Abstract
To increase granularity in human neuroimaging science, we designed and built a next-generation 7 Tesla magnetic resonance imaging scanner to reach ultra-high resolution by implementing several advances in hardware. To improve spatial encoding and increase the image signal-to-noise ratio, we developed a head-only asymmetric gradient coil (200 mT m−1, 900 T m−1s−1) with an additional third layer of windings. We integrated a 128-channel receiver system with 64- and 96-channel receiver coil arrays to boost signal in the cerebral cortex while reducing g-factor noise to enable higher accelerations. A 16-channel transmit system reduced power deposition and improved image uniformity. The scanner routinely performs functional imaging studies at 0.35–0.45 mm isotropic spatial resolution to reveal cortical layer functional activity, achieves high angular resolution in diffusion imaging and reduces acquisition time for both functional and structural imaging.
Main
Functional magnetic resonance imaging (fMRI) has become the mainstay of human imaging in neuroscience. Typical fMRI studies use an isotropic spatial resolution on the order of 3.0 mm (27 μl voxel volume) to inform many questions about human brain functional organization. However, much higher resolution would enable exploration of neural circuits at the scale of cerebral columns and cortical layers. Human fMRI at about 0.8 mm isotropic resolution (0.5 μl)1,2,3 has been used to study columnar4,5,6,7,8,9 and laminar organization10,11,12,13,14,15,16 (mesoscale circuitry), which has been achieved by limiting the image field of view (FOV) to small targeted regions of cerebral cortex. With human cortex thickness varying between 1.5 and 4.5 mm (refs. 17,18) and with cortical columnar features being 0.6–1.0 mm, higher resolution is needed to adequately sample cortical layers and columns and to minimize partial volume averaging of surrounding white matter and cerebrospinal fluid (CSF). Advancing the resolution to the next level, for example to 0.3–0.6 mm (0.03–0.2 μl)1, would allow the six neuronal layers of the human cerebral cortex to be adequately sampled. Reaching mesoscale spatial resolution is also beneficial in structural and diffusion imaging, which can reveal depth-dependent organization of axonal fiber tracks19,20,21 when performed at higher spatial and angular resolutions.
Toward the goal of reaching higher spatial resolution MRI for neuroscience, we designed and developed a next-generation (NexGen) 7 Tesla (7 T) human brain scanner to perform a wide range of MRI image acquisition techniques used in neuroscience, including functional, diffusion, physiological and structural imaging techniques. Several high-performance system components were built and adapted at the ultra-high field of 7 T, which was critical to achieve higher resolution in brain imaging.
Over the years, improvements to major MRI scanner system components (for example, magnetic field gradient coils22,23,24 and radiofrequency (RF) receiver array coils25) have led to higher signal-to-noise ratio (SNR) and higher resolution imaging or faster image acquisition (acceleration). However, such hardware improvements were typically developed in isolation from one another and at less challenging, lower field strengths (for example, 3 T). To achieve higher resolution, we took advantage of the higher signal afforded by 7 T ultra-high field scanners in which we integrated hardware systems including a head-only magnetic gradient coil, a receiver system, acquisition computer, receiver coils and transmit coils. Critically, these system components were designed, built and optimized concurrently. Our approach of comprehensive system design and integration overcame several physical challenges including space constraints, system cooling, peripheral nerve stimulation (PNS), eddy currents and the interaction of higher forces and torques that are produced by strong magnetic gradients at ultra-high magnetic field. The improvements of the gradient coil combined with the larger channel count receiver arrays enable higher acceleration of data capture and higher SNR. In aggregate, the higher signal26 and greater blood oxygenation level-dependent (BOLD) contrast of 7 T coupled with higher resolution, SNR and faster signal acquisition of the NexGen 7 T scanner enable improved diffusion imaging and mesoscale imaging in vivo.
Results
Head gradient coil
Standard shielded gradient coils include two layers of conductive wiring: the primary inner layer creates linear magnetic fields inside the coil for spatial encoding images while the secondary outer winding layer cancels the external magnetic field to reduce eddy currents in the surrounding superconducting magnet. Our high-performance ‘Impulse’ head-only asymmetric gradient coil incorporates an intermediate third layer of wire winding (Fig. 1) providing additional degrees of freedom needed for simultaneous reduction of PNS27 and optimization of gradient field quality, mechanical resonances and torque.
