Related Experiment Video
Updated: Jun 6, 2026
09:36
Outer-Boundary Assisted Segmentation and Quantification of Trabecular Bones by an Imagej Plugin
Published on: March 14, 2018
Tubular surface segmentation for extracting anatomical structures from medical imagery.
Vandana Mohan1, Ganesh Sundaramoorthi, Allen Tannenbaum
1Schools of Electrical and Computer Engineering and Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA 30332, USA. vandana@gatech.edu
IEEE Transactions on Medical Imaging
|December 2, 2010
Summary
This study introduces a novel tubular model and algorithm for automatically extracting brain fiber bundles and blood vessels from medical images. This method enhances diagnostic accuracy for neurological and cardiovascular conditions.
Area of Science:
- Medical Imaging Analysis
- Computational Anatomy
- Biomedical Engineering
Background:
- Accurate segmentation of tubular anatomical structures like brain fiber bundles and blood vessels is crucial for diagnosing neurological and cardiovascular diseases.
- Existing methods for extracting these structures often face challenges with local minima and computational efficiency.
Purpose of the Study:
- To develop a robust model for tubular structures and an algorithm for their automatic extraction from medical imagery.
- To improve the efficiency and accuracy of segmenting structures such as the cingulum bundle and blood vessel trees.
Main Methods:
- A novel tubular model represented by a 4-D curve (radius function and centerline) was developed.
- An algorithm was devised to automatically detect and evolve branches, generalizing to tubular trees.
- The algorithm was tested on diffusion-weighted magnetic resonance (DW-MRI) and computed tomography angiogram (CTA) datasets.
Main Results:
- The proposed model offers advantages over existing methods, including reduced sensitivity to local minima due to fewer degrees-of-freedom.
- The 4-D representation allows for computationally efficient extraction of both the tube and its centerline.
- The algorithm successfully extracted tubular structures, including the cingulum bundle and blood vessel trees, with performance validated on multiple datasets.
Conclusions:
- The developed tubular model and extraction algorithm provide an efficient and accurate method for segmenting complex anatomical structures in medical imaging.
- This approach has significant potential for improving the diagnosis of conditions linked to the cingulum bundle and cardiovascular diseases.
