TY - JOUR
T1 - MR image reconstruction using densely connected residual convolutional networks
AU - Aghabiglou, Amir
AU - Eksioglu, Ender M.
N1 - Publisher Copyright:
© 2021 Elsevier Ltd
PY - 2021/12
Y1 - 2021/12
N2 - MR image reconstruction techniques based on deep learning have shown their capacity for reducing MRI acquisition time and performance improvement compared to analytical methods. Despite the many challenges in training these rather large networks, novel methodologies have enhanced the capability for having clinical-grade MR image reconstruction in real-time. In recent literature, novel developments have facilitated the utilization of deep networks in various image processing inverse problems. In particular, it has been reported multiple times that the performance of deep networks can be improved by using short connections between layers. In this study, we introduce a novel MRI reconstruction method that utilizes such short connections. The dense connections are used inside densely connected residual blocks. Inside these blocks, the feature maps are concatenated to the subsequent layers. In this way, the extracted information is propagated until the last stage of the block. We have evaluated this densely connected residual block's efficiency in MRI reconstruction settings, by augmenting different types of effective deep network models with these blocks in novel structures. The quantitative and qualitative results indicate that this original introduction of the densely connected blocks to the MR image reconstruction problem improves the reconstruction performance significantly.
AB - MR image reconstruction techniques based on deep learning have shown their capacity for reducing MRI acquisition time and performance improvement compared to analytical methods. Despite the many challenges in training these rather large networks, novel methodologies have enhanced the capability for having clinical-grade MR image reconstruction in real-time. In recent literature, novel developments have facilitated the utilization of deep networks in various image processing inverse problems. In particular, it has been reported multiple times that the performance of deep networks can be improved by using short connections between layers. In this study, we introduce a novel MRI reconstruction method that utilizes such short connections. The dense connections are used inside densely connected residual blocks. Inside these blocks, the feature maps are concatenated to the subsequent layers. In this way, the extracted information is propagated until the last stage of the block. We have evaluated this densely connected residual block's efficiency in MRI reconstruction settings, by augmenting different types of effective deep network models with these blocks in novel structures. The quantitative and qualitative results indicate that this original introduction of the densely connected blocks to the MR image reconstruction problem improves the reconstruction performance significantly.
KW - DCR blocks
KW - Deep learning
KW - Image reconstruction
KW - Magnetic resonance imaging
UR - http://www.scopus.com/inward/record.url?scp=85118891404&partnerID=8YFLogxK
U2 - 10.1016/j.compbiomed.2021.105010
DO - 10.1016/j.compbiomed.2021.105010
M3 - Article
C2 - 34773757
AN - SCOPUS:85118891404
SN - 0010-4825
VL - 139
JO - Computers in Biology and Medicine
JF - Computers in Biology and Medicine
M1 - 105010
ER -