Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction
Summary: Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve image quality from sparse-view data, a multi-domain integrative Swin transformer network (MIST-net) was developed and is reported in this article. First, MIST-net incorporated la...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-06-01
|
| Series: | Patterns |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666389922000836 |
| _version_ | 1828335644300017664 |
|---|---|
| author | Jiayi Pan Heye Zhang Weifei Wu Zhifan Gao Weiwen Wu |
| author_facet | Jiayi Pan Heye Zhang Weifei Wu Zhifan Gao Weiwen Wu |
| author_sort | Jiayi Pan |
| collection | DOAJ |
| description | Summary: Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve image quality from sparse-view data, a multi-domain integrative Swin transformer network (MIST-net) was developed and is reported in this article. First, MIST-net incorporated lavish domain features from data, residual data, image, and residual image using flexible network architectures, where a residual data and residual image sub-network was considered as a data consistency module to eliminate interpolation and reconstruction errors. Second, a trainable edge enhancement filter was incorporated to detect and protect image edges. Third, a high-quality reconstruction Swin transformer (i.e., Recformer) was designed to capture image global features. The experimental results on numerical and real cardiac clinical datasets with 48 views demonstrated that our proposed MIST-net provided better image quality with more small features and sharp edges than other competitors. The bigger picture: Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve reconstructed image quality from sparse-view data, we develop a multi-domain integrative Swin transformer (MIST) network in this study. The proposed MIST-net incorporates lavish domain features from data, residual data, image, and residual image using flexible network architectures, which help deeply mine the data and image features. To detect image features and protect image edges, the trainable edge enhancement filter is further incorporated to the network for improving encode-decode ability. A high-quality reconstruction transformer was designed to improve the ability of global feature extraction. Our results from both simulation and real cardiac data demonstrated the great potential of MIST. |
| first_indexed | 2024-04-13T21:50:15Z |
| format | Article |
| id | doaj.art-9e5d1cd0392b4708a9895d6c8971928b |
| institution | Directory Open Access Journal |
| issn | 2666-3899 |
| language | English |
| last_indexed | 2024-04-13T21:50:15Z |
| publishDate | 2022-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Patterns |
| spelling | doaj.art-9e5d1cd0392b4708a9895d6c8971928b2022-12-22T02:28:27ZengElsevierPatterns2666-38992022-06-0136100498Multi-domain integrative Swin transformer network for sparse-view tomographic reconstructionJiayi Pan0Heye Zhang1Weifei Wu2Zhifan Gao3Weiwen Wu4School of Biomedical Engineering, Sun Yat-sen University, Shenzhen, Guangdong, ChinaSchool of Biomedical Engineering, Sun Yat-sen University, Shenzhen, Guangdong, ChinaDepartment of Orthopedics, The People’s Hospital of China Three Gorges University, The First People’s Hospital of Yichang, Yichang, Hubei, ChinaSchool of Biomedical Engineering, Sun Yat-sen University, Shenzhen, Guangdong, ChinaSchool of Biomedical Engineering, Sun Yat-sen University, Shenzhen, Guangdong, China; Corresponding authorSummary: Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve image quality from sparse-view data, a multi-domain integrative Swin transformer network (MIST-net) was developed and is reported in this article. First, MIST-net incorporated lavish domain features from data, residual data, image, and residual image using flexible network architectures, where a residual data and residual image sub-network was considered as a data consistency module to eliminate interpolation and reconstruction errors. Second, a trainable edge enhancement filter was incorporated to detect and protect image edges. Third, a high-quality reconstruction Swin transformer (i.e., Recformer) was designed to capture image global features. The experimental results on numerical and real cardiac clinical datasets with 48 views demonstrated that our proposed MIST-net provided better image quality with more small features and sharp edges than other competitors. The bigger picture: Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve reconstructed image quality from sparse-view data, we develop a multi-domain integrative Swin transformer (MIST) network in this study. The proposed MIST-net incorporates lavish domain features from data, residual data, image, and residual image using flexible network architectures, which help deeply mine the data and image features. To detect image features and protect image edges, the trainable edge enhancement filter is further incorporated to the network for improving encode-decode ability. A high-quality reconstruction transformer was designed to improve the ability of global feature extraction. Our results from both simulation and real cardiac data demonstrated the great potential of MIST.http://www.sciencedirect.com/science/article/pii/S2666389922000836DSML 3: Development/pre-production: Data science output has been rolled out/validated across multiple domains/problems |
| spellingShingle | Jiayi Pan Heye Zhang Weifei Wu Zhifan Gao Weiwen Wu Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction Patterns DSML 3: Development/pre-production: Data science output has been rolled out/validated across multiple domains/problems |
| title | Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction |
| title_full | Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction |
| title_fullStr | Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction |
| title_full_unstemmed | Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction |
| title_short | Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction |
| title_sort | multi domain integrative swin transformer network for sparse view tomographic reconstruction |
| topic | DSML 3: Development/pre-production: Data science output has been rolled out/validated across multiple domains/problems |
| url | http://www.sciencedirect.com/science/article/pii/S2666389922000836 |
| work_keys_str_mv | AT jiayipan multidomainintegrativeswintransformernetworkforsparseviewtomographicreconstruction AT heyezhang multidomainintegrativeswintransformernetworkforsparseviewtomographicreconstruction AT weifeiwu multidomainintegrativeswintransformernetworkforsparseviewtomographicreconstruction AT zhifangao multidomainintegrativeswintransformernetworkforsparseviewtomographicreconstruction AT weiwenwu multidomainintegrativeswintransformernetworkforsparseviewtomographicreconstruction |