Related Experiment Video
Updated: Mar 21, 2026
08:59
Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps
Published on: October 28, 2018
7.7K
Classification of Normal and Apoptotic Cells from Fluorescence Microscopy Images Using Generalized Polynomial Chaos
Yuncheng Du1, Hector M Budman1, Thomas A Duever1
1Department of Chemical Engineering,University of Waterloo,200 University Ave W,Waterloo,ON,Canada,N2L 3G1.
Summary
This study presents a new algorithm for distinguishing normal from apoptotic Chinese hamster ovary (CHO) cells using fluorescence microscopy. The method enhances cell image segmentation for more accurate apoptosis detection.
Area of Science:
- Cell biology
- Biomedical imaging
- Computational biology
Background:
- Automated quantitative analysis of cell images aids in evaluating experimental outcomes and cell culture.
- Distinguishing between normal and apoptotic cells is crucial for biological research.
Purpose of the Study:
- To develop a fast algorithm for differentiating normal and apoptotic viable Chinese hamster ovary (CHO) cells.
- To improve the accuracy of automated cell analysis using fluorescence microscopy.
Main Methods:
- Developed a stochastic segmentation algorithm combining generalized polynomial chaos expansion and level set functions.
- Extracted morphological features (boundary curvature and length) from segmented cell images.
- Utilized a support vector machine (SVM) classifier trained on these features to differentiate cell states.
Main Results:
- The stochastic level set segmentation provided a probabilistic description of cell boundaries.
- Morphological features derived from this method were used to train an SVM classifier.
- The developed approach demonstrated higher differentiation accuracy compared to the deterministic level set method.
Conclusions:
- The novel stochastic segmentation algorithm effectively captures morphological changes associated with apoptosis.
- Combining stochastic segmentation with SVM classification offers an efficient and accurate method for analyzing CHO cell viability.
- This technique enhances automated quantitative analysis in cell biology research.
