Related Experiment Videos
Extracting branching tubular object geometry via cores
Yonatan Fridman1, Stephen M Pizer, Stephen Aylward
1Medical Image Display and Analysis Group, University of North Carolina, Chapel Hill, NC, USA. fridman@cs.unc.edu
Medical Image Analysis
|September 29, 2004
Summary
This study introduces a novel method for extracting branching structures from medical images. The technique accurately identifies tubular objects and their branching geometry, even in noisy data.
Area of Science:
- Medical imaging
- Computational anatomy
- Image processing
Background:
- Human anatomy, such as blood vessels, often exhibits complex branching tubular structures.
- Accurate extraction of these structures is crucial for medical diagnosis and analysis.
Purpose of the Study:
- To develop a robust method for extracting the 3D branching geometry and tubular structures from images.
- To identify object bifurcations within these structures.
Main Methods:
- Utilizing skeletons computed as cores, derived from medialness values measuring object centrality.
- Employing an affine-invariant corner detector for bifurcation detection.
- Evaluating methods on synthetic data and head MR angiogram data.
Main Results:
- The method demonstrates high resistance to noise in image data.
- Successful detection of branches with varying widths and branching angles.
- Validation on both synthetic and real-world medical imaging data.
Conclusions:
- The core-based skeletonization effectively extracts branching tubular structures and their geometry.
- The developed method offers a robust solution for analyzing complex anatomical branching patterns.
- An extension allows for the extraction of general branching structures beyond just tubes.