PURPOSE: To use a deep neural network (DNN) for solving the optimization problem of water/fat separation and to compare supervised and unsupervised training. METHODS: The current T 2 ∗ -IDEAL algorithm for solving water/fat separation is dependent on initialization. Recently, DNN has been proposed to solve water/fat separation without the need for suitable initialization. However, this approach requires supervised training of DNN using the reference water/fat separation images. Here we propose 2 novel DNN water/fat separation methods: 1) unsupervised training of DNN (UTD) using the physical forward problem as the cost function during training, and 2) no training of DNN using physical cost and backpropagation to directly reconstruct a single dataset. The supervised training of DNN, unsupervised training of DNN, and no training of DNN methods were compared with the reference T 2 ∗ -IDEAL. RESULTS: All DNN methods generated consistent water/fat separation results that agreed well with T 2 ∗ -IDEAL under proper initialization. CONCLUSION: The water/fat separation problem can be solved using unsupervised deep neural networks.
PURPOSE: To use a deep neural network (DNN) for solving the optimization problem of water/fat separation and to compare supervised and unsupervised training. METHODS: The current T 2 ∗ -IDEAL algorithm for solving water/fat separation is dependent on initialization. Recently, DNN has been proposed to solve water/fat separation without the need for suitable initialization. However, this approach requires supervised training of DNN using the reference water/fat separation images. Here we propose 2 novel DNN water/fat separation methods: 1) unsupervised training of DNN (UTD) using the physical forward problem as the cost function during training, and 2) no training of DNN using physical cost and backpropagation to directly reconstruct a single dataset. The supervised training of DNN, unsupervised training of DNN, and no training of DNN methods were compared with the reference T 2 ∗ -IDEAL. RESULTS: All DNN methods generated consistent water/fat separation results that agreed well with T 2 ∗ -IDEAL under proper initialization. CONCLUSION: The water/fat separation problem can be solved using unsupervised deep neural networks.
Authors: Scott B Reeder; Zhifei Wen; Huanzhou Yu; Angel R Pineda; Garry E Gold; Michael Markl; Norbert J Pelc Journal: Magn Reson Med Date: 2004-01 Impact factor: 4.668
Authors: Jinwei Zhang; Zhe Liu; Shun Zhang; Hang Zhang; Pascal Spincemaille; Thanh D Nguyen; Mert R Sabuncu; Yi Wang Journal: Neuroimage Date: 2020-01-22 Impact factor: 6.556
Authors: Takeshi Yokoo; Suraj D Serai; Ali Pirasteh; Mustafa R Bashir; Gavin Hamilton; Diego Hernando; Houchun H Hu; Holger Hetterich; Jens-Peter Kühn; Guido M Kukuk; Rohit Loomba; Michael S Middleton; Nancy A Obuchowski; Ji Soo Song; An Tang; Xinhuai Wu; Scott B Reeder; Claude B Sirlin Journal: Radiology Date: 2017-09-11 Impact factor: 11.105
Authors: Junghun Cho; Jinwei Zhang; Pascal Spincemaille; Hang Zhang; Simon Hubertus; Yan Wen; Ramin Jafari; Shun Zhang; Thanh D Nguyen; Alexey V Dimov; Ajay Gupta; Yi Wang Journal: Magn Reson Med Date: 2021-10-31 Impact factor: 3.737