Automated tracking of migrating cells in phase-contrast video microscopy sequences using image registration

Area of Science:

• Cell Biology
• Biophysics
• Microscopy Techniques

Background:

• In vitro cell motility analysis is crucial for understanding cellular responses.
• Existing cell-tracking systems often rely on fluorescently labeled images, limiting their application.
• Low-contrast microscopy, such as phase-contrast, presents challenges for accurate cell tracking.

Purpose of the Study:

• To evaluate the performance of five common cell-tracking systems against a novel in-house system.
• To assess the effectiveness of image registration and optical flow for cell motility analysis.
• To compare tracking efficiency using phase-contrast versus fluorescence microscopy.

Main Methods:

• Compared Retrac (manual), CellTrack, ImageJ (MTrack2), Imaris, Volocity, and a novel in-house system.
• Tracked human epithelial cell migration over ten frames using phase-contrast time-lapse microscopy.
• Evaluated performance with both non-fluorescent and fluorescently labeled cells.

Main Results:

• The in-house image-registration system achieved the highest success rate (95%) for tracking non-dividing epithelial cells in low-contrast images.
• Fluorescence microscopy generally improved tracking efficiency across all tested systems compared to phase contrast.
• The in-house system showed a smaller improvement (<5%) with fluorescence, while Volocity and ImageJ demonstrated greater gains.

Conclusions:

• The novel in-house cell-tracking system based on image registration is highly effective for analyzing cell motility in low-contrast microscopy.
• Image registration offers a robust solution for tracking cells in challenging imaging conditions.
• While fluorescence enhances tracking, the developed system provides reliable results even without it.