DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture
IntroductionGlioma segmentation is vital for diagnostic decision-making, monitoring disease progression, and surgical planning. However, this task is hindered by substantial heterogeneity within gliomas and imbalanced region distributions, posing challenges to existing segmentation methods.MethodsTo...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2025.1449911/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832542158749433856 |
|---|---|
| author | Ruipeng Li Yuehui Liao Yueqi Huang Xiaofei Ma Guohua Zhao Yanbin Wang Chen Song |
| author_facet | Ruipeng Li Yuehui Liao Yueqi Huang Xiaofei Ma Guohua Zhao Yanbin Wang Chen Song |
| author_sort | Ruipeng Li |
| collection | DOAJ |
| description | IntroductionGlioma segmentation is vital for diagnostic decision-making, monitoring disease progression, and surgical planning. However, this task is hindered by substantial heterogeneity within gliomas and imbalanced region distributions, posing challenges to existing segmentation methods.MethodsTo address these challenges, we propose the DeepGlioSeg network, a U-shaped architecture with skip connections for continuous contextual feature integration. The model includes two primary components. First, a CTPC (CNN-Transformer Parallel Combination) module leverages parallel branches of CNN and Transformer networks to fuse local and global features of glioma images, enhancing feature representation. Second, the model computes a region-based probability by comparing the number of pixels in tumor and background regions and assigns greater weight to regions with lower probabilities, thereby focusing on the tumor segment. Test-time augmentation (TTA) and volume-constrained (VC) post-processing are subsequently applied to refine the final segmentation outputs.ResultsExtensive experiments were conducted on three publicly available glioma MRI datasets and one privately owned clinical dataset. The quantitative and qualitative findings consistently show that DeepGlioSeg achieves superior segmentation performance over other state-of-the-art methods.DiscussionBy integrating CNN- and Transformer-based features in parallel and adaptively emphasizing underrepresented tumor regions, DeepGlioSeg effectively addresses the challenges associated with glioma heterogeneity and imbalanced region distributions. The final pipeline, augmented with TTA and VC post-processing, demonstrates robust segmentation capabilities. The source code for this work is publicly available at https://github.com/smallboy-code/Brain-tumor-segmentation. |
| format | Article |
| id | doaj-art-3217ae496a3d4c1a890d1e84d0aa07f6 |
| institution | Kabale University |
| issn | 2234-943X |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj-art-3217ae496a3d4c1a890d1e84d0aa07f62025-02-04T07:41:45ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2025-02-011510.3389/fonc.2025.14499111449911DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architectureRuipeng Li0Yuehui Liao1Yueqi Huang2Xiaofei Ma3Guohua Zhao4Yanbin Wang5Chen Song6Department of Urology, Hangzhou Third People’s Hospital, Hangzhou, ChinaCollege of Medical Technology, Zhejiang Chinese Medical University, Hangzhou, ChinaDepartment of Psychiatry, Hangzhou Seventh People’s Hospital, Hangzhou, ChinaCollege of Medical Technology, Zhejiang Chinese Medical University, Hangzhou, ChinaDepartment of Magnetic Resonance Imaging, First Affiliated Hospital of Zhengzhou University, Zhengzhou, ChinaDepartment of Urology, Hangzhou Third People’s Hospital, Hangzhou, ChinaDepartment of Urology, Hangzhou Third People’s Hospital, Hangzhou, ChinaIntroductionGlioma segmentation is vital for diagnostic decision-making, monitoring disease progression, and surgical planning. However, this task is hindered by substantial heterogeneity within gliomas and imbalanced region distributions, posing challenges to existing segmentation methods.MethodsTo address these challenges, we propose the DeepGlioSeg network, a U-shaped architecture with skip connections for continuous contextual feature integration. The model includes two primary components. First, a CTPC (CNN-Transformer Parallel Combination) module leverages parallel branches of CNN and Transformer networks to fuse local and global features of glioma images, enhancing feature representation. Second, the model computes a region-based probability by comparing the number of pixels in tumor and background regions and assigns greater weight to regions with lower probabilities, thereby focusing on the tumor segment. Test-time augmentation (TTA) and volume-constrained (VC) post-processing are subsequently applied to refine the final segmentation outputs.ResultsExtensive experiments were conducted on three publicly available glioma MRI datasets and one privately owned clinical dataset. The quantitative and qualitative findings consistently show that DeepGlioSeg achieves superior segmentation performance over other state-of-the-art methods.DiscussionBy integrating CNN- and Transformer-based features in parallel and adaptively emphasizing underrepresented tumor regions, DeepGlioSeg effectively addresses the challenges associated with glioma heterogeneity and imbalanced region distributions. The final pipeline, augmented with TTA and VC post-processing, demonstrates robust segmentation capabilities. The source code for this work is publicly available at https://github.com/smallboy-code/Brain-tumor-segmentation.https://www.frontiersin.org/articles/10.3389/fonc.2025.1449911/fullautomated segmentationgliomaCTPCconvolutional neural networkmagnetic resonance imaging |
| spellingShingle | Ruipeng Li Yuehui Liao Yueqi Huang Xiaofei Ma Guohua Zhao Yanbin Wang Chen Song DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture Frontiers in Oncology automated segmentation glioma CTPC convolutional neural network magnetic resonance imaging |
| title | DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture |
| title_full | DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture |
| title_fullStr | DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture |
| title_full_unstemmed | DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture |
| title_short | DeepGlioSeg: advanced glioma MRI data segmentation with integrated local-global representation architecture |
| title_sort | deepglioseg advanced glioma mri data segmentation with integrated local global representation architecture |
| topic | automated segmentation glioma CTPC convolutional neural network magnetic resonance imaging |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2025.1449911/full |
| work_keys_str_mv | AT ruipengli deepgliosegadvancedgliomamridatasegmentationwithintegratedlocalglobalrepresentationarchitecture AT yuehuiliao deepgliosegadvancedgliomamridatasegmentationwithintegratedlocalglobalrepresentationarchitecture AT yueqihuang deepgliosegadvancedgliomamridatasegmentationwithintegratedlocalglobalrepresentationarchitecture AT xiaofeima deepgliosegadvancedgliomamridatasegmentationwithintegratedlocalglobalrepresentationarchitecture AT guohuazhao deepgliosegadvancedgliomamridatasegmentationwithintegratedlocalglobalrepresentationarchitecture AT yanbinwang deepgliosegadvancedgliomamridatasegmentationwithintegratedlocalglobalrepresentationarchitecture AT chensong deepgliosegadvancedgliomamridatasegmentationwithintegratedlocalglobalrepresentationarchitecture |