Swin-Unet vs U-Net in MRI images of right ventricular Segmentation: Comparative study

DOI

10.2991/978-94-6463-805-9_13How to use a DOI?

Keywords

Cardiac image segmentation; right ventricle; deep learning; U-Net; Swin-U-Net; transformer-based models; Dice coefficient

Abstract

Medical imaging is one of the essential technology used today for diagnosis as it provides detailed insight into internal heart structures, facilitates early detection of diseases and guides doctors in making treatment decisions. However, despite advances in technology, cardiac image segmentation remains challenging, particularly in right ventricular segmentation due to its complex anatomy and motion. the traditional methods while achieving high accuracy and efficiency, are often computationally intensive. In contrast, deep learning methods like U-net are highly regarded in medical image segmentation to their architecture and encoder-decoder structure. Furthermore, transformer-based models like Swin-U-Net enhance segmentation performance by capturing long range dependencies. this study provides comparative analysis of deep learning models U-Net and Swin-Unet, evaluating their preference in the right ventricular cardiac image segmentation with focuses on the dice coefficient.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Nouha Benzine
AU  - Tebra Abbassi
AU  - Asma Ammari
AU  - Iman Youkana
AU  - Rachida Ben Abedelaziz
PY  - 2025
DA  - 2025/08/05
TI  - Swin-Unet vs U-Net in MRI images of right ventricular Segmentation: Comparative study
BT  - Proceedings of the First International Conference on Artificial Intelligence, Smart Technologies and Communications (AISTC 2025)
PB  - Atlantis Press
SP  - 104
EP  - 117
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-805-9_13
DO  - 10.2991/978-94-6463-805-9_13
ID  - Benzine2025
ER  -