Fidelity imposed network edit (FINE) for solving ill-posed image reconstruction

Neuroimage. 2020 May 1;211:116579. doi: 10.1016/j.neuroimage.2020.116579. Epub 2020 Jan 22.


Deep learning (DL) is increasingly used to solve ill-posed inverse problems in medical imaging, such as reconstruction from noisy and/or incomplete data, as DL offers advantages over conventional methods that rely on explicit image features and hand engineered priors. However, supervised DL-based methods may achieve poor performance when the test data deviates from the training data, for example, when it has pathologies not encountered in the training data. Furthermore, DL-based image reconstructions do not always incorporate the underlying forward physical model, which may improve performance. Therefore, in this work we introduce a novel approach, called fidelity imposed network edit (FINE), which modifies the weights of a pre-trained reconstruction network for each case in the testing dataset. This is achieved by minimizing an unsupervised fidelity loss function that is based on the forward physical model. FINE is applied to two important inverse problems in neuroimaging: quantitative susceptibility mapping (QSM) and under-sampled image reconstruction in MRI. Our experiments demonstrate that FINE can improve reconstruction accuracy.

Keywords: Data fidelity; Deep learning; Inverse problem; Quantitative susceptibility mapping; Under-sampled image reconstruction.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Adult
  • Brain / diagnostic imaging*
  • Cerebral Hemorrhage / diagnostic imaging*
  • Computer Simulation
  • Deep Learning*
  • Humans
  • Image Processing, Computer-Assisted / methods*
  • Image Processing, Computer-Assisted / standards
  • Magnetic Resonance Imaging / methods*
  • Magnetic Resonance Imaging / standards
  • Multiple Sclerosis / diagnostic imaging*
  • Neuroimaging / methods*
  • Neuroimaging / standards