Related Experiment Videos
Model-based morphological segmentation and labeling of coronary angiograms
K Haris1, S N Efstratiadis, N Maglaveras
1Laboratory of Medical Informatics, Faculty of Medicine, Aristotle University, Thessalonik, Greece. haris@med.auth.gr
IEEE Transactions on Medical Imaging
|January 11, 2000
Summary
This study introduces a novel method for automatically extracting and labeling the coronary arterial tree (CAT) from single-view angiograms with minimal user input. The approach accurately identifies artery structures and dimensions, improving diagnostic capabilities.
Area of Science:
- Medical Imaging Analysis
- Cardiovascular Imaging
- Computational Anatomy
Background:
- Accurate analysis of the coronary arterial tree (CAT) is crucial for diagnosing cardiovascular diseases.
- Existing methods for CAT extraction and labeling from single-view angiograms often require significant user supervision.
- Automated and precise methods are needed to improve efficiency and reduce inter-observer variability in cardiac imaging analysis.
Purpose of the Study:
- To propose and validate a novel method for automated extraction and labeling of the coronary arterial tree (CAT) from single-view angiograms.
- To achieve minimal user supervision in the process of identifying coronary artery structures and dimensions.
- To generate a structural description (skeleton and borders) and assign coded labels to the CAT based on a coronary artery model.
Main Methods:
- A four-stage method involving CAT tracking and detection, skeleton and border estimation, feature graph creation, and artery labeling via graph matching.
- Extraction of approximate CAT centerline and borders using recursive tracking with circular template analysis.
- Refinement of skeleton and borders using morphological homotopy modification and watershed transform, followed by graph matching for labeling.
Main Results:
- The proposed method successfully extracts structural information (skeleton and borders) of the coronary arterial tree.
- Quantitative information regarding artery dimensions is obtained, and coded labels are assigned through graph matching.
- Experimental results on clinical digitized coronary angiograms demonstrate the effectiveness of the automated approach.
Conclusions:
- The developed method offers an automated solution for coronary arterial tree extraction and labeling with minimal user intervention.
- This technique provides accurate structural and quantitative information, facilitating improved analysis of coronary angiograms.
- The approach holds potential for enhancing the efficiency and reliability of cardiovascular diagnostic procedures.