Related Experiment Video
Updated: Aug 14, 2026
12:56
An Automated Method to Perform The In Vitro Micronucleus Assay using Multispectral Imaging Flow Cytometry
Published on: May 13, 2019
Automatic detection of clustered, fluorescent-stained nuclei by digital image-based cytometry
1Lawrence-Berkeley Laboratory, California 94720.
Cytometry
|September 1, 1994
Summary
New algorithms accurately detect individual nuclei in cell clusters for improved image-based cytometry (IC). This enhances the analysis of molecular distributions in complex tissue specimens.
Area of Science:
- Biomedical imaging
- Cell biology
- Computational pathology
Background:
- Automatic image-based cytometry (IC) quantifies molecular distributions in isolated cells.
- Existing methods struggle to detect clustered nuclei in tissue specimens.
Purpose of the Study:
- Develop novel algorithms for detecting individual nuclei within cell clusters.
- Improve the accuracy of nuclear detection in complex tissue samples.
Main Methods:
- Nuclei segmentation and cluster identification based on size and shape.
- Division of clustered nuclei using gradient-based pathfinding algorithms.
- Selection of optimal nuclear segmentation using morphological analysis and image filtering.
Main Results:
- High detection rates for both isolated (328/333) and clustered (254/271) nuclei.
- Successful application to 2-micron prostate and breast cancer tissue sections.
- Demonstrated improved accuracy in nuclear detection within intact specimens.
Conclusions:
- The developed algorithms significantly enhance the detection of individual nuclei in clustered cell populations.
- This advancement enables more precise quantitative analyses in biomedical imaging and pathology.
- The methods promise more accurate detection and analysis of nuclei in intact tissue specimens.
