
• Sach-SW: cellular biophysics
• Sach-SW: clumped cells
• Sach-SW: Computer Simulation
• Sach-SW: elliptical cells
• Sach-SW: image segmentation
• Sach-SW: Image segmentation
• Sach-SW: image sequences
• Sach-SW: Image sequences
• Sach-SW: integer programming
• Sach-SW: Linear programming
• Sach-SW: medical image processing
• Sach-SW: network flow integer programming
• Sach-SW: network flows
• Sach-SW: Sociology
• Sach-SW: Statistics
• Sach-SW: Target tracking
• Sach-SW: Time Factors
• Sach-SW: time-lapse image sequences
• Sach-SW: Trajectory
• K10plus-PPN: 1581033117

Abstract

We propose a novel approach to automatically tracking elliptical cell populations in time-lapse image sequences. Given an initial segmentation, we account for partial occlusions and overlaps by generating an over-complete set of competing detection hypotheses. To this end, we fit ellipses to portions of the initial regions and build a hierarchy of ellipses, which are then treated as cell candidates. We then select temporally consistent ones by solving to optimality an integer program with only one type of flow variables. This eliminates the need for heuristics to handle missed detections due to partial occlusions and complex morphology. We demonstrate the effectiveness of our approach on a range of challenging sequences consisting of clumped cells and show that it outperforms state-of-the-art techniques.