An untrained deep learning method with model-based regularization for reconstructing dynamic MR images from retrospectively accelerated data

Slavkova, Kalina P.; DiCarlo, Julie; Viraj, Wadhwa,; Yankeelov, Thomas E.; Tamir, Jonathan I.

doi:10.58530/2022/0842

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address the need for a physics-informed untrained method with a reliable stopping condition for high-resolution dynamic MRI data amendable to quantitative analysis, we previously applied the CD 31 method with an SPGR-based regularization term (CD + r) 40 as a proof-of-principle for reconstructing multi-dimensional images from simulated accelerated VFA data. Here, we formalize the CD + r method and apply it to retrospectively accelerated 3D VFA data collected from three healthy subjects.…”

Section: Introductionmentioning

confidence: 99%

An untrained deep learning method for reconstructing dynamic MR images from accelerated model‐based data

Slavkova

DiCarlo

Viraj³

et al. 2022

Magnetic Resonance in Med

View full text Add to dashboard Cite

Purpose To implement physics‐based regularization as a stopping condition in tuning an untrained deep neural network for reconstructing MR images from accelerated data. Methods The ConvDecoder (CD) neural network was trained with a physics‐based regularization term incorporating the spoiled gradient echo equation that describes variable‐flip angle data. Fully‐sampled variable‐flip angle k‐space data were retrospectively accelerated by factors of R = {8, 12, 18, 36} and reconstructed with CD, CD with the proposed regularization (CD + r), locally low‐rank (LR) reconstruction, and compressed sensing with L1‐wavelet regularization (L1). Final images from CD + r training were evaluated at the “argmin” of the regularization loss; whereas the CD, LR, and L1 reconstructions were chosen optimally based on ground truth data. The performance measures used were the normalized RMS error, the concordance correlation coefficient, and the structural similarity index. Results The CD + r reconstructions, chosen using the stopping condition, yielded structural similarity indexs that were similar to the CD (p = 0.47) and LR structural similarity indexs (p = 0.95) across R and that were significantly higher than the L1 structural similarity indexs (p = 0.04). The concordance correlation coefficient values for the CD + r T1 maps across all R and subjects were greater than those corresponding to the L1 (p = 0.15) and LR (p = 0.13) T1 maps, respectively. For R ≥ 12 (≤4.2 min scan time), L1 and LR T1 maps exhibit a loss of spatially refined details compared to CD + r. Conclusion The use of an untrained neural network together with a physics‐based regularization loss shows promise as a measure for determining the optimal stopping point in training without relying on fully‐sampled ground truth data.

show abstract

Section: Introductionmentioning

confidence: 99%