Efficient globally optimal segmentation of cells in fluorescence microscopy images using level sets and convex energy
Jan-Philip Bergeest1, Karl Rohr
1University of Heidelberg, BIOQUANT, IPMB, and DKFZ Heidelberg, Dept. of Bioinformatics and Functional Genomics, Biomedical Computer Vision Group, Im Neuenheimer Feld 267, 69120 Heidelberg, Germany. jan-philip.bergeest@bioquant.uni-heidelberg.de
Medical Image Analysis
|July 17, 2012
Summary
This study introduces a novel active contour method for accurate cell nuclei segmentation in fluorescence microscopy. The approach guarantees a global solution, overcoming limitations of previous methods for high-throughput biological analysis.
Area of Science:
- Cell Biology
- Biomedical Imaging
- Computational Biology
Background:
- Accurate cell segmentation is crucial for high-throughput biological studies.
- Existing methods often depend on initialization and can get stuck in local minima.
Purpose of the Study:
- To develop a robust and efficient cell nuclei segmentation method for fluorescence microscopy.
- To ensure segmentation results are independent of initialization by finding the global solution.
Main Methods:
- Active contours utilizing level sets and convex energy functionals.
- Convex formulations of three established energy functionals.
- An efficient numerical computation approach.
Main Results:
- The proposed method achieves global solutions for cell nuclei segmentation.
- Segmentation results are independent of initial contour placement.
- Validated on diverse fluorescence microscopy datasets with various cell types.
Conclusions:
- The novel active contour approach provides accurate and initialization-independent cell nuclei segmentation.
- This method enhances quantification of protein expression and cell function analysis in high-throughput applications.
