Related Experiment Video
Updated: Jul 4, 2026
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Robust segmentation of cerebral arterial segments by a sequential Monte Carlo method: particle filtering
Hackjoon Shim1, Dongjin Kwon, Il Dong Yun
1School of Electrical Engineering and Computer Science, Seoul National University, Seoul 151-742, Republic of Korea. hjshim@diehard.snu.ac.kr
Computer Methods and Programs in Biomedicine
|October 19, 2006
Summary
This study introduces a robust method for segmenting cerebral arteries in CT angiography (CTA) by tracking elliptical vessel segments. The adaptive particle filter effectively handles bone contact and vein contamination, improving accuracy.
Area of Science:
- Medical Imaging
- Biomedical Engineering
- Radiology
Background:
- Accurate segmentation of cerebral arteries from CT angiography (CTA) is crucial for diagnosing cerebrovascular diseases.
- Challenges in CTA segmentation include bone interference and contamination from surrounding veins.
Purpose of the Study:
- To propose and evaluate a novel method for robust cerebral arterial segmentation in CTA.
- To address the limitations of existing methods concerning bone contact and vein contamination.
Main Methods:
- A novel method treating vessel segments as 3D traveling ellipses.
- Utilizing a particle filter framework with adaptive capabilities for tracking.
- Segmentation achieved by tracking the ellipse through spatial sequences.
Main Results:
- The proposed method demonstrated robustness in experiments with both synthetic and real CTA data.
- The technique showed insensitivity to axis curvature changes, obscure boundaries, and noise.
- Reduced user intervention and parameter sensitivity compared to conventional methods.
Conclusions:
- The developed method offers improved robustness and accuracy for cerebral arterial segmentation in CTA.
- This approach has the potential to enhance the diagnosis and treatment planning of cerebrovascular conditions.
