CARE algorithm dataset
Description
This repository contains the CARE algorithm source code and the image dataset used in the following work: - Salvi et al., "Automated Segmentation of Fluorescence Microscopy Images for 3D Cell Detection in human-derived Cardiospheres", SciReports 2019 (DOI: 10.1038/s41598-019-43137-2) ABSTRACT The ‘cardiosphere’ is a 3D cluster of cardiac progenitor cells recapitulating a stem cell niche-like microenvironment with a potential for disease and regeneration modelling of the failing human myocardium. In this multicellular 3D context, it is extremely important to decrypt the spatial distribution of cell markers for dissecting the evolution of cellular phenotypes by direct quantification of fluorescent signals in confocal microscopy. In this study, we present a fully automated method, named CARE (‘CARdiosphere Evaluation’), for the segmentation of membranes and cell nuclei in human-derived cardiospheres. The proposed method is tested on twenty 3D-stacks of cardiospheres, for a total of 1160 images. Automatic results are compared with manual annotations and two open-source software designed for fluorescence microscopy. CARE performance was excellent in cardiospheres membrane segmentation and, in cell nuclei detection, the algorithm achieved the same performance as two expert operators. To the best of our knowledge, CARE is the first fully automated algorithm for segmentation inside in vitro 3D cell spheroids, including cardiospheres. The proposed approach will provide, in the future, automated quantitative analysis of markers distribution within the cardiac niche-like environment, enabling predictive associations between cell mechanical stresses and dynamic phenotypic changes.