Related Experiment Video
Updated: Jun 21, 2026
09:04
Lens-free Video Microscopy for the Dynamic and Quantitative Analysis of Adherent Cell Culture
Published on: February 23, 2018
Tracking cells in Life Cell Imaging videos using topological alignments
Axel Mosig1, Stefan Jäger, Chaofeng Wang
1Department of Combinatorics and Geometry, CAS-MPG Partner Institute for Computational Biology, 200031 Shanghai, PR China. axel@picb.ac.cn
Algorithms for Molecular Biology : AMB
|July 18, 2009
Summary
This study introduces topological alignments for cell tracking, effectively linking segmentations between video frames. The method overcomes common over- and under-segmentation issues in bioimage analysis.
Area of Science:
- Bioimage informatics
- Computational biology
- Cellular imaging analysis
Background:
- Live cell imaging generates large datasets, posing challenges for automated cell tracking.
- Existing cell tracking algorithms struggle with segmentation errors, misidentifying single cells or merging multiple cells.
Purpose of the Study:
- To develop a novel method for accurate cell tracking in live cell imaging.
- To address the persistent problem of over- and under-segmentation in cell tracking algorithms.
Main Methods:
- Proposes topological alignments to link segmentations between consecutive video frames.
- Employs a generalized bipartite matching algorithm with weights derived from segment overlap scores.
- Utilizes an integer linear program to solve the matching task for robust cell linking.
Main Results:
- The topological alignment method effectively links segmentations across frames.
- Demonstrates efficient and practical solutions for the matching task.
- Successfully applied to cell tracking in Large Scale Digital Cell Analysis System (LSDCAS) datasets.
Conclusions:
- The proposed topological alignment method is effective and useful for cell tracking.
- The approach provides a robust solution to segmentation inaccuracies in bioimage analysis.
- Source code is available for the implementation.
