Techniques for imaging small impedance changes in the human head due to neuronal depolarisation

Ahadzi, Gershon M. (2006) Techniques for imaging small impedance changes in the human head due to neuronal depolarisation. PhD thesis, Middlesex University. [Thesis]

Download (13MB) | Preview


A new imaging modality is being developed, which may be capable of imaging small impedance changes in the human head due to neuronal depolarization. One way to do this would be by imaging the impedance changes associated with ion channels opening in neuronal membranes in the brain
during activity. The results of previous modelling and experimental studies indicated that impedance changes between 0.6%and 1.7% locally in brain grey matter when recorded at DC. This reduces by a further of 10% if measured at the surface of the head, due to distance and the effect of the resistive skull. In principle, this could be measured using Electrical Impedance Tomography (ElT) but it is close to its threshold of detectability.
With the inherent limitation in the use of electrodes, this work proposed two new schemes. The first is
a magnetic measurement scheme based on recording the magnetic field with Superconducting
Quantum Interference Devices (SQUIDs), used in Magnetoencephalography (MEG) as a result of a
non-invasive injection of current into the head. This scheme assumes that the skull does not attenuate
the magnetic field. The second scheme takes into consideration that the human skull is irregular in
shape, with less and varying conductivity as compared to other head tissues. Therefore, a key issue is to
know through which electrodes current can be injected in order to obtain high percentage changes in surface potential when there is local conductivity change in the head. This model will enable the prediction of the current density distribution at specific regions in the brain with respect to the varying skull and local conductivities.
In the magnetic study, the head was modelled as concentric spheres, and realistic head shapes to mimic
the scalp, skull, Cerebrospinal Auid (CSF) and brain using the Finite Element Method (FEM). An
impedance change of 1 % in a 2cm-radius spherical volume depicting the physiological change in the
brain was modelled as the region of depolarisation. The magnetic field, 1 cm away from the scalp, was
estimated on injecting a constant current of 100 µA into the head from diametrically opposed
electrodes. However, in the second scheme, only the realistic FEM of the head was used, which
included a specific region of interest; the primary visual cortex (V1). The simulated physiological
change was the variation in conductivity of V1 when neurons were assumed to be firing during a visual
evoked response. A near DC current of 100 µA was driven through possible pairs of 31 electrodes
using ElT techniques. For a fixed skull conductivity, the resulting surface potentials were calculated
when the whole head remained unperturbed, or when the conductivity of V1 changed by 0.6%, 1 %,
and 1.6%.
The results of the magnetic measurement predicted that standing magnetic field was about 10pT and
the field changed by about 3fT (0.03%) on depolarization. For the second scheme, the greatest mean
current density through V1 was 0.020 ± 0.005 µAmm-2, and occurred with injection through two electrodes positioned near the occipital cortex. The corresponding maximum change in potential from baseline was 0.02%. Saline tank experiments confirmed the accuracy of the estimated standing
potentials. As the noise density in a typical MEG system in the frequency band is about 7fT/√Hz, it
places the change at the limit of detectability due to low signal to noise ratio. This is therefore similar
to electrical recording, as in conventional ElT systems, but there may be advantages to MEG in that
the magnetic field direcdy traverses the skull and instrumentation errors from the electrode-skin
interface will be obviated. This has enabled the estimation of electrode positions most likely to permit
recording of changes in human experiments and suggests that the changes, although tiny, may just be
discernible from noise.

Item Type: Thesis (PhD)
Additional Information: In Collaboration with University College London. Sponsor: University of Ghana, Legon-Accra, Ghana.
Research Areas: B. > Theses
A. > School of Science and Technology > Natural Sciences
Item ID: 9494
Depositing User: Adam Miller
Date Deposited: 17 Dec 2012 11:12
Last Modified: 30 Nov 2022 02:12

Actions (login required)

View Item View Item


Activity Overview
6 month trend
6 month trend

Additional statistics are available via IRStats2.