Ultrahigh-resolution Functional MRI of the Human Brain at 10.5 Tesla

MRI is the most advanced and the most used tool in our armamentarium to study the human brain. However, highest possible resolution achievable with existing methods falls short of goals set for the next generation Human Connectome Project (HCP) or the BRAIN Initiative that calls for the capability to study the organizing principles of the human cortex at the mesoscopic or sub-mesoscopic scale. Our long-term goal is to advance ultrahigh field (UHF) MRI through innovative technical developments applied at the highest possible magnetic field available for human studies. The overall objective of this study is to develop a new suite of MRI tools that, when synergistically combined with the unique UHF of 10.5 Tesla (10.5T) and optimized instrumentation (RF coils and gradients), will enable us to acquire high-quality, whole- and partial-brain functional MRI (fMRI) at unparalleled spatiotemporal resolutions, ushering in the next generation HCP and approaching the ambitious resolution target (i.e., 0.01-µL voxel volumes) set in the strategic plan developed by the second BRAIN Initiative Working Group. The overall objective will be accomplished by pursuing four specific aims: Under aim 1, we will develop and optimize new high-channel-count transmit and receive RF arrays to leverage SNR gains available at 10.5T while enabling RF parallel transmission (pTx) for flip angle homogenization and SAR reduction. For aim 2, we will devise two new pTx pulse design frameworks that incorporate RF-power-related constraints for water- selective excitation tailored for whole-brain scan without the need for additional fat saturation, and a fat-selective saturation and a water-selective excitation, suitable for partial-brain scan of a single imaging slab. Under aim 3, we will develop i) a comprehensive pTx-enabled 3D GRE EPI sequence for rapid high-quality high-resolution whole-brain and partial-brain scans, all with joint motion and field correction, ii) a computational toolkit enabling real-time motion detection, background field measurement, and GIRF-based field dynamics determination using NMR field probes, and iii) a tailored image reconstruction for producing motion- and field-corrected images based on field dynamics and motion detection. The results will be compared to those obtained with our long-standing and extensively used 2D simultaneous multi-slice EPI sequence. For aim 4, we will develop and demonstrate brain fMRI at spatiotemporal resolutions that are much higher than what is possible with existing methods. The research proposed in this application is significant because it is expected to have a broad positive impact on various fields including ultrahigh-field MRI, RF hardware, high-resolution fMRI, parallel transmission, data acquisition, image reconstruction, and neuroscience. The proposed research is innovative because it represents a new and substantive departure from the existing approaches by shifting focus to developing a synergy of various technical innovations that can help realize the full potential of 10.5T, enabling ultrahigh-resolution fMRI and increasing our ability to study brain function at the mesoscopic scale.