Related Experiment Video
Updated: Jun 23, 2026
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
Coronary artery stenosis segmentation using U-Net architecture with customised loss function
Nimra Iman1, Romana Aziz2, Mahwish Ilyas3
1Department of Software Engineering, Faculty of Computing and IT, University of Sargodha, Sargodha, Pakistan.
Insights
This study introduces a deep learning framework for automated coronary artery stenosis detection using X-ray angiography. The advanced U-Net model significantly improves stenosis segmentation accuracy compared to previous methods.
Area of Science:
- Cardiology
- Medical Imaging
- Artificial Intelligence
Background:
- Cardiovascular disorders are the leading global cause of death.
- Coronary artery stenosis is a significant cardiovascular disorder.
- Manual stenosis detection is time-consuming and error-prone.
Purpose of the Study:
- To develop a fully automated deep learning framework for binary stenosis segmentation in coronary arteries.
- To evaluate the performance of different U-Net architectures for this task.
- To improve the accuracy and efficiency of stenosis detection.
Main Methods:
- Utilized a deep learning framework based on U-Net architectures.
- Compared standard U-Net, U-Net with squeeze-and-excitation blocks, and U-Net with dense blocks.
- Employed a custom loss function and data augmentation on the ARCADE dataset.
- The dataset includes X-ray angiography images from 1,500 patients.
Main Results:
- The U-Net incorporating dense blocks achieved the best performance.
- The model achieved a precision of 0.5985, recall of 0.6319, and F1 Score of 61.47%.
- This represents a significant improvement over the previous F1 Score of 53.4%.
Conclusions:
- The proposed deep learning framework shows promising performance for automated stenosis segmentation.
- The method successfully segments stenotic regions despite challenges like small vessel size and low contrast.
- This automated approach can address the limitations of manual stenosis detection.
Abstract:
The disorders that affect our heart and blood vessels are cardiovascular disorders, and they are the leading cause of death worldwide. A significant disorder among these is coronary artery stenosis. Stenosis detection is a time-consuming process that requires an expert cardiologist and is also prone to human error. A fully automated system can handle all these challenges. Therefore, we presented a deep learning-based framework for binary stenosis segmentation in the coronary arteries using an X-ray angiography dataset. In this research work, three architectures 1st standard U-Net, then the U-Net enhanced with squeeze-and-excitation blocks, and last the U-Net incorporating dense blocks, where the final configuration achieved the best performance in the stenosis segmentation task. A custom loss function is employed to enhance model performance, utilising the publicly available ARCADE dataset. This dataset comprises X-ray angiography images from 1,500 patients, with 1,000 for training, 300 for validation, and 200 for testing. The training dataset was augmented to address limited data availability and enhance model generalizability. The model achieved a precision of 0.5985, a recall of 0.6319, and an F1 Score of 61.47%, whereas previous research in this challenge has achieved only a 53.4% F1 Score. The experimental results show that our method can achieve promising performance, which successfully segments the stenotic regions under the effect of small vessel size and low contrast.
