DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT Registration
Medical image registration is a fundamental and indispensable element in medical image analysis, which can establish spatial consistency among corresponding anatomical structures across various medical images. Since images with different modalities exhibit different features, it remains a challenge...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/1/95 |
| _version_ | 1827384921559138304 |
|---|---|
| author | Aolin Yang Tiejun Yang Xiang Zhao Xin Zhang Yanghui Yan Chunxia Jiao |
| author_facet | Aolin Yang Tiejun Yang Xiang Zhao Xin Zhang Yanghui Yan Chunxia Jiao |
| author_sort | Aolin Yang |
| collection | DOAJ |
| description | Medical image registration is a fundamental and indispensable element in medical image analysis, which can establish spatial consistency among corresponding anatomical structures across various medical images. Since images with different modalities exhibit different features, it remains a challenge to find their exact correspondence. Most of the current methods based on image-to-image translation cannot fully leverage the available information, which will affect the subsequent registration performance. To solve the problem, we develop an unsupervised multimodal image registration method named DTR-GAN. Firstly, we design a multimodal registration framework via a bidirectional translation network to transform the multimodal image registration into a unimodal registration, which can effectively use the complementary information of different modalities. Then, to enhance the quality of the transformed images in the translation network, we design a multiscale encoder–decoder network that effectively captures both local and global features in images. Finally, we propose a mixed similarity loss to encourage the warped image to be closer to the target image in deep features. We extensively evaluate methods for MRI-CT image registration tasks of the abdominal cavity with advanced unsupervised multimodal image registration approaches. The results indicate that DTR-GAN obtains a competitive performance compared to other methods in MRI-CT registration. Compared with DFR, DTR-GAN has not only obtained performance improvements of 2.35% and 2.08% in the dice similarity coefficient (DSC) of MRI-CT registration and CT-MRI registration on the Learn2Reg dataset but has also decreased the average symmetric surface distance (ASD) by 0.33 mm and 0.12 mm on the Learn2Reg dataset. |
| first_indexed | 2024-03-08T15:12:22Z |
| format | Article |
| id | doaj.art-61c6ad5feba244998d777a131cd02ddd |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-08T15:12:22Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-61c6ad5feba244998d777a131cd02ddd2024-01-10T14:50:54ZengMDPI AGApplied Sciences2076-34172023-12-011419510.3390/app14010095DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT RegistrationAolin Yang0Tiejun Yang1Xiang Zhao2Xin Zhang3Yanghui Yan4Chunxia Jiao5School of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, ChinaSchool of Artificial Intelligence and Big Data, Henan University of Technology, Zhengzhou 450001, ChinaSchool of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, ChinaSchool of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, ChinaSchool of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, ChinaSchool of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, ChinaMedical image registration is a fundamental and indispensable element in medical image analysis, which can establish spatial consistency among corresponding anatomical structures across various medical images. Since images with different modalities exhibit different features, it remains a challenge to find their exact correspondence. Most of the current methods based on image-to-image translation cannot fully leverage the available information, which will affect the subsequent registration performance. To solve the problem, we develop an unsupervised multimodal image registration method named DTR-GAN. Firstly, we design a multimodal registration framework via a bidirectional translation network to transform the multimodal image registration into a unimodal registration, which can effectively use the complementary information of different modalities. Then, to enhance the quality of the transformed images in the translation network, we design a multiscale encoder–decoder network that effectively captures both local and global features in images. Finally, we propose a mixed similarity loss to encourage the warped image to be closer to the target image in deep features. We extensively evaluate methods for MRI-CT image registration tasks of the abdominal cavity with advanced unsupervised multimodal image registration approaches. The results indicate that DTR-GAN obtains a competitive performance compared to other methods in MRI-CT registration. Compared with DFR, DTR-GAN has not only obtained performance improvements of 2.35% and 2.08% in the dice similarity coefficient (DSC) of MRI-CT registration and CT-MRI registration on the Learn2Reg dataset but has also decreased the average symmetric surface distance (ASD) by 0.33 mm and 0.12 mm on the Learn2Reg dataset.https://www.mdpi.com/2076-3417/14/1/95multimodal image registrationimage-to-image translationunsuperviseddeep learning |
| spellingShingle | Aolin Yang Tiejun Yang Xiang Zhao Xin Zhang Yanghui Yan Chunxia Jiao DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT Registration Applied Sciences multimodal image registration image-to-image translation unsupervised deep learning |
| title | DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT Registration |
| title_full | DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT Registration |
| title_fullStr | DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT Registration |
| title_full_unstemmed | DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT Registration |
| title_short | DTR-GAN: An Unsupervised Bidirectional Translation Generative Adversarial Network for MRI-CT Registration |
| title_sort | dtr gan an unsupervised bidirectional translation generative adversarial network for mri ct registration |
| topic | multimodal image registration image-to-image translation unsupervised deep learning |
| url | https://www.mdpi.com/2076-3417/14/1/95 |
| work_keys_str_mv | AT aolinyang dtrgananunsupervisedbidirectionaltranslationgenerativeadversarialnetworkformrictregistration AT tiejunyang dtrgananunsupervisedbidirectionaltranslationgenerativeadversarialnetworkformrictregistration AT xiangzhao dtrgananunsupervisedbidirectionaltranslationgenerativeadversarialnetworkformrictregistration AT xinzhang dtrgananunsupervisedbidirectionaltranslationgenerativeadversarialnetworkformrictregistration AT yanghuiyan dtrgananunsupervisedbidirectionaltranslationgenerativeadversarialnetworkformrictregistration AT chunxiajiao dtrgananunsupervisedbidirectionaltranslationgenerativeadversarialnetworkformrictregistration |