Siam Tahsin Bhuiyan
- Intern, CCDS
Bachelors
Independent University, Bangladesh
Published at
Scientific Reports
Conference Paper2025
Deep Learning-Based Stenosis Segmentation in X-ray Angiography: Vision Transformers vs. CNNs
Md Jahidul Hossain Mekat, Kazi Samin Nawal, Jasper Oliver D Rozario, Siam Tahsin Bhuiyan, Md Asif Bin Khaled, Md Rashedur Rahman, Ashraful Islam, Saadia Binte Alam
2025 28th International Conference on Computer and Information Technology (ICCIT)
IEEE, pp. 3553-3558
Abstract
The accurate segmentation of stenosis in coronary angiography images is a critical task for the diagnosis of Coronary Artery Disease (CAD). Manual analysis of these images is often time-consuming and subject to interobserver variability. In this study, the performance of several popular Convolutional Neural Network (CNN) models was evaluated against a modern Vision Transformer (ViT) architecture for the task of automatic stenosis segmentation. Eight CNN architectures, including variants of ResNet and EfficientNet, were compared with the two SegFormer models, which utilize a Mix-Transformer (MiT) backbone, on a publicly available angiography dataset. Model performance was quantitatively assessed using the Dice Score Coefficient (DSC) and Intersection over Union (IoU). Among the evaluated models, the SegFormer-MiT-B3 architecture achieved the highest performance, yielding a DSC of 0.632 and an IoU of 0.462, which surpassed all the other tested CNN models. These findings demonstrate the superior capability of Transformerbased models in modeling complex vessel morphology, establishing them as a promising foundation for the development of robust automated diagnostic tools for CAD.
Keywords
- EfficientNet
- vision transformer
- convolutional neural networks
- ResNet
- X-ray angiography
- coronary artery disease
- SegFormer
- automated image analysis
- vascular morphology
CCDS Authors
MD Asif Bin Khaled
- Co-Director, MIRA Wing
Published at
IEEE Internet of Things Journal
Md. Rashedur Rahman, D. Eng.
- Co-Director, MIRA & NEST Wings
Assistant Professor
Department of Computer Science and Engineering
Independent University, Bangladesh
Published at
CHI, HCII, Scientific Reports, Cognitive Computation, Neural Computing and Applications
Ashraful Islam, PhD
- Center Director, CCDS
- Director, HCI Wing
Assistant Professor
Department of Computer Science and Engineering
Independent University, Bangladesh (IUB)Published at
CHI, UIST, UbiComp, IJCNN, ICIP, HCII, ICDMW, BLP, IEEE Internet of Things Journal, JMIR, JMIR Human Factors, JMIR Formative Research, IEEE Access, Smart Agricultural Technology, Diabetes Research and Clinical Practice, PLOS ONE
Research Interests
Human-Computer Interaction, AI for Social Good, AI for Public Health, AI for Impact
Saadia Binte Alam, PhD
- Director, MIRA & WiSE Wings
Associate Professor
Department of Computer Science and Engineering
Independent University, Bangladesh
Published at
HCII, Scientific Reports, Cognitive Computation, Neural Computing and Applications
References
- 1.Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. (2016). Deep Residual Learning for Image Recognition. , 770–778[10.1109/cvpr.2016.90]
- 2.Gao Huang, Zhuang Liu, Laurens Van Der Maaten, Kilian Q. Weinberger. (2017). Densely Connected Convolutional Networks. , 2261–2269[10.1109/cvpr.2017.243]
- 3.Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen. (2018). MobileNetV2: Inverted Residuals and Linear Bottlenecks. , 4510–4520[10.1109/cvpr.2018.00474]
- 4.Saining Xie, Ross Girshick, Piotr Dollar, Zhuowen Tu, Kaiming He. (2017). Aggregated Residual Transformations for Deep Neural Networks. , 5987–5995[10.1109/cvpr.2017.634]
- 5.Ali Rostami, Fatemeh Fouladi, Hedieh Sajedi. (2024). Segmentation of Coronary Artery Stenosis in X-Ray Angiography Using Mamba Models. , 326–331[10.1109/iccke65377.2024.10874801]
- 6.Vibha Gupta, Petur Petursson, Aidin Rawshani, Jan Boren, Truls Ramunddal, Deepak L Bhatt, Elmir Omerovic, Oskar Angerås, Gustav Smith, Naveed Sattar, Erik Andersson, Björn Redfors, Lukas Hilgendorf, Göran Bergström, Carlo Pirazzi, Kristofer Skoglund, Araz Rawshani. (2025). End-to-end deep-learning model for the detection of coronary artery stenosis on coronary CT images. Open Heart, 12(1), e002998[10.1136/openhrt-2024-002998]
- 7.Maxim Popov, Akmaral Amanturdieva, Nuren Zhaksylyk, Alsabir Alkanov, Adilbek Saniyazbekov, Temirgali Aimyshev, Eldar Ismailov, Ablay Bulegenov, Arystan Kuzhukeyev, Aizhan Kulanbayeva, Almat Kalzhanov, Nurzhan Temenov, Alexey Kolesnikov, Orazbek Sakhov, Siamac Fazli. (2024). Dataset for Automatic Region-based Coronary Artery Disease Diagnostics Using X-Ray Angiography Images. Scientific Data, 11(1), 20[10.1038/s41597-023-02871-z]