Related Experiment Video
Updated: Aug 8, 2025
Identifying Coronary Artery Calcification on Non-gated Computed Tomography Scans
Published on: August 28, 2018
An automated segmentation of coronary artery calcification using deep learning in specific region limitation
Asmae Mama Zair1, Assia Bouzouad Cherfa2, Yazid Cherfa2
1University of Blida 01, B.P 270, Soumaa road, Blida, Algeria. asmaz.zair@gmail.com.
Insights
This study presents a deep learning method for automatically segmenting coronary artery calcification (CAC) in CT images. The approach accurately quantifies the Agatston score, aiding in the assessment of cardiovascular disease risk.
Area of Science:
- Medical imaging
- Artificial intelligence in healthcare
- Cardiovascular diagnostics
Background:
- Coronary artery calcification (CAC) is a common arterial disease.
- Untreated severe CAC can lead to permanent damage.
- Accurate quantification of CAC, often via the Agatston score, is crucial for patient management.
Purpose of the Study:
- To develop an automated method for segmenting CAC in 2D CT images.
- To accurately measure the Agatston score for CAC quantification.
- To improve the efficiency and precision of CAC assessment.
Main Methods:
- Utilized deep learning, specifically U-Net and SegNet-VGG16 models with transfer learning, for CAC segmentation.
- Implemented image processing techniques including thresholding and 2D connectivity for region isolation.
- Extracted the heart cavity using the convex hull of the lungs.
- Calculated the Agatston score for quantitative analysis.
Main Results:
- Achieved encouraging outcomes in automatically segmenting CAC from 2D CT images.
- Demonstrated the feasibility of using deep learning for accurate Agatston score prediction.
- The proposed strategy shows promise for clinical application.
Conclusions:
- Deep learning models effectively segment CAC in CT images.
- Automated Agatston score measurement is achievable with the proposed method.
- This approach offers a potential advancement in diagnosing and managing coronary artery disease.
Abstract:
Coronary artery calcification (CAC) is a frequent disease of the arteries that supply the surface of the heart muscle. Leaving a severe disease untreated can make it permanent. Computer tomography (CT), which is well known for its ability to quantify the Agatston score, is used to visualize high-resolution CACs. CAC segmentation is still an important topic. Our goal is to automatically segment CAC in a specific area and measure the Agatston score in 2D images. The heart region is limited using a threshold, unused structures are removed using 2D connectivity (muscle, lung, ribcage), the heart cavity is extracted using the convex hull of the lungs, and the CAC is then segmented in 2D using a convolutional neural network (U-Net models/SegNet-VGG16 with transfer learning). The Agatston score prediction is calculated for CAC quantification. The proposed strategy is tested through experiments, which yield encouraging outcomes. Graphical Abstract Deep learning for CAC segmentation in CT images.
More Related Videos
06:57Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
Published on: September 22, 2023
04:48Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
