ACDC: Automated Cell Detection and Counting for Time-Lapse Fluorescence Microscopy.

Area of Science:

• Cell biology
• Bioimaging
• Computational microscopy

Background:

• Time-lapse microscopy is crucial for observing live-cell dynamics.
• Current methods face challenges in training and time efficiency, limiting lab application.
• Accurate cell nuclei detection and counting are essential for biological research.

Purpose of the Study:

• To introduce Automated Cell Detection and Counting (ACDC), a novel method for fluorescent cell nuclei activity detection in time-lapse microscopy.
• To address limitations of existing methods concerning training and time constraints.
• To provide a practical and efficient solution for laboratory-based cell imaging analysis.

Main Methods:

• ACDC employs bilateral filtering for image smoothing while preserving edges.
• The method utilizes watershed transform and morphological filtering for nuclei segmentation.
• A Parent-Workers implementation leverages multi-core architectures for efficient processing of large datasets.

Main Results:

• ACDC achieved accurate cell counting and nuclei segmentation on diverse datasets.
• The method demonstrated high accuracy without requiring large annotated datasets.
• Performance was validated by Dice Similarity Coefficients (76.84, 88.64) and Pearson coefficients (0.99, 0.96).

Conclusions:

• ACDC provides an effective and accurate solution for cell nuclei detection and counting in time-lapse microscopy.
• The method's efficiency and scalability make it suitable for large-scale imaging datasets.
• ACDC overcomes limitations of prior techniques, offering a practical tool for biological research.