Contact

  • Academic
    linym@stanford.edu
    University - Student Department: Computer Science Position: Graduate
    University - Student Department: Electrical Engineering Position: Graduate
    University - Student Department: Management Science and Engineering Position: Graduate
    University - Student Department: Electrical Engineering Position: Graduate

All Publications

  • Fast Reconstruction of Motion-Corrupted Data With Mobile-GRAPPA: Motion and δ B 0 Inhomogeneity Correction Leveraging Efficient GRAPPA. Magnetic resonance in medicine Lin, Y., Wang, N., Abraham, D., Polak, D., Cao, X., Nurdinova, A., Cauley, S., Setsompop, K. 2026

    Abstract

    To develop an accurate and computationally efficient motion-corrected MRI reconstruction framework that incorporates hundreds to thousands of motion and δ B 0 estimates from high-temporal-resolution tracking.We propose Mobile-GRAPPA, a k-space preprocessing approach that uses MLP-parameterized local GRAPPA operators to jointly correct trajectory perturbations, coil reweighting, and δ B 0 -induced phase changes before standard downstream reconstruction. Reconstruction accuracy, noise propagation, spatial resolution, and runtime were evaluated using 3D MPRAGE, multi-echo 3D GRE, and 3D EPTI.Experiments with discrete motion states demonstrated that Mobile-GRAPPA followed by standard SENSE achieved image quality comparable to Aligned-SENSE. In 3D GRE with 1620 tracked states and 3D EPTI with 544 tracked states, Mobile-GRAPPA incorporated all state estimates with minimal motion-correction overhead. Total reconstruction times were approximately 15 s for GRE and 20 min for EPTI, whereas full-state Aligned-SENSE was computationally prohibitive (approximately 10 h for GRE and multiple days for EPTI). Pseudo-replica and PSF analyses showed limited additional noise amplification and negligible spatial-resolution loss.Mobile-GRAPPA enables dense motion and δ B 0 information to be incorporated with minimal motion-correction overhead while preserving standard SENSE, subspace, and other downstream reconstruction pipelines.

    View details for DOI 10.1002/mrm.70575

    View details for PubMedID 42693979

  • Neural Space-Time Modeling for Motion-Corrected MR Reconstruction Nurdinova, A., Huang, W., Abraham, D., Bae, J., Lin, Y., Setsompop, K., Hargreaves, B. edited by Felsner, L., Kustner, T., Maier, A., Qin, C., Ahmadi, S. A., Kazi, A., Hu SPRINGER INTERNATIONAL PUBLISHING AG. 2026: 118-128

    View details for DOI 10.1007/978-3-032-06103-4_12

    View details for Web of Science ID 001685837600012

  • In Vivo Meso-Scale Whole-Brain Quantitative Imaging With Tailored MRF on the NexGen 7T Scanner. Magnetic resonance in medicine Cao, X., Beckett, A., Liao, C., Walker, E., Zhu, Z., Qian, Y., Gao, M., Wang, N., Lin, Y., Gong, L., McCready, M. A., Wang, Z., Li, Z., Vu, A., Ma, S., Ramos-Llordén, G., Tian, Q., Kerr, A., Yang, Y., Feinberg, D. A., Setsompop, K. 2025

    Abstract

    To push the speed and resolution limit of in vivo quantitative imaging and enable estimation of quantitative tissue parameters of subtle brain structures that were previously difficult to assess.This study implemented an efficient quantitative imaging approach, 3D-SPI MRF, on the NexGen 7T scanner equipped with a high-performance head-only gradient and 96-channel receiver array. To address challenges associated with performing rapid mesoscale MRF on this system, acquisition and reconstruction mitigation methods were developed and incorporated into the MRF framework, including: (i) flip-angle-aware dictionary fitting to account for both B1 + inhomogeneity and voxel-specific RF frequency response, (ii) gradient imperfection corrections via Skope measurements that incorporates a new per-TR trajectory rewinder compensation, (iii) incorporation of rapid B1 + and B0 mappings into the MRF sequence, and (iv) high-temporal motion navigation.Whole-brain T1 and T2 maps were obtained at 560-μm isotropic resolution within 4 min, where ablation studies demonstrated the necessity of the various mitigation methods implemented in removing bias and artifacts. For comparison, MRF data were acquired using current state-of-the-art method but limited to typical whole-body gradient specifications to demonstrate that the proposed developments resulted in ∼3× shorter scan time while producing more accurate parameter maps. Data were also acquired at ∼3.8× smaller voxel size, 360-μm isotropic, using the developed technique, to achieve mesoscale multi-parameter quantitative mapping in vivo.Tailored 3D MRF acquisition and reconstruction were developed to enable fast and accurate T1 and T2 mapping across the whole-brain at mesoscale resolution on the NexGen 7T scanner.

    View details for DOI 10.1002/mrm.70234

    View details for PubMedID 41472402