Diffusion weighted magnetic resonance imaging (DW-MRI) has opened unprecedented opportunities to simultaneously study local microarchitecture and global structural connectivity. Yet, nearly two decades after the initial presentation of diffusion tensor imaging (DTI), DW-MRI remains plagued by basic theoretical challenges to its interpretation in regions of complex tissues (e.g., crossing fibers). Understanding these complex regions with advanced DW-MRI methods is critical to quantitative assessment of the brain architectural organization, diagnosing connectivity abnormalities, and developing useful biomarkers. Although numerous potential acquisition and analysis techniques have been proposed, to date, there has not been a systematic characterization of the impacts of acquisition design and quality across advanced techniques. The motivating hypothesis of this work is that different advanced DW-MRI methods are appropriate for different imaging contexts given practical image acquisition considerations (e.g., feasible scan time, hardware, propensity for patient motion). Investigators considering studies involving advanced DW-MRI are faced with two critical questions: (1) What is the expected performance (specificity/sensitivity) of possible advanced DW-MRI analyses given a particular imaging scenario? and (2) How can an imaging scenario be optimized to achieve target level of performance given a particular DW-MRI analysis scenario? The overall goal of this project is resolve these important long-standing questions. We will evaluate the motivating hypothesis through a series of three experiments: First, we will perform the most extensive phantom study to date to map the empirical sensitivity and specificity of advanced DW-MRI metrics. This study will generate scan-rescan data of known physical fiber structures. Second, we will perform the most extensive scan-rescan study to date to map the in vivo reproducibility of advanced DW-MRI metrics. This study will generate scan-rescan data across all three major scan vendors with neurologically normal control volunteers who are age and gender stratified. Third, we will collaborate with on-going clinical research studies to generate a large database of scan-rescan validation data. This study will enable mapping of normal inter-subject variance, lead to the construction of a normative database of advanced DW-MRI measures, and enable single-subject inference. Together, these efforts will (1) provide a quantitative basis for biophysical interpretation of biomarkers derived from advanced DW-MRI in the context of noise and sampling strategies and (2) provide a solid theoretical and empirical basis for optimization of protocols within scan-time and hardware constraints. The data, analysis software, and visualization tools will be made freely available to facilitate continued improvement and innovation. These efforts will drive quantitative exploration for biomarkers based on advanced DW-MRI and, eventually, improve patient care.
Advanced diffusion-weighted magnetic resonance imaging (DW-MRI) techniques offer the potential to resolve complex structures with micron-level organization at millimeter resolution. While there is general consensus in the research community that advanced DW-MRI methods contribute meaningful biological information, specifically which technique is optimal or advantageous given practical considerations remains an open problem. This research will improve understanding of how practical imaging concerns impact advanced DW-MRI and provide a quantitative basis for interpreting biomarkers.
| Hainline, Allison E; Nath, Vishwesh; Parvathaneni, Prasanna et al. (2018) Evaluation of inter-site bias and variance in diffusion-weighted MRI. Proc SPIE Int Soc Opt Eng 10574: |
| Lyu, Ilwoo; Kang, Hakmook; Woodward, Neil D et al. (2018) Sulcal Depth-based Cortical Shape Analysis in Normal Healthy Control and Schizophrenia Groups. Proc SPIE Int Soc Opt Eng 10574: |
| Hainline, Allison E; Nath, Vishwesh; Parvathaneni, Prasanna et al. (2018) Empirical single sample quantification of bias and variance in Q-ball imaging. Magn Reson Med 80:1666-1675 |
| Parvathaneni, Prasanna; Lyu, Ilwoo; Huo, Yuankai et al. (2018) Constructing Statistically Unbiased Cortical Surface Templates Using Feature-Space Covariance. Proc SPIE Int Soc Opt Eng 10574: |
| Nath, Vishwesh; Schilling, Kurt G; Hainline, Allison E et al. (2018) SHARD: Spherical Harmonic-based Robust Outlier Detection for HARDI Methods. Proc SPIE Int Soc Opt Eng 10574: |
| Huo, Yuankai; Liu, Jiaqi; Xu, Zhoubing et al. (2018) Robust Multicontrast MRI Spleen Segmentation for Splenomegaly Using Multi-Atlas Segmentation. IEEE Trans Biomed Eng 65:336-343 |
| Rogers, Baxter P; Blaber, Justin; Newton, Allen T et al. (2018) Phantom-based field maps for gradient nonlinearity correction in diffusion imaging. Proc SPIE Int Soc Opt Eng 10573: |
| Schilling, Kurt G; Gao, Yurui; Li, Muwei et al. (2018) Functional tractography of white matter by high angular resolution functional-correlation imaging (HARFI). Magn Reson Med : |
| Parvathaneni, Prasanna; Nath, Vishwesh; Blaber, Justin A et al. (2018) Empirical reproducibility, sensitivity, and optimization of acquisition protocol, for Neurite Orientation Dispersion and Density Imaging using AMICO. Magn Reson Imaging 50:96-109 |
| Damon, Stephen M; Boyd, Brian D; Plassard, Andrew J et al. (2017) DAX - The Next Generation: Towards One Million Processes on Commodity Hardware. Proc SPIE Int Soc Opt Eng 2017: |
Showing the most recent 10 out of 17 publications
