A transformer-based generative adversarial network for brain tumor segmentation
Brain tumor segmentation remains a challenge in medical image segmentation tasks. With the application of transformer in various computer vision tasks, transformer blocks show the capability of learning long-distance dependency in global space, which is complementary to CNNs. In this paper, we propo...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-11-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2022.1054948/full |
| _version_ | 1811211322968244224 |
|---|---|
| author | Liqun Huang Enjun Zhu Long Chen Zhaoyang Wang Senchun Chai Baihai Zhang |
| author_facet | Liqun Huang Enjun Zhu Long Chen Zhaoyang Wang Senchun Chai Baihai Zhang |
| author_sort | Liqun Huang |
| collection | DOAJ |
| description | Brain tumor segmentation remains a challenge in medical image segmentation tasks. With the application of transformer in various computer vision tasks, transformer blocks show the capability of learning long-distance dependency in global space, which is complementary to CNNs. In this paper, we proposed a novel transformer-based generative adversarial network to automatically segment brain tumors with multi-modalities MRI. Our architecture consists of a generator and a discriminator, which is trained in min–max game progress. The generator is based on a typical “U-shaped” encoder–decoder architecture, whose bottom layer is composed of transformer blocks with Resnet. Besides, the generator is trained with deep supervision technology. The discriminator we designed is a CNN-based network with multi-scale L1 loss, which is proved to be effective for medical semantic image segmentation. To validate the effectiveness of our method, we conducted exclusive experiments on BRATS2015 dataset, achieving comparable or better performance than previous state-of-the-art methods. On additional datasets, including BRATS2018 and BRATS2020, experimental results prove that our technique is capable of generalizing successfully. |
| first_indexed | 2024-04-12T05:10:57Z |
| format | Article |
| id | doaj.art-e39bf5258f014b1aa13ca1d5a472353d |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-04-12T05:10:57Z |
| publishDate | 2022-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-e39bf5258f014b1aa13ca1d5a472353d2022-12-22T03:46:45ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2022-11-011610.3389/fnins.2022.10549481054948A transformer-based generative adversarial network for brain tumor segmentationLiqun Huang0Enjun Zhu1Long Chen2Zhaoyang Wang3Senchun Chai4Baihai Zhang5The School of Automation, Beijing Institute of Technology, Beijing, ChinaDepartment of Cardiac Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, ChinaThe School of Automation, Beijing Institute of Technology, Beijing, ChinaThe School of Automation, Beijing Institute of Technology, Beijing, ChinaThe School of Automation, Beijing Institute of Technology, Beijing, ChinaThe School of Automation, Beijing Institute of Technology, Beijing, ChinaBrain tumor segmentation remains a challenge in medical image segmentation tasks. With the application of transformer in various computer vision tasks, transformer blocks show the capability of learning long-distance dependency in global space, which is complementary to CNNs. In this paper, we proposed a novel transformer-based generative adversarial network to automatically segment brain tumors with multi-modalities MRI. Our architecture consists of a generator and a discriminator, which is trained in min–max game progress. The generator is based on a typical “U-shaped” encoder–decoder architecture, whose bottom layer is composed of transformer blocks with Resnet. Besides, the generator is trained with deep supervision technology. The discriminator we designed is a CNN-based network with multi-scale L1 loss, which is proved to be effective for medical semantic image segmentation. To validate the effectiveness of our method, we conducted exclusive experiments on BRATS2015 dataset, achieving comparable or better performance than previous state-of-the-art methods. On additional datasets, including BRATS2018 and BRATS2020, experimental results prove that our technique is capable of generalizing successfully.https://www.frontiersin.org/articles/10.3389/fnins.2022.1054948/fullgenerative adversarial networktransformerdeep learningautomatic segmentationbrain tumor |
| spellingShingle | Liqun Huang Enjun Zhu Long Chen Zhaoyang Wang Senchun Chai Baihai Zhang A transformer-based generative adversarial network for brain tumor segmentation Frontiers in Neuroscience generative adversarial network transformer deep learning automatic segmentation brain tumor |
| title | A transformer-based generative adversarial network for brain tumor segmentation |
| title_full | A transformer-based generative adversarial network for brain tumor segmentation |
| title_fullStr | A transformer-based generative adversarial network for brain tumor segmentation |
| title_full_unstemmed | A transformer-based generative adversarial network for brain tumor segmentation |
| title_short | A transformer-based generative adversarial network for brain tumor segmentation |
| title_sort | transformer based generative adversarial network for brain tumor segmentation |
| topic | generative adversarial network transformer deep learning automatic segmentation brain tumor |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2022.1054948/full |
| work_keys_str_mv | AT liqunhuang atransformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT enjunzhu atransformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT longchen atransformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT zhaoyangwang atransformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT senchunchai atransformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT baihaizhang atransformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT liqunhuang transformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT enjunzhu transformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT longchen transformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT zhaoyangwang transformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT senchunchai transformerbasedgenerativeadversarialnetworkforbraintumorsegmentation AT baihaizhang transformerbasedgenerativeadversarialnetworkforbraintumorsegmentation |