Skip to main content
. Author manuscript; available in PMC: 2021 Apr 1.
Published in final edited form as: IEEE Trans Med Imaging. 2019 Oct 9;39(4):1268–1277. doi: 10.1109/TMI.2019.2946501

Fig. 4.

Fig. 4.

The proposed hybrid MoDL-MUSSELS architecture resulting from the alternating scheme shown in (23)–(25). Here the Dk and DI blocks represent the k-space and the image-space denoising networks, respectively. The Dk and DI networks have identical structures as shown in Fig 3(a). The learnable convolution weights are differnt for networks Dk and DI but remain constant across iterations.