OSCAR: a framework to identify and quantify cells in densely packed three-dimensional biological samples
ledesma-terron, M.; perez-dones, D.; Miguez, d. G.
Show abstract
We have developed an Object Segmentation, Counter and Analysis Resource (OSCAR) that is designed specifically to quantify densely packed biological samples with reduced signal-to-background ratio. OSCAR uses as input three dimensional images reconstructed from confocal 2D sections stained with dies such as nuclear marker and immunofluorescence labeling against specific antibodies to distinguish the cell types of interest. Taking advantage of a combination of arithmetic, geometric and statistical algorithms, OSCAR is able to reconstruct the objects in the 3D space bypassing segmentation errors due to the typical reduced signal to noise ration of biological tissues imaged in toto. When applied to the zebrafish developing retina, OSCAR is able to locate and identify the fate of each nuclei as a cycling progenitor or a terminally differentiated cell, providing a quantitative characterization of the dynamics of the developing vertebrate retina in space and time with unprecedented accuracy.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- WaveletSEG: Automatic wavelet-based 3D nuclei segmentation and analysis for multicellular embryo quantification 96%
- Machine learning-based estimation of spatial gene expression pattern during ESC-derived retinal organoid development 95%
- Annotation-free Learning of Plankton for Classification and Anomaly Detection 94%
Similar papers in this journal
Similar papers in this journal
- A deep learning approach for staging embryonic tissue isolates with small data 96%
- ConfluentFUCCI for fully-automated analysis of cell-cycle progression in a highly dense collective of migrating cells 95%
- Caveolae and scaffold detection from single molecule localization microscopy data using deep learning 95%
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.