OSCAR: a framework to identify and quantify cells in densely packed three-dimensional biological samples

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.

Quantitative Approaches to Study Retinal Neurogenesis

“…Finally, a digital representation is generated based on the geometry of the objects detected. Our results show that OSCAR is able to outperform other tools used commonly for image analysis in conditions of low resolution and low signal-to-background, typical of biological 3D images registered in toto or in vivo [19].…”

“…The field of automated image processing and analysis of biological and biomedical data is very active, and many recent publications present updated tools and/or propose new methods based on state-of-the-art algorithms [19,40,41] designed to overcome the typical limitations of images from biological samples.…”

“…In this direction, we have recently developed OSCAR: an object segmentation counter and analysis resource that is designed to work with three-dimensional images where the resolution is low and/or the object density is high [19]. Our tool combines nonlinear fitting algorithms with statistical analysis to bypass segmentation errors that frequently take place when segmenting automatically low-resolution images.…”

“…Earlier studies approached the developing vertebrate retina with a focus on key aspects at the genetic and/or molecular levels, providing a highly important but mostly qualitative characterization of the processes, signals, and regulation involved in the retinal specification [18]. This basic qualitative knowledge provided the foundation for more recent contributions that studied retinogenesis from a quantitative spatial and dynamics perspective, taking advantage of more powerful and specialized tools, both at the software and hardware levels [19].…”

Cell competition promotes metastatic intestinal cancer through a multistage process

“…Our routine has been thoroughly tested and optimized for the phenotypical characterization of 3D nuclei in multicellular samples of embryos and hearth microtissues. 46,60 The spatial dimensions of each 3D nucleus were used to quantify the volume of positive signal within each nucleus (referred to as marker volume) and calculation of the average intensity. For classification of cells (e.g.…”