Improving segmentation of 3D touching cell nuclei using flow tracking on surface meshes.

Area of Science:

• Bioimage informatics
• Computational biology
• 3D microscopy imaging

Background:

• Accurate segmentation of touching cell nuclei in 3D microscopy images is crucial for quantitative analysis in biological research.
• Existing methods may struggle with the precise separation of adjacent cell nuclei, impacting downstream analyses.
• Developing robust segmentation techniques is essential for advancing fields like cancer research and developmental biology.

Purpose of the Study:

• To present a novel computational method for improving the segmentation of touching cell nuclei in 3D microscopy images.
• To enhance the accuracy and reliability of cell nuclei segmentation beyond existing approaches.
• To provide a validated tool for researchers in bioimage informatics and computational biology.

Main Methods:

• The proposed method utilizes surface mesh reconstruction and curvature estimation from initial binary nuclei.
• Direction field diffusion and flow tracking are performed on these surface meshes to delineate individual nuclei.
• Segmentation results from the surface meshes are then projected back to the volumetric images for final output.

Main Results:

• The method successfully segmented touching cell nuclei in both synthesized and real 3D microscopy datasets.
• Validation demonstrated the effectiveness of the surface mesh-based approach in separating complex nuclear structures.
• The technique showed significant improvements in segmentation accuracy compared to baseline methods.

Conclusions:

• The novel surface mesh-based method offers a significant advancement in 3D touching cell nuclei segmentation.
• This technique provides a robust solution for challenging segmentation tasks in bioimage analysis.
• The validated approach holds promise for improving the accuracy of quantitative biological studies reliant on cell nuclei segmentation.