Related Experiment Video
Updated: May 14, 2026
06:25
Using Computer Vision Libraries to Streamline Nuclei Quantification
Published on: June 6, 2025
A novel dictionary based computer vision method for the detection of cell nuclei.
Jonas De Vylder1, Jan Aelterman, Trees Lepez
1Department of Telecommunications and Information Processing, iMinds, Ghent University, Ghent, Belgium. jonas.devylder@telin.ugent.be
Plos One
|January 30, 2013
Summary
This study introduces a new active contour method for accurate cell nuclei detection in microscopic images. The approach improves detection in cluttered images, achieving high accuracy for cell nuclei identification.
Area of Science:
- Biomedical Imaging
- Computational Biology
- Cell Biology
Background:
- Accurate cell nuclei detection is crucial for biological research but challenging due to image artifacts like blur and occlusion.
- Automated methods are needed to overcome the time-consuming nature of manual nuclei identification.
Purpose of the Study:
- To develop a novel and robust method for cell nuclei detection in fluorescent microscopic images.
- To improve upon existing automated nuclei detection techniques by incorporating prior knowledge of nucleus shape.
Main Methods:
- The proposed method utilizes an active contour framework enhanced with prior knowledge of nucleus shape.
- This prior knowledge is defined via a dictionary-based approach and formulated as a convex energy function optimization.
Main Results:
- The method demonstrates accurate detection of individual nuclei, even in dense clusters.
- Achieved an F-measure of 0.96 for cell nuclei detection in peripheral blood mononuclear cells, outperforming state-of-the-art methods (F-measure of 0.90).
Conclusions:
- The novel active contour method offers a robust and accurate solution for cell nuclei detection.
- Exploiting prior nucleus shape knowledge significantly enhances detection performance in challenging microscopic images.
