A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor Segmentation
Brain tumors are characterized by their relatively high incidence and mortality rates, highlighting the utmost importance of precise automatic segmentation for subsequent diagnosis and treatment. Although deep learning has significantly advanced the field of accurate and efficient automatic brain tu...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10815934/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841533425083744256 |
|---|---|
| author | Qiwei Xing Zhihua Li Yongxia Jing Xiaolin Chen |
| author_facet | Qiwei Xing Zhihua Li Yongxia Jing Xiaolin Chen |
| author_sort | Qiwei Xing |
| collection | DOAJ |
| description | Brain tumors are characterized by their relatively high incidence and mortality rates, highlighting the utmost importance of precise automatic segmentation for subsequent diagnosis and treatment. Although deep learning has significantly advanced the field of accurate and efficient automatic brain tumor segmentation, several challenges still persist. In this paper, we introduce a novel architecture called the Dual Encoder Mirror Difference Residual U-Net (DEMD-ResUNet). This approach incorporates dual encoders that process both the original and horizontally flipped images. Additionally, residual blocks are employed to substitute the original convolutional blocks in the encoder section of the U-Net structure. This modification not only streamlines network training but also mitigates issues related to network degradation and the loss of detailed information. To further enhance feature representation, we propose a Multimodal Difference Feature Augmentation (MDFA) module, which effectively highlights abnormal regions in both the original and mirrored brain tumor images to facilitate better feature discrimination. Moreover, a Mirror Difference Feature Fusion (MDFF) module is integrated between the dual encoders and the decoder. This module efficiently transfers features from both the original and mirrored images to the decoder, leveraging the symmetrical information inherent in the images and subsequently boosting the segmentation performance of the model. Ablation experiments conducted on the DEMD-ResUNet model demonstrate the efficacy of its various modules and hyperparameter settings. When evaluated on the BraTS 2018 and BraTS 2019 datasets, our model achieves impressive Dice similarity coefficient (DSC) values of 0.862, 0.925, and 0.905 for Enhanced tumor (ET), Whole tumor (WT), and Tumor core (TC) in the former, and 0.869, 0.922, and 0.916 in the latter, respectively. |
| format | Article |
| id | doaj-art-94221fd911674254807abc39d6ba31b0 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-94221fd911674254807abc39d6ba31b02025-01-16T00:01:47ZengIEEEIEEE Access2169-35362025-01-01131621163510.1109/ACCESS.2024.352268210815934A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor SegmentationQiwei Xing0https://orcid.org/0000-0002-5060-8669Zhihua Li1https://orcid.org/0000-0001-6564-6927Yongxia Jing2Xiaolin Chen3https://orcid.org/0000-0001-5910-0661Institute for Educational Big Data and Artificial Intelligence, Qiongtai Normal University, Haikou, ChinaInstitute for Educational Big Data and Artificial Intelligence, Qiongtai Normal University, Haikou, ChinaInstitute for Educational Big Data and Artificial Intelligence, Qiongtai Normal University, Haikou, ChinaInstitute for Educational Big Data and Artificial Intelligence, Qiongtai Normal University, Haikou, ChinaBrain tumors are characterized by their relatively high incidence and mortality rates, highlighting the utmost importance of precise automatic segmentation for subsequent diagnosis and treatment. Although deep learning has significantly advanced the field of accurate and efficient automatic brain tumor segmentation, several challenges still persist. In this paper, we introduce a novel architecture called the Dual Encoder Mirror Difference Residual U-Net (DEMD-ResUNet). This approach incorporates dual encoders that process both the original and horizontally flipped images. Additionally, residual blocks are employed to substitute the original convolutional blocks in the encoder section of the U-Net structure. This modification not only streamlines network training but also mitigates issues related to network degradation and the loss of detailed information. To further enhance feature representation, we propose a Multimodal Difference Feature Augmentation (MDFA) module, which effectively highlights abnormal regions in both the original and mirrored brain tumor images to facilitate better feature discrimination. Moreover, a Mirror Difference Feature Fusion (MDFF) module is integrated between the dual encoders and the decoder. This module efficiently transfers features from both the original and mirrored images to the decoder, leveraging the symmetrical information inherent in the images and subsequently boosting the segmentation performance of the model. Ablation experiments conducted on the DEMD-ResUNet model demonstrate the efficacy of its various modules and hyperparameter settings. When evaluated on the BraTS 2018 and BraTS 2019 datasets, our model achieves impressive Dice similarity coefficient (DSC) values of 0.862, 0.925, and 0.905 for Enhanced tumor (ET), Whole tumor (WT), and Tumor core (TC) in the former, and 0.869, 0.922, and 0.916 in the latter, respectively.https://ieeexplore.ieee.org/document/10815934/Multimodal MRIbrain tumor segmentationmirror differenceresidual U-Net |
| spellingShingle | Qiwei Xing Zhihua Li Yongxia Jing Xiaolin Chen A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor Segmentation IEEE Access Multimodal MRI brain tumor segmentation mirror difference residual U-Net |
| title | A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor Segmentation |
| title_full | A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor Segmentation |
| title_fullStr | A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor Segmentation |
| title_full_unstemmed | A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor Segmentation |
| title_short | A 3D Dual Encoder Mirror Difference ResU-Net for Multimodal Brain Tumor Segmentation |
| title_sort | 3d dual encoder mirror difference resu net for multimodal brain tumor segmentation |
| topic | Multimodal MRI brain tumor segmentation mirror difference residual U-Net |
| url | https://ieeexplore.ieee.org/document/10815934/ |
| work_keys_str_mv | AT qiweixing a3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation AT zhihuali a3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation AT yongxiajing a3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation AT xiaolinchen a3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation AT qiweixing 3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation AT zhihuali 3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation AT yongxiajing 3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation AT xiaolinchen 3ddualencodermirrordifferenceresunetformultimodalbraintumorsegmentation |