Fast Functional MRI with Sparse Sampling and Model-Based Reconstruction Functional brain imaging using MRI (functional MRI or fMRI) has grown rapidly over the past 25 years and is widely used for basic cognitive neuroscience research and for presurgical planning. It is increasingly being used for developing biomarkers for neurological and psychiatric disorders and for population based studies of, for example, normal and abnormal development and aging. There have also been developments in imaging hardware and methods as well as processing methods to correct for artifacts and analyze functional activity. The overarching goal of this project is to develop a novel ultra-fast whole-brain fMRI acquisition approach that expands the spatiotemporal resolution envelope by roughly 3-fold. For example, new methods will allow 2mm isotropic resolution image with 250ms temporal resolution or 1.5mm isotropic resolution images with 500ms temporal resolution. Current state-of-the-art acquisition approaches for fMRI use the simultaneous multislice (SMS, and also known as multiband) method; these single-shot acquisitions use parallel imaging concepts and array coils to provide acceleration in the slice direction and possibly, the in-plane direction as well. Our approach is fundamentally different and uniquely powerful because: 1) it uses parallel imaging concepts for the slice and in-plane directions similar to multiband methods, while 2) also exploiting the temporal dimension that has a substantial data redundancy, and 3) incorporating novel image reconstruction methods based on low- rank (LR) spatiotemporal representations and ?sparse? sampling patterns that extend farther out in k-space to improve spatial resolution. Together, these methods promise to enable new faster and more robust fMRI acquisition technology than is currently possible, while also improving spatial resolution. The project has three main aims: (1) Develop new low-rank and sparse (L+S) acquisition and reconstruction methods that model temporal basis functions using multi-coil array data, and account for magnetic field inhomogeneity; (2) Develop and evaluate methods to address several well-recognized issues associated with fMRI acquisition, notably physiological noise, head motion, and susceptibility-induced signal losses; and (3) Evaluate the low-rank and sparse acquisition approach and compare to state-of-the-art SMS (multiband) acquisition methods for task and resting state fMRI. The proposed technology will greatly improve spatiotemporal resolution for a given set of hardware (gradient and RF coils). Faster fMRI will allow improved physiological noise correction, improved statistical power and sensitivity for network analysis, and discovery of temporally ordered network processes. Higher spatial resolution will lead to less partial volume and susceptibility artifacts, improved surface-based analyses, and potentially layer-specific BOLD dynamics. These methods also may lead to fMRI that is more robust to head motion making it more useful for patient studies and studies of language.

Public Health Relevance

Functional magnetic resonance imaging (fMRI) is used to measure brain function, and has revolutionized our understanding of cognitive processes during the last 20 years and has also been used as a tool for presurgical mapping of language and other sensitive brain regions. More recently, fMRI is being used as a biomarker for the progression of neurologic and psychiatric diseases, for example, in Alzheimer's disease and multiple sclerosis. In this project we will develop faster fMRI techniques that will improve the robustness and sensitivity of fMRI and broaden the scope of applications for fMRI.

National Institute of Health (NIH)
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Research Project (R01)
Project #
Application #
Study Section
Biomedical Imaging Technology Study Section (BMIT)
Program Officer
Liu, Guoying
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Michigan Ann Arbor
Biomedical Engineering
Biomed Engr/Col Engr/Engr Sta
Ann Arbor
United States
Zip Code
Olafsson, Valur T; Noll, Douglas C; Fessler, Jeffrey A (2018) Fast Spatial Resolution Analysis of Quadratic Penalized Least-Squares Image Reconstruction With Separate Real and Imaginary Roughness Penalty: Application to fMRI. IEEE Trans Med Imaging 37:604-614
Ravishankar, Saiprasad; Moore, Brian E; Nadakuditi, Raj Rao et al. (2017) Low-Rank and Adaptive Sparse Signal (LASSI) Models for Highly Accelerated Dynamic Imaging. IEEE Trans Med Imaging 36:1116-1128