Segmentation and tracking of cytoskeletal filaments using open active contours

Authors

  • Matthew B. Smith,

    1. Department of Physics, Lehigh University, Bethlehem, Pennsylvania
    Search for more papers by this author
    • Matthew B. Smith, Hongsheng Li, and Tian Shen contributed equally to this work.

  • Hongsheng Li,

    1. Department of Computer Science and Engineering, Lehigh University, Bethlehem, Pennsylvania
    Search for more papers by this author
    • Matthew B. Smith, Hongsheng Li, and Tian Shen contributed equally to this work.

  • Tian Shen,

    1. Department of Computer Science and Engineering, Lehigh University, Bethlehem, Pennsylvania
    Search for more papers by this author
    • Matthew B. Smith, Hongsheng Li, and Tian Shen contributed equally to this work.

  • Xiaolei Huang,

    Corresponding author
    1. Department of Computer Science and Engineering, Lehigh University, Bethlehem, Pennsylvania
    • Department of Computer Science and Engineering, Lehigh University, 19 Memorial Drive West, Bethlehem, PA 18015, USA
    Search for more papers by this author
  • Eddy Yusuf,

    1. Department of Physics, Lehigh University, Bethlehem, Pennsylvania
    Current affiliation:
    1. Physics Department, Surya Institute School of Education, Plaza Summarecon Serpong, Serpong, Tangerang 15810, Indonesia
    Search for more papers by this author
  • Dimitrios Vavylonis

    Corresponding author
    1. Department of Physics, Lehigh University, Bethlehem, Pennsylvania
    • Department of Physics, Lehigh University, 16 Memorial Drive East, Bethlehem, PA 18015, USA
    Search for more papers by this author

  • Monitoring Editor: Bruce Goode

Abstract

We use open active contours to quantify cytoskeletal structures imaged by fluorescence microscopy in two and three dimensions. We developed an interactive software tool for segmentation, tracking, and visualization of individual fibers. Open active contours are parametric curves that deform to minimize the sum of an external energy derived from the image and an internal bending and stretching energy. The external energy generates (i) forces that attract the contour toward the central bright line of a filament in the image, and (ii) forces that stretch the active contour toward the ends of bright ridges. Images of simulated semiflexible polymers with known bending and torsional rigidity are analyzed to validate the method. We apply our methods to quantify the conformations and dynamics of actin in two examples: actin filaments imaged by TIRF microscopy in vitro, and actin cables in fission yeast imaged by spinning disk confocal microscopy. © 2010 Wiley-Liss, Inc.

Ancillary