Abstract
With the ever-increasing quality and quantity of imaging data in biomedical research comes the demand for computational methodologies that enable efficient and reliable automated extraction of the quantitative information contained within these images. One of the challenges in providing such methodology is the need for tailoring algorithms to the specifics of the data, limiting their areas of application. Here we present a broadly applicable approach to quantification and classification of complex shapes and patterns in biological or other multi-component formations. This approach integrates the mapping of all shape boundaries within an image onto a global information-rich graph and machine learning on the multidimensional measures of the graph. We demonstrated the power of this method by (1) extracting subtle structural differences from visually indistinguishable images in our phenotype rescue experiments using the endothelial tube formations assay, (2) training the algorithm to identify biophysical parameters underlying the formation of different multicellular networks in our simulation model of collective cell behavior, and (3) analyzing the response of U2OS cell cultures to a broad array of small molecule perturbations.
Highlights
Quantitative characterization of cell shapes and their organization within multicellular formations is critically important for many biomedical applications, including tissue engineering [1], phenotypic cell-based screening [2,3], and testing platforms for drug discovery [4,5]
A number of software tools have been developed for the analysis of morphological changes among individual cells, such as CellProfiler for quantification of cell phenotypes [6], CellC for counting cell numbers and quantifying cell characteristics [7], and CellGeo for identifying, tracking, and characterizing cell protrusions [8]
Our shape-to-graph mapping is a generalization of the Voronoi Diagram to accept the edges outlining a shape as inputs
Summary
Quantitative characterization of cell shapes and their organization within multicellular formations is critically important for many biomedical applications, including tissue engineering [1], phenotypic cell-based screening [2,3], and testing platforms for drug discovery [4,5]. We present an approach that allows for an efficient and precise extraction and classification of structural features in arbitrarily complex cellular patterns, including subtle variations that are difficult to decipher using visual inspection or a set of standard geometric measures. The graph-based method that we presented here goes beyond the characterization of individual objects and is applicable to a set of images that include compact objects, branching and mesh-like structures, or any combination of such shapes and structures. Textural features that are extracted from original, unsegmented images can potentially compensate for some possible pre-processing artifacts associated with the segmentation process
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call