Time-Dependent Deep Image Prior for Dynamic MRI
| Autor: | Harshit Gupta, Michael Unser, Matthias Stuber, Jaejun Yoo, Kyong Hwan Jin, Jérôme Yerly |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning reconstruction Computer science Computer Vision and Pattern Recognition (cs.CV) Physics::Medical Physics Computer Science - Computer Vision and Pattern Recognition Latent variable cardiac cine mri unsupervised learning low-rank Convolutional neural network Machine Learning (cs.LG) 030218 nuclear medicine & medical imaging 03 medical and health sciences Consistency (database systems) 0302 clinical medicine Data acquisition framework FOS: Electrical engineering electronic engineering information engineering Image Processing Computer-Assisted medicine accelerated mri grappa Electrical and Electronic Engineering Retrospective Studies combination Sequence k-t sense Radiological and Ultrasound Technology medicine.diagnostic_test business.industry Image and Video Processing (eess.IV) manifold recovery Magnetic resonance imaging Pattern recognition Electrical Engineering and Systems Science - Image and Video Processing Magnetic Resonance Imaging Manifold Computer Science Applications regularization network Dynamic contrast-enhanced MRI Neural Networks Computer Artificial intelligence business Algorithms Software |
| Zdroj: | IEEE Transactions on Medical Imaging. 40:3337-3348 |
| ISSN: | 1558-254X 0278-0062 |
| Popis: | 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. Existing reconstruction methods suffer from restrictions either in the model design or in the absence of ground-truth data, resulting in low image quality. We introduce a generalized version of the deep-image-prior approach, which optimizes the network weights 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 k-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. 11 pages, 6 figures. First Author has been changed |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|