Noninvasive Acoustoelectric Imaging of Spatiotemporally Varying Currents to Diagnose Neuronal Electrical Dysfunction

Abstract

Transcranial Acoustoelectric Brain Imaging (tABI) offers what Scalp Electroencephalography (EEG) alone fails to offer in the complete characterization of epilepsy. Invasive Depth Electrode (DE) EEG techniques may achieve localized detection of neuronal function and noninvasive scalp EEG sacrifices accuracy of signal detection for safety. The goal of this work was defined through noninvasively visualizing travelling current densities of the brain at varying locations and times, determining signal detection limits and calculating conduction velocities of travelling deep neuronal waves. I proposed a method to detect signal profiles that changed over a period of 50 ms and 35mm distance. Neural currents were simulated ex-vivo in a human head model using a 12-site Spencer/AdTech Depth Electrode (DE) and Infinity Deep Brain Stimulator (DBS) with varying current geometries (dipole, monopole). Slow ≤ 200 Hz and varying magnitude signals were used to emulate neuronal signals in an agarose/saline gel brain phantom inside a real human skull. TABI movies were produced using a 0.6 MHz center frequency, 2D Ultrasound (US) array. These movies provide demonstrable evidence of effective correction for transcranial US imaging of specific current locations with millimeter (mm) and millisecond (ms) accuracy, more specifically, track these signals as they travel spatially and temporally regardless of their direction of travel. Conduction velocities were found through tracking centroid peak magnitude coordinates (x,y,z) over time to map more effectively specific (1.5mm and 4.5mm site separation on Infinity and AdTech electrodes, respectively) spanning wide field of view (35mm) at sub-cranial depth of ~30mm. Current detection limits were nominally found to be ≤300µA. Site to site conduction velocities were determined to be 4.71 m/s and 3.9 m/s on the AdTech and Infinity electrodes, respectively. TABI offers a safe, noninvasive, accurate and less expensive route to better track epileptic (ictal) signals, neural degenerative disease, and cardiac fibrillation in the human brain and heart model. In the final chapter of this work, a prototypical cardiac arrhythmia ex-vivo model is described. Baseline tests were made and there remains room for improvement with design ideas listed for better emulation of the heart electrical dysfunction included.