Three-dimensional cell and tissue image analysis for cellular and sub-cellular morphological modeling and classification
The ability to automate the processes of specimen collection, image acquisition, data pre-processing, computation of derived biomarkers, modeling, classification and analysis can significantly impact clinical decision-making and fundamental investigation of cell deformation. This disclosure combine 3D cell nuclear shape modeling by robust smooth surface reconstruction and extraction of shape morphometry measure into a highly parallel pipeline workflow protocol for end-to-end morphological analysis of thousands of nuclei and nucleoli in 3D. This approach allows efficient and informative evaluation of cell shapes in the imaging data and represents a reproducible technique that can be validated, modified, and repurposed by the biomedical community. This facilitates result reproducibility, collaborative method validation, and broad knowledge dissemination.
1. An automated method for analyzing biological tissues and cells, comprising:
staining constituents of a biological sample, where the biological sample includes tissue or cell culture containing at least one cell;
receiving image data of the biological sample, where the image data provides a three-dimensional representation of the biological sample;
labeling constituents of the at least one cell in the image data;
for each labeled tissue biomarker or cell or sub-cellular organelle or component in the image data, constructing a mathematical representation of boundaries defining a given cell nuclei, including by chromosome territories, topological associated domains (TADs), lamina associated domains (LADs), or telomeric boundaries;
extracting features for each of the labeled tissue biomarker or cell or sub-cellular organelle or component using the mathematical representation of boundaries defining a given cell nuclei, where the features are measures of shape and size of the labeled tissue biomarker or cell or sub-cellular organelle or component;
storing two or more models for cell classification; and
classifying the biological sample by comparing the extracted features of the labeled tissue biomarker or cell or sub-cellular organelle or component to the stored models.
2. The method of claim 1 further comprises generating the image data using a three-dimensional image acquisition technique.
3. The method of claim 1 wherein labeling constituents further comprises segmenting the image data into volumes, which represent the constituents.
4. The method of claim 1 wherein constructing a mathematical representation further comprises using iterative Laplace-Beltrami eigen-projection and boundary deformation.
5. The method of claim 1 wherein the features extracted from the labeled tissue biomarker or cell or sub-cellular organelle or component include determining one or more of a volume of a labeled tissue biomarker or cell or sub-cellular organelle or component, a surface area of a labeled tissue biomarker or cell or sub-cellular organelle or component, mean curvature of a labeled tissue biomarker or cell or sub-cellular organelle or component, shape index of a labeled tissue biomarker or cell or sub-cellular organelle or component, curvedness index of a labeled tissue biomarker or cell or sub-cellular organelle or component, and fractal dimension of a labeled tissue biomarker or cell or sub-cellular organelle or component.
6. The method of claim 1 further comprises classifying the biological sample using a random forest classification method.
7. The method of claim 1 further comprises classifying the biological sample using a classification method selected from a group consisting of linear classifiers, k nearest neighbor method, decision tree methods, neural networks, and support vector machines.
8. The method of claim 1 further comprises classifying the biological sample only if the biological sample contains more than three cells.
9. The method of claim 1 wherein staining constituents of the biological sample using a label for a tissue biomarker or cell or sub-cellular organelle or component, such as fibrillarin antibody, ethidium bromide and 4′,6-diamidino-2-phenylindole.
10. The method of claim 1 wherein the biological sample is from a cell culture or cell tissue.
11. An automated method for analyzing biological cells, comprising:
staining constituents of a biological sample, where the biological sample includes at least one cell;
receiving image data of the biological sample, where the image data provides a three-dimensional representation of the biological sample;
labeling constituents of the at least one cell in the image data;
for each labeled cell nuclei in the image data, constructing a mathematical representation of boundaries defining a given cell nuclei;
extracting features for each of the labeled cell nuclei using the mathematical representation of the cell nuclei, where the features are measures of shape and size of the labeled cell nuclei;
for each labeled cell nucleoli, constructing a mathematical representation of boundaries defining a given cell nucleoli;
extracting features from each of the labeled cell nucleoli using the mathematical representation of the cell nucleoli, where the features are measures of shape and size of the labeled cell nucleoli;
storing two or more models for cell classification; and
classifying the biological sample by comparing the extracted features of the labeled cell nucleus and the labeled cell nucleolus to the stored models.
12. The method of claim 11 further comprises generating the image data using a confocal microscope.
13. The method of claim 11 wherein labeling constituents further comprises segmenting the image data into volumes which represent the constituents.
14. The method of claim 11 wherein constructing a mathematical representation further comprises using iterative Laplace-Beltrami eigen-projection and boundary deformation.
15. The method of claim 11 wherein the features extracted from the labeled cell nucleus include determining one or more of a volume of a labeled cell nuclei, a surface area of a labeled cell nuclei, mean curvature of a labeled cell nuclei, shape index of a labeled cell nuclei, curvedness index of a labeled cell nuclei, and fractal dimension of a labeled cell nuclei.
16. The method of claim 15 wherein the features extracted from the labeled cell nucleolus include determining one or more of a count of nucleoli in corresponding cell nuclei, a volume of a labeled cell nucleoli, surface area of a labeled cell nucleoli, mean curvature of a labeled cell nucleoli, shape index of a labeled cell nucleoli, curvedness index of a labeled cell nucleoli, and fractal dimension of a labeled cell nucleoli.
17. The method of claim 11 further comprises classifying the biological sample using a random forest classification method.
18. The method of claim 11 further comprises classifying the biological sample using a classification method selected from a group consisting of linear classifiers, k nearest neighbor method, decision tree methods, neural networks and support vector machines.
19. The method of claim 11 further comprises classifying the biological sample only if the biological sample contains more than three cells.
20. The method of claim 11 wherein staining constituents of the biological sample using fibrillarin antibody, ethidium bromide and 4′,6-diamidino-2-phenylindole.
21. The method of claim 11 wherein the biological sample is from a cell culture.