Microstructure & Connectivity Lab
Publications (Image Processing)
Microstructure & Connectivity Lab
Publications (Image Processing)
Embedded Files
Publications (Image Processing)
Head Motion in Diffusion Magnetic Resonance Imaging: Quantification, Mitigation, and Structural Associations in Large, Cross-Sectional Datasets Across the Lifespan
Head Motion in Diffusion Magnetic Resonance Imaging: Quantification, Mitigation, and Structural Associations in Large, Cross-Sectional Datasets Across the Lifespan
Authors: Kurt G. Schilling, Karthik Ramadass, Viljami Sairanen, Michael E. Kim, Francois Rheault, Nancy Newlin, Tin Nguyen, Laura Barquero, Micah D'archangel, Chenyu Gao, Ema Topolnjak, Nazirah Mohd Khairi, Derek Archer, Lori L. Beason-Held, Susan M. Resnick, Timothy Hohman, Laurie Cutting, Julie Schneider, Lisa L. Barnes, David A. Bennett, Konstantinos Arfanakis, Sophia Vinci-Booher, Marilyn Albert, The BIOCARD Study Team‡, The Alzheimer's Disease Neuroimaging Initiative (ADNI)†, Aging Brain: Vasculature, Ischemia, and Behavior (ABVIB), Daniel Moyer, Bennett A. Landman
Journal: Human Brain Mapping (2025)
Summary: Head motion in diffusion MRI (dMRI) introduces artifacts that bias structural connectivity analyses. This study quantifies motion across 13 cohorts (16,995 sessions, ages 0.1–100 years) and evaluates preprocessing strategies for mitigating motion-induced biases. Findings show that modern pipelines effectively reduce these biases, ensuring reliable dMRI-derived metrics across diverse populations.
Synthesized b0 for diffusion distortion correction (Synb0-DisCo)
Synthesized b0 for diffusion distortion correction (Synb0-DisCo)
Authors: urt G. Schilling, Justin Blaber, Yuankai Huo, Allen Newton, Colin Hansen, Vishwesh Nath, Andrea T. Shafer, Owen Williams, Susan M. Resnick, Baxter Rogers, Adam W. Anderson, Bennett A. Landman
Journal: Magnetic Resonance Imaging (2019)
Summary: Diffusion MRI is commonly distorted by magnetic field inhomogeneities, leading to misalignment with anatomical images. This study introduces Synb0-DisCo, a deep-learning-based method that synthesizes undistorted diffusion images from structural scans, enabling distortion correction without specialized acquisitions. The method significantly improves image alignment and diffusion model accuracy, making advanced correction techniques accessible to a broader range of clinical and research applications.
Why this is a Notable Lab publication: This paper introduces an AI-powered solution to correct distortions in brain scans, allowing researchers and clinicians to improve MRI accuracy even when specialized correction scans are unavailable.
Page updated
Google Sites
Report abuse