Related Experiment Video
Updated: May 27, 2025
Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
Published on: September 22, 2023
Automated Coronary Artery Segmentation with 3D PSPNET using Global Processing and Patch Based Methods on CCTA Images
Kavita Chachadi1, S R Nirmala2, Pavan G Netrakar2
1KLE Technological University, Hubballi, Karnataka, India. kavita.chachadi@kletech.ac.in.
Insights
This study modified a 2D deep learning model into a 3D Pyramid Scene Parsing Neural Network (PSPNet) for segmenting coronary arteries in 3D Coronary Computed Tomography Angiography (CCTA) images, achieving promising results for disease diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Cardiovascular Disease Research
Background:
- Coronary artery disease (CAD) is a leading global cause of death.
- Accurate coronary artery segmentation is crucial for diagnosing CAD, including stenosis and plaque analysis.
- Deep learning (DL) shows promise in medical image analysis, but 2D models have limitations for 3D data.
Purpose of the Study:
- To adapt the 2D Pyramid Scene Parsing Neural Network (PSPNet) into a 3D model for segmenting coronary arteries.
- To evaluate the performance of the proposed 3D PSPNet using both Global and Patch-based processing methods.
- To assess the potential of 3D PSPNet for improving the analysis of 3D Coronary Computed Tomography Angiography (CCTA) images.
Main Methods:
- Modification of the 2D PSPNet architecture to a 3D version.
- Application of the 3D PSPNet for semantic segmentation of coronary arteries in 3D CCTA datasets.
- Comparative evaluation of Global processing versus Patch-based processing strategies for segmentation.
- Utilizing the ImageCAS dataset for experimental validation.
Main Results:
- The 3D PSPNet achieved a Dice Similarity Coefficient (DSC) of 0.76 using the Global processing method.
- The Patch-based processing method yielded a DSC of 0.73.
- These results demonstrate the feasibility of the proposed 3D PSPNet for coronary artery segmentation.
Conclusions:
- The developed 3D PSPNet effectively segments coronary arteries from 3D CCTA images.
- The Global processing approach showed slightly superior performance compared to Patch-based processing in this study.
- This work contributes a novel deep learning approach for enhancing CAD diagnosis through improved image segmentation.
Abstract:
The prevalence of coronary artery disease (CAD) has become the major cause of death across the world in recent years. The accurate segmentation of coronary artery is important in clinical diagnosis and treatment of coronary artery disease (CAD) such as stenosis detection and plaque analysis. Deep learning techniques have been shown to assist medical experts in diagnosing diseases using biomedical imaging. There are many methods which employ 2D DL models for medical image segmentation. The 2D Pyramid Scene Parsing Neural Network (PSPNet) has potential in this domain but not explored for the segmentation of coronary arteries from 3D Coronary Computed Tomography Angiography (CCTA) images. The contribution of present research work is to propose the modification of 2D PSPNet into 3D PSPNet for segmenting the coronary arteries from 3D CCTA images. The innovative factor is to evaluate the network performance by employing Global processing and Patch based processing methods. The experimental results achieved a Dice Similarity Coefficient (DSC) of 0.76 for Global process method and 0.73 for Patch based method using a subset of 200 images from the ImageCAS dataset.
