Automatic Cell Segmentation in Fluorescence Images of Confluent Cell Monolayers Using Multi-object Geometric...

Area of Science:

• Cell Biology
• Biomedical Imaging
• Computational Biology

Background:

• Accurate cell morphology analysis requires robust image segmentation.
• Existing automatic cell segmentation methods face challenges in accuracy, efficiency, and adaptability.
• Confluent cell monolayers present unique segmentation difficulties due to cell-cell interactions and potential image defects.

Purpose of the Study:

• To develop a fully automatic and accurate method for cell segmentation in fluorescence microscopy images of confluent cell monolayers.
• To improve the efficiency and adaptability of cell segmentation software for quantitative cell morphology studies.
• To address common defects in fluorescence images that hinder accurate segmentation.

Main Methods:

• A fully automatic segmentation process initiated by detecting cell nuclei as seeds.
• Utilizing a multi-object geometric deformable model (MGDM) for precise final cell segmentation.
• Employing order-statistic filters and principal curvature analysis to enhance cell junctions and mitigate image defects.

Main Results:

• The proposed method achieves fully automatic segmentation of cells in confluent monolayers.
• Enhanced cell junction detection and robust segmentation using MGDM.
• Achieved an average Dice coefficient of 0.88 when compared to manual cell delineations, indicating high accuracy.

Conclusions:

• The developed method offers a significant advancement in automatic cell segmentation for fluorescence microscopy.
• The approach provides accurate, efficient, and adaptable cell segmentation, crucial for quantitative cell morphology analysis.
• The MGDM-based segmentation ensures seamless separation of neighboring cells without overlaps or gaps.