Direct neuronal current detection by means of Low field Magnetic Resonance

   The detection of neuronal currents (NCs) is a tool to monitor the flow of neuronal information. The interpretation of existing methods of NC detection like electro- (EEG)  and magnetoencephalography (MEG) is limited by the ambiguity of the inverse problem, while imaging methods like functional magnetic resonance imaging (fMRI) monitor secondary effects such as the blood oxygenation level dependency (BOLD). To overcome these limitations research within the department of  Biosignals is carried out which pursues the use of low-field magnetic resonance (LFMR) techniques as an alternative approach to detect directly NCs non-invasively.

LFMR by SQUIDs utilizes magnetic fields around 1 microTesla. Due to the scaling of absolute field inhomogeneities of coil systems with the applied magnetic field strength, the resulting inhomogeneities at low-fields do not limit the spectral resolution. Furthermore, susceptibility effects like the BOLD effect become negligible.

Two different methods to detect NCs at low magnetic fields have been proposed. The first approach, a resonant mechanism, aims at observing the influence of the local neuronal induced magnetic fields on the spin dynamics when the Larmor frequencies of the cerebral ¹H protons match the spectral content of local cerebral electrical activities. As this extents only up to ~ kHz this mechanism is exclusively accessible to LFMR.

The second approach, a DC mechanism, is based on the observation of local frequency and NMR lineshape changes induced by long-lasting NCs. These long-lasting NCs are active during the complete MR cycle and their locally evoked neuromagnetic field will superimpose the applied field. The spatio-temporal structure strongly depends on the distribution of the underlying NCs and leads to different local effects on the lineshape of the MR signal.

Based on these considerations and funded by the Federal Ministry for Education and Research within the project „Bernstein Focus Neurotechnology“, demonstrator experiments are in progress which utilize dipole phantoms to mimic neuronal activity. As a first step we verified the principal working mechanisms in a dipole phantom of both the AC - (see Figure 1) and the DC mechanism (see Figure 2) and obtained a current dipole resolution of some hundred nAm for both effects for our current NMR set up. In order to detect NCs which have current dipole moments of some ten nAm we are presently working on the improvement of the signal to noise ratio which should enable the direct detection of neuronal activity in the human brain. The successful demonstration might form the basis of a new functional neuroimaging modality.

Figure 1: Left: Saline dipole phantom. Right: Demonstration of AC mechanism: NMR amplitude for increasing dipole strength for constant time of phantom operation. Driving the phantom at the Larmor frequency leads to resonant spin absorption. The resulting precessing magnetisation is detected by a SQUID. Minimum detectable current dipole ~ 300 nAm.

Figure 2: Demonstration of DC mechanism. Left: Original and differenced spectra showing that the influence on the NMR line due to the phantom operation is clearly visible in the latter. Right: Differenced spectra for varying current dipole strength with a minimum detectable dipole of ~ 300 nAm.

[1] M. Burghoff et al., “SQUID system for MEG and Low Filed Magnetic Resonance”, Metrol. Meas. Syst., Vol. XVI, 371-375, (2009)

[2] M. Burghoff et al., „On the feasibility of neurocurrent imaging by low field NMR“, in preparation

Contact: