Related Experiment Video
Updated: Jul 10, 2026
10:59
Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands
Published on: July 26, 2014
Topology correction of segmented medical images using a fast marching algorithm.
Pierre-Louis Bazin1, Dzung L Pham
1Laboratory of Medical Image Computing, Neuroradiology Division, Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, MD 21218, USA. pbazin1@jhmi.edu
Computer Methods and Programs in Biomedicine
|October 19, 2007
Summary
We developed a new method to correct object topology in medical images. This technique works directly on image intensities, improving topology propagation for all object surfaces.
Area of Science:
- Medical image analysis
- Computational anatomy
- Image processing
Background:
- Accurate object topology is crucial for medical image analysis.
- Existing methods often alter segmented surfaces, limiting their application.
- Probabilistic or fuzzy segmentations present unique topological challenges.
Purpose of the Study:
- To present a novel method for correcting object topology in medical image segmentation.
- To enable topology correction directly from image intensities, accommodating probabilistic segmentations.
- To provide a robust algorithm for enforcing desired topologies in medical imaging.
Main Methods:
- Developed a topology propagation algorithm based on analyzing topological changes and critical points in implicit surfaces.
- Utilized a fast marching technique to enforce desired topologies.
- Applied the method directly to image intensities of probabilistic or fuzzy segmentations.
Main Results:
- Successfully corrected the topology of the cortical gray matter/white matter interface in segmented brain images.
- Demonstrated topology propagation across all isosurfaces of an object.
- The method proved effective for complex topological corrections.
Conclusions:
- The new method offers a significant advancement in correcting object topology from medical image intensities.
- It provides a versatile tool applicable to various probabilistic and fuzzy segmentation scenarios.
- The publicly released software plug-in facilitates broader adoption and research in medical image analysis.
