A complete understanding of both normal brain function and neurological disorders / mental illness will require the deciphering of the complex brain networks that underlie behavior and cognition, at both whole-brain and microscopic scales. The difficulty of structurally and functionally mapping these brain networks in living human subjects with sufficient sensitivity and resolution to understand normal function and detect pathological change represents a fundamental challenge. Recognizing these challenges, the NIH and other funding agencies have supported new initiatives in brain mapping for decades, starting with the Decade of the Brain, followed by the Decade after the Decade of the Brain, the New Century of the Brain, the Human Connectome Project, and most recently the BRAIN Initiative. Most of these brain mapping initiatives have featured MRI, the premier tool for studying the intact living human brain, non-invasively, at high resolution, and with high sensitivity to many subtle physiological and pathological processes. For example, the NIH-funded Human Connectome Project (HCP) was launched in 2009 to comprehensively map brain circuitry in 1,200 healthy subjects using MRI, and has already had a large impact on the neuroscience field. The HCP uses diffusion MRI (dMRI) and BOLD-based functional MRI (fMRI) to derive whole-brain structural and functional connectivity maps for individual subjects at 1.25-2mm resolution (2-8L voxels). Such coarse resolution results in the spatial blurring of single-voxel responses over 105-106 neurons. The NIH BRAIN Initiative calls for disruptive new approaches to resolving brain circuit connections and function at dramatically higher spatiotemporal and microstructure resolution. This 2-year proof-of-concept BRAIN R01 will focus on producing a complete and validated design for a next-generation, compact, low-cost, high-performance ultra-high-field (UHF) MRI system, capable of resolving neural connections and circuitry at the scale of 104 neurons (~0.2L), throughout the entire living human brain non-invasively. This would address a specific target identified in BRAIN 2025, which calls for significant developments in MR methodology targeting whole brain studies with voxel volumes of 0.3-0.4L in the short term, and 0.1L or better in the long term. Our specific goals in this project are to re- engineer all front-end hardware components of the UHF MRI system, thereby resolving the technological and physics challenges that have been severely holding back this human brain imaging modality. We will develop several disruptive technologies, designed synergistically to create a dramatically lower cost but higher performing UHF MRI system: 1) ultra-compact UHF magnet technology; 2) ultra-high-performance gradient hardware; and 3) innovative shim and RF array technologies designed to fully correct both main field and RF inhomogeneity problems. The practical impact of our work will be to enable greatly improved human brain mapping using dMRI and fMRI, while simultaneously solving the major technological and cost limitations of UHF MRI.
Our long-term goal is to develop MRI methods capable of generating maps of neural connections in the brain at microscopic resolution, over the entire living human brain. In this proof-of-concept BRAIN R01 project, we propose to re-engineer all front-end hardware components of the ultra-high-field MRI system, thereby resolving the technological and physics challenges that have been preventing this human brain imaging modality from reaching this long-term goal. In so doing, we will show the feasibility of dramatically improved spatiotemporal resolution and sensitivity compared to existing MRI. We will also uniquely address and resolve the high cost of UHF MRI, demonstrating how this technology can be made available to a far larger community.
| Winkler, Simone A; Schmitt, Franz; Landes, Hermann et al. (2018) Gradient and shim technologies for ultra high field MRI. Neuroimage 168:59-70 |
