Automated and semi-automated cell tracking: addressing portability challenges

Area of Science:

• Cell biology
• Computational biology
• Bioimage analysis

Background:

• Cell tracking is crucial for high-throughput biological studies.
• Existing trackers struggle with variable cell density and dynamics, limiting their portability.
• Challenges include noise in cell segmentation, cell divisions, and cell deaths.

Purpose of the Study:

• To develop a portable cell tracking algorithm addressing current limitations.
• To improve the robustness and generalizability of cell tracking across different experimental videos.
• To introduce a parameter-free variation and a semi-automated framework for enhanced tracking.

Main Methods:

• Developed a novel cell tracking algorithm with a new method for recovering cell displacement distributions.
• Proposed a mathematically justified procedure for determining optimal gating distances.
• Introduced a parameter-free variation of the tracker.
• Created a semi-automated framework for challenging tracking conditions.

Main Results:

• The algorithm correctly recovers an average of 96% of cell movements in real videos.
• Outperforms an advanced probabilistic tracker in high cell detection quality scenarios.
• Demonstrated scalability up to 200 cells per frame in synthetic videos.
• The semi-automated framework improved correctly recovered tracks by 12% with minimal manual input.

Conclusions:

• The developed cell tracking algorithm offers enhanced portability and robustness.
• The parameter-free variation and semi-automated framework provide flexible solutions for diverse cell tracking challenges.
• This work advances quantitative analysis in cell biology by improving tracking accuracy and efficiency.