Medical image segmentation by combining feature enhancement Swin Transformer and UperNet

Abstract Medical image segmentation plays a crucial role in assisting clinical diagnosis, yet existing models often struggle with handling diverse and complex medical data, particularly when dealing with multi-scale organ and tissue structures. This paper proposes a novel medical image segmentation...

Full description

Saved in:
Bibliographic Details
Main Authors: Lin Zhang, Xiaochun Yin, Xuqi Liu, Zengguang Liu
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-97779-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849713575610810368
author Lin Zhang
Xiaochun Yin
Xuqi Liu
Zengguang Liu
author_facet Lin Zhang
Xiaochun Yin
Xuqi Liu
Zengguang Liu
author_sort Lin Zhang
collection DOAJ
description Abstract Medical image segmentation plays a crucial role in assisting clinical diagnosis, yet existing models often struggle with handling diverse and complex medical data, particularly when dealing with multi-scale organ and tissue structures. This paper proposes a novel medical image segmentation model, FE-SwinUper, designed to address these challenges by integrating the strengths of the Swin Transformer and UPerNet architectures. The objective is to enhance multi-scale feature extraction and improve the fusion of hierarchical organ and tissue representations through a feature enhancement Swin Transformer (FE-ST) backbone and an adaptive feature fusion (AFF) module. The FE-ST backbone utilizes self-attention mechanisms to efficiently extract rich spatial and contextual features across different scales, while the AFF module adapts to multi-scale feature fusion, mitigating the loss of contextual information. We evaluate the model on two publicly available medical image segmentation datasets: Synapse multi-organ segmentation dataset and the ACDC cardiac segmentation dataset. Our results show that FE-SwinUper outperforms existing state-of-the-art models in terms of Dice coefficient, pixel accuracy, and Hausdorff distance. The model achieves a Dice score of 91.58% on the Synapse dataset and 90.15% on the ACDC dataset. These results demonstrate the robustness and efficiency of the proposed model, indicating its potential for real-world clinical applications.
format Article
id doaj-art-f8f264ae3e074c53adba1eedb6d0dc2f
institution DOAJ
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-f8f264ae3e074c53adba1eedb6d0dc2f2025-08-20T03:13:55ZengNature PortfolioScientific Reports2045-23222025-04-0115111310.1038/s41598-025-97779-6Medical image segmentation by combining feature enhancement Swin Transformer and UperNetLin Zhang0Xiaochun Yin1Xuqi Liu2Zengguang Liu3College of Computer Science, Weifang University of Science and TechnologyCollege of Computer Science, Weifang University of Science and TechnologyCollege of Computer Science, Weifang University of Science and TechnologySchool of Information Engineering, Shandong Vocational College of Science and TechnologyAbstract Medical image segmentation plays a crucial role in assisting clinical diagnosis, yet existing models often struggle with handling diverse and complex medical data, particularly when dealing with multi-scale organ and tissue structures. This paper proposes a novel medical image segmentation model, FE-SwinUper, designed to address these challenges by integrating the strengths of the Swin Transformer and UPerNet architectures. The objective is to enhance multi-scale feature extraction and improve the fusion of hierarchical organ and tissue representations through a feature enhancement Swin Transformer (FE-ST) backbone and an adaptive feature fusion (AFF) module. The FE-ST backbone utilizes self-attention mechanisms to efficiently extract rich spatial and contextual features across different scales, while the AFF module adapts to multi-scale feature fusion, mitigating the loss of contextual information. We evaluate the model on two publicly available medical image segmentation datasets: Synapse multi-organ segmentation dataset and the ACDC cardiac segmentation dataset. Our results show that FE-SwinUper outperforms existing state-of-the-art models in terms of Dice coefficient, pixel accuracy, and Hausdorff distance. The model achieves a Dice score of 91.58% on the Synapse dataset and 90.15% on the ACDC dataset. These results demonstrate the robustness and efficiency of the proposed model, indicating its potential for real-world clinical applications.https://doi.org/10.1038/s41598-025-97779-6Medical imageSemantic segmentationDeep learningFeature enhancement
spellingShingle Lin Zhang
Xiaochun Yin
Xuqi Liu
Zengguang Liu
Medical image segmentation by combining feature enhancement Swin Transformer and UperNet
Scientific Reports
Medical image
Semantic segmentation
Deep learning
Feature enhancement
title Medical image segmentation by combining feature enhancement Swin Transformer and UperNet
title_full Medical image segmentation by combining feature enhancement Swin Transformer and UperNet
title_fullStr Medical image segmentation by combining feature enhancement Swin Transformer and UperNet
title_full_unstemmed Medical image segmentation by combining feature enhancement Swin Transformer and UperNet
title_short Medical image segmentation by combining feature enhancement Swin Transformer and UperNet
title_sort medical image segmentation by combining feature enhancement swin transformer and upernet
topic Medical image
Semantic segmentation
Deep learning
Feature enhancement
url https://doi.org/10.1038/s41598-025-97779-6
work_keys_str_mv AT linzhang medicalimagesegmentationbycombiningfeatureenhancementswintransformerandupernet
AT xiaochunyin medicalimagesegmentationbycombiningfeatureenhancementswintransformerandupernet
AT xuqiliu medicalimagesegmentationbycombiningfeatureenhancementswintransformerandupernet
AT zengguangliu medicalimagesegmentationbycombiningfeatureenhancementswintransformerandupernet