Abstract
A method is presented for reconstructing images of fast neural evoked activity in rat cerebral cortex recorded with electrical impedance tomography (EIT) and a 6 × 5 mm2 epicortical planar 30 electrode array. A finite element model of the rat brain and inverse solution with Tikhonov regularization were optimized in order to improve spatial resolution and accuracy. The optimized FEM mesh had 7 M tetrahedral elements, with finer resolution (0.05 mm) near the electrodes. A novel noise-based image processing technique based on t-test significance improved depth localization accuracy from 0.5 to 0.1 mm. With the improvements, a simulated perturbation 0.5 mm in diameter could be localized in a region 4 × 5 mm2 under the centre of the array to a depth of 1.4 mm, thus covering all six layers of the cerebral cortex with an accuracy of <0.1 mm. Simulated deep brain hippocampal or thalamic activity could be localized with an accuracy of 0.5 mm with a 256 electrode array covering the brain. Parallel studies have achieved a temporal resolution of 2 ms for imaging fast neural activity by EIT during evoked activity; this encourages the view that fast neural EIT can now resolve the propagation of depolarization-related fast impedance changes in cerebral cortex and deeper in the brain with a resolution equal or greater to the dimension of a cortical column.
Highlights
IntroductionOver the past two decades, there have been huge advances in the ability to image brain function
Over the past two decades, there have been huge advances in the ability to image brain function. Techniques such as functional MRI and positron emission tomography have enabled imaging of changes in blood flow and metabolic changes, but these are only over seconds as they are secondary to brain activity
Electrical impedance tomography (EIT) is an emerging imaging technique which enables tomographic images of the electrical properties of a subject to be produced with rings of electroencephalography (EEG)-type electrodes
Summary
Over the past two decades, there have been huge advances in the ability to image brain function. In 2011, our group published a method for measuring impedance changes due to fast neural activity in the somatosensory cortex using a planar epicortical electrode array during repeated sensory stimulation (Oh et al 2011). This method would only allow for imaging with temporal and spatial resolution of 8 ms and >500 μm. Other methods have limitations: visible light optical mapping has limited penetration into the brain (Steinbrink et al 2005), MRI of neural currents has a low sensitivity and limited temporal resolution of about 30 ms due to echo-planar data acquisition (Parkes et al 2007), and the new method of MR-EIT, in which a dc current is applied to an object and distortions in the resulting magnetic field are imaged with MRI (Sadleir et al 2006) at present does not appear to have sufficient sensitivity
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
