NeuroFlex: Wireless, Mechanically Flexible, Stimulation-Capable Depth and Surface High-Density Microelectrode Arrays for Epilepsy - Project Summary/Abstract Intracranial electrocorticography (ECoG) is widely used in the evaluation of patients with drug-resistant epilepsy. Current neural interface technologies suffer from tradeoffs between the density of electrodes, ability to sample from broad networks, and limitations on data transmission and power. Commercial subdural and stereotactic depth electrode arrays used for continuous clinical monitoring require wired connections through the skin to external electronics for days to weeks of in hospital monitoring. These arrays are costly to manufacture or customize, cause significant brain tissue reactivity, and typically have small numbers (<64) of millimeter-scale electrodes, significantly limiting recording resolution and detection of informative seizure features. Here, we seek to translate to human use the NeuroFlex system of modular, fully implantable wireless depth and subdural high density microelectrode arrays with support for up to 1024 recording and stimulating electrodes on each implant. NeuroFlex builds on a custom complementary metal-oxide-semiconductor (CMOS) application- specific integrated circuit (ASIC) developed under the DARPA Neuroelectronic Systems Design program that provides wireless powering and high-throughput data telemetry as well as data conversion and front-end analog electrical microstimulation and recording electronics in a fully implantable 1.2 cm × 1.2 cm chip, which is thinned to less than 20 µm. Mechanically flexible, customizable polyimide extender depth or surface arrays are bonded to each implanted circuit. Multiple arrays can be implanted with no wires or connections, and minimal volume displacement. An external, wearable “relay station” powers and communicates with one or more arrays indefinitely with no implanted batteries. Wireless operation mitigates several sources of patient risk and discomfort and has the potential to untether patients from the hospital with ambulatory recording, while dramatically improving the quantity and quality of data transmitted. The implants minimize tissue damage and displaced volume, can be efficiently implanted and removed, and are inexpensive to manufacture and customize. In the UG3 phase of the project, we will complete enhancements to the wearable relay station, surgical implant technique, and data visualization software (Aim 1). In a porcine model, we will demonstrate recording quality and stability for acute pharmacologically induced seizures and somatosensory evoked potentials, define microstimulation parameters for cortico-cortical evoked potentials and electrocortical stimulation mapping, and demonstrate safety. In Aim 3, we will perform additional biocompatibility, safety, and accelerate aging studies, including prolonged device operation in a second animal model, and engage with the FDA for an Investigational Device Exemption. In the UH3 phase, we will demonstrate feasibility of recording and stimulation for mapping cortical function in human subjects in the intraoperative setting (Aim 4). After demonstrating initial human translation, we will demonstrate safety, tolerability, and feasibility for fully-implanted, short-term (<30 days) epilepsy intracranial monitoring (Aim 5). These studies are designed to bring the device to larger clinical trials.
