Segmentation and tracking of cytoskeletal filaments using open active contours

Area of Science:

• Cell Biology
• Biophysics
• Image Analysis

Background:

• Cytoskeletal structures, such as actin filaments, are crucial for cellular functions.
• Quantifying their conformation and dynamics requires advanced imaging and analysis techniques.
• Existing methods may lack the precision or interactivity needed for detailed fiber analysis.

Purpose of the Study:

• To develop and validate a novel computational method for quantifying cytoskeletal structures in fluorescence microscopy.
• To create an interactive software tool for segmentation, tracking, and visualization of individual fibers.
• To apply the method to analyze actin dynamics in vitro and in vivo.

Main Methods:

• Utilized open active contours, which are parametric curves that deform to minimize energy functions.
• Developed an external energy function incorporating image-derived forces for contour attraction to filament centers and ridge ends.
• Implemented an internal energy function to model contour bending and stretching properties.

Main Results:

• Validated the open active contour method using simulated semiflexible polymers with known mechanical properties.
• Successfully applied the method to quantify actin filaments in vitro using TIRF microscopy.
• Demonstrated the method's utility in analyzing actin cables within fission yeast using spinning disk confocal microscopy.

Conclusions:

• Open active contours provide a robust and accurate approach for quantifying cytoskeletal elements in 2D and 3D.
• The developed interactive software tool facilitates detailed analysis of fiber conformation and dynamics.
• This method advances the study of cytoskeletal organization and function in various biological contexts.