Automated Segmentation of Fluorescence Microscopy Images for 3D Cell Detection in human-derived Cardiospheres

Area of Science:

• Biomedical Engineering
• Cell Biology
• Regenerative Medicine

Background:

• Cardiospheres mimic cardiac stem cell niches, valuable for studying heart disease and regeneration.
• Understanding spatial marker distribution in cardiospheres is crucial for dissecting cellular phenotype evolution.
• Current methods for analyzing fluorescent signals in 3D cardiospheres are often manual and time-consuming.

Purpose of the Study:

• To develop a fully automated method for segmenting cell membranes and nuclei within human-derived cardiospheres.
• To establish a quantitative tool for analyzing cellular structures in a 3D cardiac niche model.
• To enable advanced research into cardiac disease and regeneration modeling.

Main Methods:

• A novel, fully automated algorithm named CARE (CARdiosphere Evaluation) was developed.
• CARE was applied to segment membranes and cell nuclei in twenty 3D cardiosphere image stacks (1160 images total).
• Performance was evaluated by comparing CARE's results against manual annotations and two open-source microscopy software packages.

Main Results:

• CARE demonstrated excellent performance in segmenting cardiosphere membranes.
• The algorithm achieved performance comparable to expert operators in cell nuclei detection.
• CARE is the first fully automated algorithm for segmentation within 3D cell spheroids, including cardiospheres.

Conclusions:

• The CARE algorithm provides a robust and automated solution for analyzing cellular structures in cardiospheres.
• This method facilitates quantitative analysis of marker distribution within the cardiac niche-like environment.
• Future applications include enabling predictive associations between cellular stresses and phenotypic changes in cardiac research.