Related Experiment Video
Updated: Nov 22, 2025
10:23
Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
3.3K
Bayesian Estimation of Geometric Morphometric Landmarks for Simultaneous Localization of Multiple Anatomies in
Byunghwan Jeon1, Sunghee Jung2, Hackjoon Shim2
1School of Computer Science, Kyungil University, Gyeongsan 38428, Korea.
Entropy (Basel, Switzerland)
|January 6, 2021
Summary
This study introduces a robust method for simultaneously localizing multiple structures in cardiac computed tomography angiography (CTA) images. The approach effectively uses anatomical priors and relative distances to precisely identify objects like pulmonary veins (PVs) and the left atrial appendage (LAA).
Area of Science:
- Medical Imaging
- Computer Vision
- Anatomy
Background:
- Accurate localization of multiple anatomical structures in cardiac computed tomography angiography (CTA) is crucial for diagnosis and treatment planning.
- Existing methods may struggle with objects exhibiting unclear boundaries or complex spatial relationships.
Purpose of the Study:
- To develop a robust method for simultaneous multi-object localization in 3D cardiac CTA images.
- To leverage anatomical prior knowledge and relative object positioning for improved localization accuracy.
Main Methods:
- Utilized the maximum a posteriori (MAP) estimator incorporating anatomical priors.
- Integrated geometric morphological relationships between target and reference objects (e.g., aorta cross-sections).
- Introduced a novel pixel feature based on relative distances to define objects with unclear boundaries.
Main Results:
- Demonstrated robust simultaneous localization of four pulmonary veins (PVs) and the left atrial appendage (LAA) in cardiac CTA.
- The proposed method effectively handles objects with indistinct boundaries.
- Experimental validation confirmed the method's reliability.
Conclusions:
- The developed method offers a robust solution for simultaneous multi-object localization in cardiac CTA.
- The approach shows potential for extension to other anatomical structures and imaging applications.
- This technique enhances the precision of anatomical landmark identification in cardiovascular imaging.
