A Very Deep Densely Connected Network for Compressed Sensing MRI
Convolutional neural network (CNN) has achieved great success in the compressed sensing-based magnetic resonance imaging (CS-MRI). Latest deep networks for CS-MRI usually consist of a stack of sub-networks, each of which refines the former image prediction to a more accurate one. However, as the sub...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8744318/ |
| _version_ | 1818415911935148032 |
|---|---|
| author | Kun Zeng Yu Yang Guobao Xiao Zhong Chen |
| author_facet | Kun Zeng Yu Yang Guobao Xiao Zhong Chen |
| author_sort | Kun Zeng |
| collection | DOAJ |
| description | Convolutional neural network (CNN) has achieved great success in the compressed sensing-based magnetic resonance imaging (CS-MRI). Latest deep networks for CS-MRI usually consist of a stack of sub-networks, each of which refines the former image prediction to a more accurate one. However, as the sub-network number increases, the information in prior sub-networks has a little influence on subsequent ones, which increases the training difficulties and limits the reconstruction performance of the deep model. In this paper, we propose a novel network, named very deep densely connected network (VDDCN), for CS-MRI. Dense connections are introduced to connect any two sub-networks of VDDCN, so each sub-network can make full use of all former predictions, boosting the reconstruction performance of the whole network. The sub-network of VDDCN is composed of feature extraction and fusion block (FEFB) processing data in the image domain and data consistency (DC) layer enforcing the data fidelity in k-space. Specifically, in FEFB, multi-level features are extracted by the recursive feature extraction and fusion sub-blocks (RFEFSBs) and fused locally to obtain the compact features. The VDDCN is much deeper than the prior deep learning models and able to discover more MR image details. The experimental results demonstrate that our proposed VDDCN outperforms other state-of-the-art CS-MRI methods. |
| first_indexed | 2024-12-14T11:42:31Z |
| format | Article |
| id | doaj.art-f1acddb0579047cf9c50f95ec86daf83 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T11:42:31Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-f1acddb0579047cf9c50f95ec86daf832022-12-21T23:02:46ZengIEEEIEEE Access2169-35362019-01-017854308543910.1109/ACCESS.2019.29246048744318A Very Deep Densely Connected Network for Compressed Sensing MRIKun Zeng0https://orcid.org/0000-0002-6713-2871Yu Yang1https://orcid.org/0000-0002-5514-455XGuobao Xiao2Zhong Chen3Department of Electronic Science, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, School of Electronic Science and Engineering, Xiamen University, Xiamen, ChinaDepartment of Electronic Science, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, School of Electronic Science and Engineering, Xiamen University, Xiamen, ChinaFujian Provincial Key Laboratory of Information Processing and Intelligent Control, College of Computer and Control Engineering, Minjiang University, Fuzhou, ChinaDepartment of Electronic Science, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, School of Electronic Science and Engineering, Xiamen University, Xiamen, ChinaConvolutional neural network (CNN) has achieved great success in the compressed sensing-based magnetic resonance imaging (CS-MRI). Latest deep networks for CS-MRI usually consist of a stack of sub-networks, each of which refines the former image prediction to a more accurate one. However, as the sub-network number increases, the information in prior sub-networks has a little influence on subsequent ones, which increases the training difficulties and limits the reconstruction performance of the deep model. In this paper, we propose a novel network, named very deep densely connected network (VDDCN), for CS-MRI. Dense connections are introduced to connect any two sub-networks of VDDCN, so each sub-network can make full use of all former predictions, boosting the reconstruction performance of the whole network. The sub-network of VDDCN is composed of feature extraction and fusion block (FEFB) processing data in the image domain and data consistency (DC) layer enforcing the data fidelity in k-space. Specifically, in FEFB, multi-level features are extracted by the recursive feature extraction and fusion sub-blocks (RFEFSBs) and fused locally to obtain the compact features. The VDDCN is much deeper than the prior deep learning models and able to discover more MR image details. The experimental results demonstrate that our proposed VDDCN outperforms other state-of-the-art CS-MRI methods.https://ieeexplore.ieee.org/document/8744318/Deep learningcompressed sensingmagnetic resonance imaging (MRI)densely connected network |
| spellingShingle | Kun Zeng Yu Yang Guobao Xiao Zhong Chen A Very Deep Densely Connected Network for Compressed Sensing MRI IEEE Access Deep learning compressed sensing magnetic resonance imaging (MRI) densely connected network |
| title | A Very Deep Densely Connected Network for Compressed Sensing MRI |
| title_full | A Very Deep Densely Connected Network for Compressed Sensing MRI |
| title_fullStr | A Very Deep Densely Connected Network for Compressed Sensing MRI |
| title_full_unstemmed | A Very Deep Densely Connected Network for Compressed Sensing MRI |
| title_short | A Very Deep Densely Connected Network for Compressed Sensing MRI |
| title_sort | very deep densely connected network for compressed sensing mri |
| topic | Deep learning compressed sensing magnetic resonance imaging (MRI) densely connected network |
| url | https://ieeexplore.ieee.org/document/8744318/ |
| work_keys_str_mv | AT kunzeng averydeepdenselyconnectednetworkforcompressedsensingmri AT yuyang averydeepdenselyconnectednetworkforcompressedsensingmri AT guobaoxiao averydeepdenselyconnectednetworkforcompressedsensingmri AT zhongchen averydeepdenselyconnectednetworkforcompressedsensingmri AT kunzeng verydeepdenselyconnectednetworkforcompressedsensingmri AT yuyang verydeepdenselyconnectednetworkforcompressedsensingmri AT guobaoxiao verydeepdenselyconnectednetworkforcompressedsensingmri AT zhongchen verydeepdenselyconnectednetworkforcompressedsensingmri |