Computational and Phantom-Based Feasibility Study of 3D dcNCI With Ultra-Low-Field MRI
OPEN ACCESS
Loading...
Author / Producer
Date
2021-04
Publication Type
Journal Article
ETH Bibliography
yes
Citations
Altmetric
OPEN ACCESS
Data
Rights / License
Abstract
The possibility to directly and non-invasively localize neuronal activities in the human brain, as for instance by performing neuronal current imaging (NCI) via magnetic resonance imaging (MRI), would be a breakthrough in neuroscience. In order to assess the feasibility of 3-dimensional (3D) NCI, comprehensive computational and physical phantom experiments using low-noise ultra-low-field (ULF) MRI technology were performed using two different source models within spherical phantoms. The source models, consisting of a single dipole and an extended dipole grid, were calibrated enabling the quantitative emulation of a long-lasting neuronal activity by the application of known current waveforms. The dcNCI experiments were also simulated by solving the Bloch equations using the calculated internal magnetic field distributions of the phantoms and idealized MRI fields. The simulations were then validated by physical phantom experiments using a moderate polarization field of 17 mT. A focal activity with an equivalent current dipole of about 150 nAm and a physiologically relevant depth of 35 mm could be resolved with an isotropic voxel size of 25 mm. The simulation tool enabled the optimization of the imaging parameters for sustained neuronal activities in order to predict maximum sensitivity.
Permanent link
Publication status
published
External links
Editor
Book title
Journal / series
Volume
9
Pages / Article No.
647376
Publisher
Frontiers Media
Event
Edition / version
Methods
Software
Geographic location
Date collected
Date created
Subject
Ultra-low-field magnetic resonance imaging; Neuronal current imaging; Current dipole phantom; MEG; Simulation