3DCellComposer - A Versatile Pipeline Utilizing 2D Cell Segmentation Methods for 3D Cell Segmentation.

Area of Science:

• Bioimage Informatics
• Computational Biology
• Cellular Imaging

Background:

• Accurate cell segmentation is vital for cellular analysis in bioimage informatics.
• 3D cell segmentation is less developed than 2D methods due to data scarcity and complexity.
• Existing 3D methods often lack generalizability across different imaging modalities.

Purpose of the Study:

• To develop a generalizable approach for accurate 3D cell segmentation using 2D methods.
• To introduce 3DCellComposer, an open-source package for versatile 3D cell segmentation.
• To enhance quality evaluation for 3D segmentation without human annotation.

Main Methods:

• Implemented a novel approach within the 3DCellComposer package.
• Enabled users to integrate existing 2D cell segmentation models.
• Enhanced CellSegmentationEvaluator for 3D image quality assessment.

Main Results:

• Demonstrated high-quality 3D segmentation of tissue images using the developed approach.
• Showcased superior performance compared to existing 3D methods on cell culture images.
• Validated the generalizability of the approach across different imaging modalities.

Conclusions:

• 3DCellComposer offers a valuable alternative to manual 3D annotation and model training.
• The tool facilitates large-scale bioimaging projects like the Human BioMolecular Atlas Program.
• It reduces the need for extensive data acquisition and training for new sample types.