UC San Diego

Neuromodulation devices are increasingly used in interrogating and treating neurological dysfunction in the human brain, spinal cord, and peripheral nerves. Recently, there has been an explosion of interest in applying these devices for brain-machine interfaces and advancing the state-of-the-art interface with the human brain. Toward this end, the Utah array has been the powerhouse of the BrainGate project that partially restored motor and sensory function to patients with neurological injury, though the array itself has advanced little in the last three decades. In this work, we leverage advanced dual-side lithographic microfabrication processes to demonstrate a 1024 channel penetrating Si microneedle array (SiMNA) that is scalable in its recording capabilities and cortical coverage. The SiMNA is built on flexible and transparent substrates permitting simultaneous optical and electrophysiological interrogation of the brain activity and is compliant to brain movements. We use the SiMNA to demonstrate reliable recordings of spontaneous and of evoked field potentials and of single unit activities in chronically implanted mice for up to 196 days in response to optogenetic and to whisker air-puff stimuli. Significantly, the 1024 channel SiMNA established detailed spatiotemporal mapping of broadband brain activity in rats. This novel scalable and biocompatible SiMNA with its multi-modal capability and sensitivity to broadband brain activity will accelerate our progress in fundamental neurophysiological investigations and establishes a new milestone for penetrating and large area coverage microelectrode arrays for brain-machine interfaces.The reach of soft substrates extends well beyond the curvilinear and pulsating brain toward spinal cord that relays the information bidirectionally to the brain and that has its own processing circuits for fundamental locomotion tasks. Materials that comprise of superior contact properties for recording and stimulation, that are minimally-invasive, and that are biocompatible and flexible have profound impact on the way we record and stimulate activity of the spinal cord. One important application of such devices is to aid in restoring function in spinal cord injury (SCI). However, thresholds for motor recruitment and for tissue damage during direct current stimulation in the spinal cord using recent microelectrode technologies are not yet established. Additionally, stimulation on the ventral side of the spinal cord in a closer proximity to the motor fibers is advantageous but systematic studies on the efficacy of microelectrode arrays for ventral stimulation versus dorsal stimulation and the stimulation parameters needed for initiation of motor response have not been studied before. This work reports the initiation thresholds for motor recruitment of rodent hindlimb from dorsal and ventral stimulation with various electrode sizes using a newly developed microelectrode material in our laboratory, the platinum nanorod (PtNR) contacts using two device configurations. Device type 1 comprises of microelectrodes of 9 diameters (40/60/80/100/120/140/160/180/200 μm) and a macroelectrode of 250 μm in diameter on sub-10 μm thin flexible parylene-C substrate. Using Device type 1 in the acute setting for dorsal or ventral-lateral spinal cord implantation and electrical stimulation in rats, we quantified lower current thresholds and charge densities, and a lower critical diameter for evoking responses in the sciatic nerves and electromyography responses in hindlimb muscles. Device type 2 consists of three representative diameters (40/100/200 μm) of PtNR electrodes from Device type 1 on sub-10 μm thin flexible polyimide substrate for investigating the stability of the platform on semi-chronic rat implants accompanied by a rigorous stimulation paradigm with a total of over 1 million pulse pairs. This stimulation paradigm is designed based on the a 7-hour stimulation session initially implemented by McCreery et al. In our work, we used a much higher pulse frequency of ~ 200 Hz over the duration of 84 minutes doubling the amount of charge used in McCreery et al.’s work. Device type 3 consists of 128 channels (4 × 32 array) of PtNR electrodes with 30 µm diameter on sub-10 μm thin flexible parylene-C substrate which was used to detect spatiotemporal compound action potentials (CAP) from an acute pig model and demonstrate the high scalability of the device platform.