Related Experiment Video
Updated: Jan 17, 2026
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
EigenU-Net: integrating eigenvalue decomposition of the Hessian into U-Net for 3D coronary artery segmentation
Cathy Ong Ly1,2,3,4, Chris McIntosh1,2,3,4,5
1Department of Medical Biophysics, University of Toronto, Toronto, Canada.
Insights
A new EigenU-Net model improves coronary artery segmentation in cardiac CT angiography (CCTA) by embedding vessel geometry. This AI approach achieves 87% accuracy, aiding cardiovascular disease diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Cardiovascular Disease Research
Background:
- Manual coronary artery segmentation from cardiac computed tomography angiography (CCTA) is labor-intensive and requires specialized expertise.
- Accurate segmentation is crucial for diagnosing and treating cardiovascular diseases.
Purpose of the Study:
- To develop an automated coronary artery segmentation method for CCTA images.
- To integrate geometrical properties of tubular structures directly into a deep learning model.
Main Methods:
- Developed EigenU-Net, a U-Net based architecture incorporating eigenvalues of the Hessian matrix to analyze local image structures.
- Integrated a closed-form solution for Hessian matrix eigenvalues as input features.
- Evaluated the model on the public IMAGECAS dataset (1000 CCTAs).
Main Results:
- The EigenU-Net model, with Gaussian pre-filtering, achieved a centerline Dice score of 87% on the IMAGECAS dataset.
- Demonstrated the model's feasibility and potential for accurate coronary artery segmentation.
- EigenU-Net successfully segmented regions missed by other segmentation models.
Conclusions:
- Directly integrating eigenvalue calculations into EigenU-Net effectively captures coronary vessel structural information.
- EigenU-Net offers a promising automated solution for coronary artery segmentation in CCTA.
- This approach has the potential to improve cardiovascular disease diagnosis and treatment planning.
Abstract:
Objective. Coronary artery segmentation is critical in medical imaging for the diagnosis and treatment of cardiovascular disease. However, manual segmentation of the coronary arteries is time-consuming and requires a high level of training and expertise.Approach. Our model, EigenU-Net, presents a novel approach to coronary artery segmentation of cardiac computed tomography angiography (CCTA) images that seeks to directly embed the geometrical properties of tubular structures, i.e. arteries, into the model. To examine the local structure of objects in the image we have integrated a closed-form solution of the eigenvalues of the Hessian matrix of each voxel for input into an U-Net based architecture.Main results. We demonstrate the feasibility and potential of our approach on the public IMAGECAS dataset consisting of 1000 CCTAs. The best performing model at 87% centerline Dice was EigenU-Net with Gaussian pre-filtering of the images.Significance. We were able to directly integrate the calculation of eigenvalues into our model EigenU-Net, to capture more information about the structure of the coronary vessels. EigenU-Net was able to segment regions that were overlooked by other models.
More Related Videos
05:56Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
06:18Intravascular Ultrasound Image-Based Finite Element Modeling Approach for Quantifying In Vivo Mechanical Properties of Human Coronary Artery
Published on: December 6, 2024
