|
Microscopic conductivity profile in the cerebral cortex of Wistar rats| old_uid | 6210 |
|---|
| title | Microscopic conductivity profile in the cerebral cortex of Wistar rats |
|---|
| start_date | 2009/02/06 |
|---|
| schedule | 10h30 |
|---|
| online | no |
|---|
| summary | In general, local field potentials (LFP) are low-frequency (0.7-170Hz) extracellular signals
recorded by high resolution arrays of electrodes acutely/chronically implanted in the brain. The LFPs are supposed to be caused by microscopic current sources embedded in a conductive medium, the brain. Hence, under the validity of quasi-stationary approach, which is the case in this frequency range, the divergence of current sources relates to the electric potential by the Poisson equation. The distribution and time course of current sources inside a cortical microcolumn can be estimated by solving this equation from the LFPs. For that end, in most previous works the conductivity of each cortical layer has been assumed to be homogeneous and isotropic; and for simplicity, the Poisson equation has been discretized with the step size of the distance between each electrode of multisided probes. However, this oversimplification causes spatial aliasing and limitation of the resolution in estimating current sources.
Furthermore, the cell body distribution, density and neuronal fibers are different in each layer; hence, the actual conductivity is neither homogeneous nor isotropic.
In this work, we propose a new method to estimate current sources continuously inside a
cortical microcolumn from LFPs obtained through an implanted 3D high resolution electrodes array. To accomplish that, we firstly developed a device to determine the conductivity profile in several areas of the cerebral cortex of Wistar rats. The device is composed by a combination of 2D and 1D silicon-based probes. We developed a circuit to select electrodes by pair as sources and sinks for current injection in the 1D probe and record the resulting LFPs in the 2D probe. The sources and sinks were selected in order to reduce the bias in determining radial and tangential conductivities in each layer. From the entire data, and assuming a piece-wise homogeneous and anisotropic spherical volume conductor model for the cortex, we estimated the conductivity profile by using a least-square optimization algorithm. In order to verify the proposed methodology, we applied it to determine the conductivity profile in a phantom constituted by two spherical layers made from low concentration agarose solutions with sodium chlorides at different concentration, each layer of 500?m thickness. Finally, we applied this methodology to characterize the conductivity profile in the somatosensory cortex of Wistar rats (eight weeks). We compared our results with those obtained by Logothetis et al. (2007) using
two anesthetized monkeys. We will present preliminary data obtained from three rats that shows not only conductivity changes among layers but also a clear distinction in the tangential and radial components. |
|---|
| responsibles | Béranger, Giard |
|---|
| |
|