Time-Dependent Deep Image Prior for Dynamic MRI

• Published in: IEEE transactions on medical imaging Vol. 40; no. 12; pp. 3337 - 3348
• Main Authors: Yoo, Jaejun, Jin, Kyong Hwan, Gupta, Harshit, Yerly, Jerome, Stuber, Matthias, Unser, Michael
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accelerated MRI; Algorithms; Artificial neural networks; Data acquisition; Deep learning; Electronics packaging; Fetuses; Heart; Heuristic algorithms; Image reconstruction; Imaging; Machine learning; Magnetic resonance imaging; Manifolds; Neural networks; Sequences; Spatial discrimination; Spatial resolution; Spokes; Temporal variations; Unsupervised learning
• ISSN: 0278-0062; 1558-254X
• DOI: 10.1109/TMI.2021.3084288

We propose a novel unsupervised deep-learning-based algorithm for dynamic magnetic resonance imaging (MRI) reconstruction. Dynamic MRI requires rapid data acquisition for the study of moving organs such as the heart. We introduce a generalized version of the deep-image-prior approach, which optimizes the weights of a reconstruction network to fit a sequence of sparsely acquired dynamic MRI measurements. Our method needs neither prior training nor additional data. In particular, for cardiac images, it does not require the marking of heartbeats or the reordering of spokes. The key ingredients of our method are threefold: 1) a fixed low-dimensional manifold that encodes the temporal variations of images; 2) a network that maps the manifold into a more expressive latent space; and 3) a convolutional neural network that generates a dynamic series of MRI images from the latent variables and that favors their consistency with the measurements in <inline-formula> <tex-math notation="LaTeX">{k} </tex-math></inline-formula>-space. Our method outperforms the state-of-the-art methods quantitatively and qualitatively in both retrospective and real fetal cardiac datasets. To the best of our knowledge, this is the first unsupervised deep-learning-based method that can reconstruct the continuous variation of dynamic MRI sequences with high spatial resolution.