Logo image
Dynamic MRI using model‐based deep learning and SToRM priors: MoDL‐SToRM
Journal article   Peer reviewed

Dynamic MRI using model‐based deep learning and SToRM priors: MoDL‐SToRM

Sampurna Biswas, Hemant K Aggarwal and Mathews Jacob
Magnetic resonance in medicine, Vol.82(1), pp.485-494
07/2019
DOI: 10.1002/mrm.27706
PMCID: PMC7895318
PMID: 30860286
url
https://www.ncbi.nlm.nih.gov/pmc/articles/7895318View
Open Access

Abstract

Purpose To introduce a novel framework to combine deep‐learned priors along with complementary image regularization penalties to reconstruct free breathing & ungated cardiac MRI data from highly undersampled multi‐channel measurements. Methods Image recovery is formulated as an optimization problem, where the cost function is the sum of data consistency term, convolutional neural network (CNN) denoising prior, and SmooThness regularization on manifolds (SToRM) prior that exploits the manifold structure of images in the dataset. An iterative algorithm, which alternates between denoizing of the image data using CNN and SToRM, and conjugate gradients (CG) step that minimizes the data consistency cost is introduced. Unrolling the iterative algorithm yields a deep network, which is trained using exemplar data. Results The experimental results demonstrate that the proposed framework can offer fast recovery of free breathing and ungated cardiac MRI data from less than 8.2s of acquisition time per slice. The reconstructions are comparable in image quality to SToRM reconstructions from 42s of acquisition time, offering a fivefold reduction in scan time. Conclusions The results show the benefit in combining deep learned CNN priors with complementary image regularization penalties. Specifically, this work demonstrates the benefit in combining the CNN prior that exploits local and population generalizable redundancies together with SToRM, which capitalizes on patient‐specific information including cardiac and respiratory patterns. The synergistic combination is facilitated by the proposed framework.
free breathing cardiac MR model‐based learned prior subject specific prior alternating minimization non‐local prior

Details

Metrics

Logo image