Related Experiment Video
Updated: Dec 7, 2025
06:57
Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
Published on: September 22, 2023
1.3K
Coronary artery segmentation in angiographic videos utilizing spatial-temporal information
Lu Wang1,2, Dongxue Liang3, Xiaolei Yin1,2
1The Future Laboratory, Tsinghua University, Beijing, 100084, China.
BMC Medical Imaging
|September 25, 2020
Summary
This study introduces a new method for segmenting coronary arteries using a 3D/2D convolutional network that analyzes image sequences. The approach improves blood vessel segmentation in coronary angiography, aiding cardiac interventions.
Area of Science:
- Medical Imaging
- Computer Vision
- Cardiovascular Interventions
Background:
- Coronary artery angiography is crucial for cardiac interventional surgery.
- Accurate segmentation of blood vessels in angiographic images is essential for diagnosing conditions like plaques and stenosis.
Purpose of the Study:
- To propose a novel coronary artery segmentation framework.
- To leverage temporal information from image sequences for improved segmentation accuracy.
Main Methods:
- A hybrid framework combining a 3D convolutional input layer and a 2D convolutional network.
- Inputting a sequence of coronary angiographic images to capture temporal dynamics.
- Utilizing down-sampling encoders, up-sampling decoders, bottleneck modules, and skip connections for segmentation.
Main Results:
- The proposed spatial-temporal model achieves effective segmentation even with low-quality coronary angiographic videos.
- The framework demonstrates superior performance compared to existing state-of-the-art segmentation techniques.
Conclusions:
- Integrating spatial and temporal information from image sequences significantly enhances the analysis and understanding of coronary angiographic videos.
- The developed framework offers a promising tool for improving the interpretation of coronary angiographic data.
