Search papers, labs, and topics across Lattice.
FlowMoDL is an unrolled neural network designed for highly accelerated 4D flow MRI reconstruction, optimizing for both anatomical accuracy and phase-derived velocity precision. By integrating a learned spatiotemporal denoiser with conjugate-gradient data-consistency updates, FlowMoDL effectively adapts to varying acceleration factors from 10x to 50x through a dual-pathway conditioning scheme. Evaluated against classical and advanced deep-learning methods, FlowMoDL consistently outperforms competitors in key metrics, demonstrating superior efficiency and accuracy in reconstructing sharp structural details and coherent velocity fields.
FlowMoDL achieves unprecedented accuracy in 4D flow MRI reconstruction, outperforming traditional and contemporary models even under extreme acceleration conditions.
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.
