FlowMoDL robustly converges and strictly outperforms all competitors across all acceleration factors in magnitude SSIM, nRMSE, relative velocity error, and angular error, successfully recovering sharp structural details and temporally coherent velocity fields.
Abstract
We present FlowMoDL, an unrolled neural network for highly accelerated 4D flow MRI reconstruction that directly optimizes for both anatomical magnitude and phase-derived velocity accuracy. Building on the MoDL framework, FlowMoDL alternates a learned (3+1)D spatiotemporal denoiser with conjugate-gradient data-consistency updates based on the SENSE forward model. A novel dual-pathway conditioning scheme adapts the denoiser features and data-consistency weighting, enabling a single model to handle varying acceleration factors ($10\times$ to $50\times$). To ensure physiological accuracy, the network is trained using a deep-supervision composite loss that explicitly penalizes velocity magnitude and angular errors, stabilized by a curriculum schedule. We evaluate FlowMoDL on the multi-center CMRx4DFlow dataset against classical and deep-learning baselines (CG-SENSE, MoDL, FlowVN, and FlowMRI-Net). A key advantage of FlowMoDL is its superior gradient step efficiency. When evaluated under an equivalent, limited budget of gradient steps, competing flow-specific networks degrade significantly. In contrast, FlowMoDL robustly converges and strictly outperforms all competitors across all acceleration factors in magnitude SSIM, nRMSE, relative velocity error, and angular error, successfully recovering sharp structural details and temporally coherent velocity fields.
This work incorporates a joint supervised and self-supervised training regime and utilize both magnitude and velocity data during supervision, and adds multiple acquisition-robust and conditioning strategies based on the acceleration factors.
Purpose To determine whether structural MRI information can improve reconstruction of highly accelerated 3D snapshot chemical exchange saturation transfer (CEST) MRI. Methods Fully sampled brain CEST data were retrospectively undersampled at acceleration factors (AFs) of 4, 6, 8, 10, and 12. We compared reconstruction...
E. Mensah, Abrar Faiyaz, G. Schifitto et al.· bioRxiv· 0 citations
Objective. The primary objective of this research is to address the challenges of limited-angle computed tomography (CT) reconstruction, which arises when projection data is acquired over an angular range smaller than required for theoretically exact reconstruction. This ill-posed problem leads to serious artifacts tha...
Gen-Wei Ma, Di-Meng Xia, Yu-Tong He et al.· Physics in Medicine and Biol...· 0 citations
DiffINR shows high robustness to data distribution shifts at test time, including dataset changes, limited-angle setting variations, and different MRI sampling sparsity levels, and achieves fast and accurate deformation-driven volumetric image estimation.
Xiao-Xue Qian, Hua-Chieh Shao, You Zhang· Physics in Medicine and Biol...· 0 citations
The results show that exploiting paired anatomical correspondence enables diffusion-level LDCT reconstruction with a single step reconstruction, and across all REFM variants, one-step inference consistently provides the highest reconstruction quality.
T. D. Sotta, J. M. Saavedra, Violeta Chang et al.· medRxiv· 0 citations
This work reports on subject-adaptive acquisition from measured energy, combining self-supervision with learned acquisition, and amortising the data-consistency weight in generative priors reconstruct accelerated 3D MRI well but pay heavily at deployment.
Rui-Bo Wang· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.