Related Experiment Videos
Automated seed placement for colon segmentation in computed tomography colonography
Gheorghe Iordanescu1, Perry J Pickhardt, J Richard Choi
1Department of Radiology, National Institutes of Health, Building 10, Room 1C660, 10 Center Drive MSC 1182, Bethesda, MD 20892-1182, USA.
Academic Radiology
|February 22, 2005
Summary
This study presents an automated algorithm for placing seeds to segment the colon in computed tomography colonography (CTC). The method successfully segmented the colon in most cases, demonstrating feasibility for automated colon segmentation.
Area of Science:
- Medical Imaging
- Radiology
- Computational Anatomy
Background:
- Accurate colon segmentation is crucial for detecting colorectal diseases using computed tomography colonography (CTC).
- Manual seed placement for colon segmentation in CTC can be time-consuming and operator-dependent.
Purpose of the Study:
- To develop and evaluate an automated algorithm for precise seed localization within the colon lumen for segmentation.
- To improve the efficiency and reproducibility of colon segmentation in CTC.
Main Methods:
- The algorithm employs 2D morphological operators to identify colonic air pockets for seed placement in the cecum and rectum.
- Seed placement strategies were optimized for specific anatomical locations and CT slice characteristics.
- The automated algorithm was applied to a dataset of 292 consecutive CTC cases (146 prone, 146 supine).
Main Results:
- Complete colon segmentation was achieved in 83.2% of cases.
- Partial segmentation occurred in 9.6% of cases due to colon collapse or fluid interference.
- Segmentation leakage outside the colon occurred in 7.2% of cases, attributed to fluid detection algorithm limitations.
Conclusions:
- Fully automatic seed placement for colon segmentation in CTC is feasible in a majority of cases.
- The developed algorithm minimizes the seeding of undesired extracolonic air, enhancing segmentation accuracy.